Re: oh no something is definitely wrong adieu debian.

2013-08-30 Thread berenger . morel

Le 30.08.2013 10:23, Joe a écrit :

On Thu, 29 Aug 2013 20:09:44 +0200
Ralf Mardorf  wrote:


On Thu, 2013-08-29 at 18:58 +0100, Joe wrote:
> Some of the English also like beer, and while German lagers are 
well

> respected, so are Belgian and Danish ones, and some of us prefer
> unpasteurised bitters. Some even prefer fermented apple juice...

Yes, some Danish beer and even some English/Irish beers are 
passable.
Belgian is known for good, but exotic beer flavours. I can't say 
much
pro or con Belgian beer, I don't know their beer good enough. 
There's

a lot of drinkable Lager on our planet, I would drink a Fosters when
thirsty and nothing else should be available. The English and Irish
folks perhaps should learn to use refrigerators, I guess no beer 
from

any country does taste good when it's not could.



Would you chill red wine? Unpasteurised bitter is drunk for its 
taste,

much of which is lost when chilled. Agreed, lager must be very cold,
like white wine.

--
Joe


Technically, it depends on the red wine. I am not good with wines, but 
in France I know people which drink *some* red wines "cold" ( around 12° 
IIRC ). Same for some white wines, some are better to ambient temper. I 
could ask my father about that, he knows wine better than me, but I 
think it depends on if the wine is "sec" or another variant.


I like to compare beers (that I prefer, although I do not say no to 
some wine) with wines: there are too many sorts to have rules of thumb. 
Except that the foster is for when there is nothing else, of course, but 
you know what? French guys are able to make worse... (I will not name 
them except if asked for, because I have big problems to name those 
stuff beers)
I think that most countries are able to brew some drinkable beers, 
anyway. In France I like the jeanlain, even if I know that we are not as 
good as Belgians for beers (I have no real knowledge about Germany's 
beers. As I never say that I know beer, only that I know them better 
than average French people ;) ).



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/00359a1cea1e133933b74edbe060f...@neutralite.org



Re: difference Debian, solaris, freebsd

2013-08-30 Thread berenger . morel



Le 29.08.2013 19:09, Ralf Mardorf a écrit :

On Thu, 2013-08-29 at 19:00 +0200, Ralf Mardorf wrote:
PS: I wouldn't install Debian's FreeBSD, test the "real" FreeBSD 
first.


PPS: The reason for this is, that there's a FreeBSD community and I
guess there is not a huge Debian GNU/kFreeBSD community, but I might 
be

mistaken.

http://www.debian.org/ports/kfreebsd-gnu/


Is not that list including that community? After all, KFreeBSD is now 
an official part of Debian, I think ;)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8783bba79f5b5fc7d8f5ca6169e61...@neutralite.org



Re: TEST

2013-08-30 Thread berenger . morel



Le 30.08.2013 09:15, Anthony Campbell a écrit :

On 29 Aug 2013, staticsafe wrote:

On Thu, Aug 29, 2013 at 01:04:35PM +0200, Ralf Mardorf wrote:
> My MUA replied unwanted several times to "Re: apt-get upgrade 
problem
> with libenchant1c2a". I got several times an error, but seemingly 
it was

> sent several times to the list, but didn't came through.
>
> Something is fishy, I was able to reply to several other threads 
to

> several people and lists.
>

Received. Perhaps your e-mail provider is having issues?
--


A reply to a thread I sent a couple of days ago didn't arrive. Let's 
see

if this one does.

--
Anthony Campbell - a...@acampbell.org.uk
http://www.acupuncturecourse.org.uk
http://www.smashwords.com/profile.view/acampbell
https://itunes.apple.com/ca/artist/anthony-campbell/id73235412


That thread seems fine: your message is currently the last one of the 
suit of 3 messages.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/526c07354588f7b729f7bb478bacc...@neutralite.org



Re: A notification tray?

2013-08-30 Thread berenger . morel



Le 29.08.2013 17:20, Sharon Kimble a écrit :

Cinnamon and Gnome 3 both have a tray that holds all notification
messages until their read and deleted. These messages are ones like
saying that your backup has just started, or you’ve got 3 new emails,
nice simple easily forgotten ones. But, what is this 'tray'? Can it 
be
used in fluxbox, or other lighter DE's? Is there anything comparable 
to

the tray please, because I'd really like it in my fluxbox setup. It
was very useful when I was using cinnamon, but apt-get removed 
cinnamon
yesterday, and when I went to re-download it today there was a big 
bug

against it, so its not working at the moment.

But can anybody help re the tray? Could it be accomplished in a bash
script? Does someone with the skill to code it have the same itch and
would find a use for it?

Hopefully, thanks
Sharon.


I am confused by your message.

Are you speaking about the tray area, which is where the icons rely, 
like for opera, transmission-gtk, skype, the hour, or whatever?
Or are you speaking about bubbles that some softwares are able to spawn 
on the desktop?


If you are speaking about the bubbles, notifications, then you should 
take a look at package "notification-daemon" and which packages depends 
on it, it may help you in your quest :) .



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/48eb9a827a85450ee25f76416a50b...@neutralite.org



Re: oh no something is definitly wrong adieu debian.

2013-08-30 Thread berenger . morel



Le 30.08.2013 12:59, Phi Debian a écrit :

Amazingly I needed to do a check on ubuntu raring, so I quickly setup
a raring guest. And as I am at it, dit the
panel->properties->[background] test, and it got killed with out any
'oh no' message, so even harder to understand what happen :) I may be
back to debian for raring era :)

So this is more a general GUI linux world problem, all distro's are
victimised by questionable bus architecture, then wanted to get as
nice as MAC, and got as worth as windooze :) (just kidin, don't
flame).

And about opening a ticket, well why opening at debian since it is 
general...


I am not involved enough to know the process of reporting this

Cheers,
Phi


Debian have a tool named reportbug, which should be installed by 
default: 
http://packages.debian.org/search?keywords=reportbug&searchon=names§ion=all


Maybe you could use it. I did once, for a package on which I discovered 
a dependency bug (one of the real dependencies was not in aptitude, so 
the program crashed on my minimal install of Debian.).
It is a command-line tool, but it uses a sequence of questions quite 
clear, if you want my opinion. It seem there is another one, 
"reportbug-ng", but never tried it.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/92026f55ad5f85ac1573ba584...@neutralite.org



Re: oh no something is definitly wrong adieu debian.

2013-08-30 Thread berenger . morel



Le 30.08.2013 17:55, Sharon Kimble a écrit :

On Fri, 30 Aug 2013 17:35:39 +0200
berenger.mo...@neutralite.org wrote:




Le 30.08.2013 05:04, Phi Debian a écrit :
> Hi Gary,
>
> Thanx you for getting back to the original post :) while I admit I
> > did learn a bit of history about distro, display manager,
> > etc... :)
>
> [snip]

And what about cloning the git of vncserver, install development
dependencies, and configure/make/make install? That would mean you
would have fixes from vnc server's team, so that bug would be absent
like in Ubuntu, and you could stay on Debian.

There should be no big conflicts, since "usually" make install will
install in /usr/local/. The only thing I can think about is that you
will need to update it by hand, with git pull/make/make install. 
(and

people here could know how to suppress that problem, but not me :) )




You could use 'myrepos' to control the git sessions, when to update
your local repo its just a simple 'mr update' and let it roll. Makes
life a lot simpler :)

Sharon.


I known that someone would know way to enhance that :D
I wonder, maybe that mr update can be integrated in deb system, so that 
the git repo would be check as if it was just a line in 
/etc/apt/sources.list ? That would be really awesome!



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8ddd309e30365a948848f72f10de8...@neutralite.org



Re: oh no something is definitly wrong adieu debian.

2013-08-30 Thread berenger . morel



Le 30.08.2013 05:04, Phi Debian a écrit :

Hi Gary,

Thanx you for getting back to the original post :) while I admit I 
did

learn a bit of history about distro, display manager, etc... :)

[snip]


And what about cloning the git of vncserver, install development 
dependencies, and configure/make/make install?
That would mean you would have fixes from vnc server's team, so that 
bug would be absent like in Ubuntu, and you could stay on Debian.


There should be no big conflicts, since "usually" make install will 
install in /usr/local/. The only thing I can think about is that you 
will need to update it by hand, with git pull/make/make install. (and 
people here could know how to suppress that problem, but not me :) )



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/3d5655d61b27ea1c46f1e90928ee9...@neutralite.org



Re: What's the easiest and/or simplest part of Linux Kernel?

2013-08-30 Thread berenger . morel



Le 30.08.2013 15:11, Jerry Stuckle a écrit :

On 8/30/2013 2:00 AM, Joel Rees wrote:


Okay, so, for you, supporting inheritance and polymorphism at 
run-time

rather than at compile time is not sufficiently OOP.

And I don't particularly care about that distinction.

I'm fine with ending the discussion there.

--
Joel Rees




You keep claiming it can be done, but have done nothing to show how
it can be done.  So all I can assume is you cannot support your
statement, because it can't be done.

Don't worry - I've heard similar statements over the years from
others who don't understand OOP.  None of them have been able to
support their statements, either.

Jerry



Here is what you defined for inheritance and polymorphism:

3. Inheritance: the ability to extend an existing class, to provide
additional or different functionality via additional messages in the
derived class.  Inheritance takes advantage of the similarities in the
base an derived classes.  The base class has no knowledge of the 
derived

class and, in fact, may not even know it is being used as a base class.
Additional classes can be derived from the original base and derived
classes with no change to the existing code.  This cannot be done in C.

4. Polymorphism: the ability to send messages to a derived class object
when you believe you have an object of the base class.  This allows
functions to operate on any class in the derived hierarchy, while only
having to worry about the messages defined in the base class.  This 
also

cannot be done in C.


Note that we agree on those definitions.

Now, can you explain why the link I provided does not meet those 
requirements?


For now, except saying to everyone that 1) they do not understand OOP 
and 2) you teach it from 25 years, you never gave any example of things 
we could do in any OOP language that we could not make in C according to 
your definition of OOP (on which I agree, again) not you destroyed the 
source code pointed by the link I gave.


Oh, and, you also said something which implied that I said that SDL is 
a language, which is something I never said. Well, to be more precise, 
the functions and structures I referred to where obviously owned by 
libsdl1.2-dev (to write the Debian package's name) which is a library.
So, you avoided replying to real arguments with yours, and you even 
used a straw man to discredit me?


I will refrain my envy of irony here, instead I will be direct: I have 
seen people which were using a lot the argument of being older than me 
to convince me that I was wrong. They never convinced reality, when what 
they did failed, why my solutions were working fine. Wisdom and 
knowledge are not only a matter of age and teaching. Including teachers.
This is the reason why I rarely accept an argument if it is not 
correctly built: explanation + example, so that it can be countered or 
not.


While I am at it, let me say you that, a machine, in common language 
(I'm not very good with mathematics), have internal states that you do 
not need to know to use it's function. I have no idea about all the 
states of my car when I am driving. For example, I do not know, or need 
to know, the states of the spark plugs.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/a505a5a67ef41d7132e262f4cf0e2...@neutralite.org



Re: Choosing among "Desktop Enviroments" and/or "Windows Managers"

2013-09-18 Thread berenger . morel
To switch between DEs, you can do it with the usual connection 
managers, IIRC.
But I personally prefer to login on TTY and then start the window 
manager/DE of my choice. Usually, DEs have a 'start-my_favorite_de' 
command to do that. If you are tired about starting them by hand 
everytime, you can add those lines to your ~/.bash_profile file:



if [ -z "$DISPLAY" ] && [ $(tty) == /dev/tty1 ];then
startx
fi

Of course, starting from here can allow you to think about other 
things, like using startx for tty1, gnome for tty2, kde for tty3, etc. 
It is like the usual connection managers, but without the mouse or 
eyecandy stuff, and more lightweight. My current only problem with that 
is that I still need to enter my login and password at start, which is 
boring on my desktop, but I never tried to fix that for now.



Le 16.09.2013 20:57, Richard Owlett a écrit :

Dan Ritter wrote:

On Mon, Sep 16, 2013 at 10:06:32AM -0500, Richard Owlett wrote:

As I have minimal to *NO INTERNET* access, when Debian 7.1 DVDs
became available from vendors I purchased the 10 DVD set. I also
purchased the live DVDs for Gnome, lxde, kde, and xfce.  I did some
test runs of each. I chose xfce for my initial Wheezy install based
almost exclusively on "look and feel". It felt "comfortable" and
seemed to express my somewhat vague ideas of how a GUI should 
'act'.


It was an educational experience. Each had something I'd like to 
see

in an ideal installation. I also did more web browsing comparing
"Desktop Environments" to "Windows Managers" and product 
comparisons

within those classifications.

/Questions
1. Given that I prefer LILO over GRUB and xfce over Gnome/lxde/kde,
what GUIs should I investigate further.


If you're happy with xfce, you shouldn't need to look further.
xfce is a fully supported, first-class environment.


I don't _need_ to, but I'm involved with other projects which might
benefit. Also I'm retired with lots of free time and I find the topic
interesting.


If you like XFCE, you might also look at LXDE. It's tools are less 
linked between them than those of XFCE, and it's even more lightweight. 
I now use some carefully chosen parts of it in combination with i3, such 
as the primitive text editor (for notes. I use vim when I need a strong 
one) and the terminal emulator.


Now, it seem you only know about stacking window managers. Let me give 
you a different direction to explore: tiling window managers.
I will try to be as objective as possible, but remember that when I 
tried them, I simply fell in love with that paradigm, probably because I 
always was a computer tinkerer. So you are warned.
There are, AFAIK, no DE with some of them (well, I have heard that KDE 
had an optional one in previous versions, but it was removed. No idea 
about the truth in those words, KDE is too heavy for my computer uses) 
but if you use lightweight DE, you can usually change the window 
manager.


If you want the name of one on official DVDs, I think I would speak 
about i3, which is the first one I successfully tried, and the one which 
remain on my computers, bot netbook and dekstop with 2 screens.
It's documentation claims that it is for powerusers but... I think it 
is only to not have people come and bother them with stupid questions 
like how to edit a text file, because the doc itself is very clear, and 
the configuration of that DE is *NOT* made with a programming language, 
which is a strong point and something which makes it very easy to use. 
In short, it uses a real configuration file, and it works when you 
install it, unlike all other TWM I have tried (but others have other 
features which can be useful as well).


Now, why could someone be interested in that kind of managers?
The difference is simply that, instead of having a stack of windows of 
different sizes that you have to manually set, each new window is 
automatically placed on the desktop and all windows are resized to fill 
the entire space without overlapping.
This allows, for example, to control everything with your keyboard 
(pretty useful if you are mostly working with text, because you do not 
have to move a hand to the mouse.), or if you have more than one screen, 
it is really easier to be efficient with them, when it's painful with 
classic window managers (moving a mouse from a screen to another just to 
focus another application is painful, and using alt-tab is not always 
very fast).


The problem is that they are usually not beautiful. I guess that the 
reason is that they are not very well known, and so, often made by power 
users for power users, like i3 claim.
The beauty problem resides in the fact that there is no transparent 
windows, no rounded corners, no buttons to control the windows, etc (but 
there is still an optional title bar). This imply that heavy mouse users 
will find them harder to use, but maybe ( I never tried that ) it can be 
fixed by using window decorators or integrating them into 

Re: Choosing among "Desktop Enviroments" and/or "Windows Managers"

2013-09-18 Thread berenger . morel



Le 18.09.2013 14:22, berenger.mo...@neutralite.org a écrit :

To switch between DEs, you can do it with the usual connection
managers, IIRC.
But I personally prefer to login on TTY and then start the window
manager/DE of my choice. Usually, DEs have a 'start-my_favorite_de'
command to do that. If you are tired about starting them by hand
everytime, you can add those lines to your ~/.bash_profile file:


if [ -z "$DISPLAY" ] && [ $(tty) == /dev/tty1 ];then
startx
fi

Of course, starting from here can allow you to think about other
things, like using startx for tty1, gnome for tty2, kde for tty3, 
etc.

It is like the usual connection managers, but without the mouse or
eyecandy stuff, and more lightweight. My current only problem with
that is that I still need to enter my login and password at start,
which is boring on my desktop, but I never tried to fix that for now.


Le 16.09.2013 20:57, Richard Owlett a écrit :

Dan Ritter wrote:

On Mon, Sep 16, 2013 at 10:06:32AM -0500, Richard Owlett wrote:

As I have minimal to *NO INTERNET* access, when Debian 7.1 DVDs
became available from vendors I purchased the 10 DVD set. I also
purchased the live DVDs for Gnome, lxde, kde, and xfce.  I did 
some
test runs of each. I chose xfce for my initial Wheezy install 
based

almost exclusively on "look and feel". It felt "comfortable" and
seemed to express my somewhat vague ideas of how a GUI should 
'act'.


It was an educational experience. Each had something I'd like to 
see

in an ideal installation. I also did more web browsing comparing
"Desktop Environments" to "Windows Managers" and product 
comparisons

within those classifications.

/Questions
1. Given that I prefer LILO over GRUB and xfce over 
Gnome/lxde/kde,

what GUIs should I investigate further.


If you're happy with xfce, you shouldn't need to look further.
xfce is a fully supported, first-class environment.


I don't _need_ to, but I'm involved with other projects which might
benefit. Also I'm retired with lots of free time and I find the 
topic

interesting.


If you like XFCE, you might also look at LXDE. It's tools are less
linked between them than those of XFCE, and it's even more
lightweight. I now use some carefully chosen parts of it in
combination with i3, such as the primitive text editor (for notes. I
use vim when I need a strong one) and the terminal emulator.

Now, it seem you only know about stacking window managers. Let me
give you a different direction to explore: tiling window managers.
I will try to be as objective as possible, but remember that when I
tried them, I simply fell in love with that paradigm, probably 
because

I always was a computer tinkerer. So you are warned.
There are, AFAIK, no DE with some of them (well, I have heard that
KDE had an optional one in previous versions, but it was removed. No
idea about the truth in those words, KDE is too heavy for my computer
uses) but if you use lightweight DE, you can usually change the 
window

manager.

If you want the name of one on official DVDs, I think I would speak
about i3, which is the first one I successfully tried, and the one
which remain on my computers, bot netbook and dekstop with 2 screens.
It's documentation claims that it is for powerusers but... I think it
is only to not have people come and bother them with stupid questions
like how to edit a text file, because the doc itself is very clear,
and the configuration of that DE is *NOT* made with a programming
language, which is a strong point and something which makes it very
easy to use. In short, it uses a real configuration file, and it 
works

when you install it, unlike all other TWM I have tried (but others
have other features which can be useful as well).

Now, why could someone be interested in that kind of managers?
The difference is simply that, instead of having a stack of windows
of different sizes that you have to manually set, each new window is
automatically placed on the desktop and all windows are resized to
fill the entire space without overlapping.
This allows, for example, to control everything with your keyboard
(pretty useful if you are mostly working with text, because you do 
not

have to move a hand to the mouse.), or if you have more than one
screen, it is really easier to be efficient with them, when it's
painful with classic window managers (moving a mouse from a screen to
another just to focus another application is painful, and using
alt-tab is not always very fast).

The problem is that they are usually not beautiful. I guess that the
reason is that they are not very well known, and so, often made by
power users for power users, like i3 claim.
The beauty problem resides in the fact that there is no transparent
windows, no rounded corners, no buttons to control the windows, etc
(but there is still an optional title bar). This imply that heavy
mouse users will find them harder to use, but maybe ( I never tried
that ) it can be fixed by using window decorators o

Re: Re: One-Stop Debian Box Config Tool: Call for Collaborators!

2013-09-18 Thread berenger . morel



Le 14.09.2013 10:25, Jarrod O'Flaherty a écrit :

Yes! That's it, essentially. You have put it very well here. I wish I
had described
it in that way. I am, in a nutshell, looking to make the process of
applying what
we find on Wikis and message boards -- all of which is fantastically
helpful --
to our systems just that much easier.


So, what you are planning to do is a meta search-engine, which would be 
able to build template from current internet's content, and then 
generate the final solution for the user.


Two lines of text. Probably dozens of month-man work. Not simply "a 
couple of hours a month". Do not think that programming is easy.


Well, I am not against that idea, but I will not contribute, I do not 
think you will be able to even reach the beta state.


Anyway, let me give you some hints, for both your technical and 
communication points.
_ stop using as much '!'. No need to mimic advertisements like "lose 
30Kg in 2 weeks! Without private yourself to eat good stuff!". I always 
wondered how those ads could work by taking people as fools like that... 
and your posts really shares such kind of points: too many enthusiasm, 
only showing points on a subjective and very positive point, etc. Plus, 
being too enthusiastic is a characteristic of people which does not have 
a lot of programming experience. I might be wrong, but, what if I assert 
you know programming since less than 3 years?
_ stop using html when you are writing to a mailing list. I do not know 
for other clients, but mine support it, and it makes your text appear 
way too big. I can zoom to adjust it, but decent sized text then are too 
small, and it's boring.
_ make a real description of your project. Show that you thought about 
it at least a little and it's not just an idea. People will not be 
convinced by just an idea. Even better, try to have a scale model to 
show. This will prove to people that you are actually serious and have 
real skills. Remember that projects always have drawbacks, and do not 
hesitate to say what they are. Be honest, in fact.
_ take a look at debtags. It might help you to have ideas about how to 
do what I think you want to do, because installing softwares is the 
really first step of system administration. I do not know for synaptic, 
but aptitude did not really evolved to manage debtags and multi-arch 
correctly (in other problems it have, but it still is a good tool).
_ being too enthusiastic for a project without clear objectives usually 
ends by over-engineering, which in turn makes the project hard to 
maintain, and dying before even reaching the beta stage.


Now, do not feel insulted by things I said, I had the same problem 
before, was very, very enthusiast for my projects, at least for the 1st 
month of real work. Then, all of them died, it took more or less time, 
but they all did without reaching the beta stage. I think it's the 
normal way for people who likes programming.


But liking programs is not enough to manage a project. First, show your 
skill with simple projects, that you can finish yourself, and then, 
slowly, move on bigger and bigger projects, when you'll have more 
experience.


Anyway, I sincerely wish you good luck.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/b83b189f9221f0bf49bd166096442...@neutralite.org



Re: Thank you - was [Re: Choosing among "Desktop Enviroments" and/or "Windows Managers"]

2013-09-18 Thread berenger . morel



Le 18.09.2013 13:31, Richard Owlett a écrit :

Richard Owlett wrote:

/Background
[snip]

/Questions
1. Given that I prefer LILO over GRUB and xfce over
Gnome/lxde/kde, what GUIs should I investigate further.
2. Given that internet is effectively non-existent and
internal/external disk space is effectively unlimited, how can I
make as many as possible of the DE &/or WM on the distribution
DVD simply available to experiment with?



Thank you to all who replied. I was able to load multiple DE's/WM's
and compare various permutations.

What's the best?


There is no best solution that we can give you. i3 is probably the best 
solution for me, but will be the poorest solution for some other people 
for which Gnome3 will be the best. Which in turn will not even meet the 
requirements of KDE's users.
Every user is different, so should be every Debian installation, at 
least for personal needs (and sometimes you have different needs 
depending on the hardware too... ). The only thing other people can do 
to help you is showing you other roads.



I still don't know. The exercise demonstrated that I
had misconceptions about Debian "philosophy" and underlying structure
[ e.g. led me to read more about package management].


This is Debian's philosophy as *I* see it: you will be able to choose 
all tools you want. Even the kernel, since you can use KFreeBSD ( ok it 
is recent, and maybe not fully tested, I do not know. I should try it 
someday. )
Debian is not a linux system, it is an operating system, which happen 
to be (often) relying on the linux kernel.
This imply, for example, that systemd will probably not become the only 
one init system. Same for gnome for DEs or for grub2 for bootloaders 
(but, yes, they can be the default tool in a period of time).


I guess this is why Debian meet my needs: you can tinker things easily, 
and so learn a lot even if you have poor starting knowledge.
But do not worry, it seem that on that list, everyone is still learning 
things regularly.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/410fa7b8b21c6fda2fbd47591943b...@neutralite.org



Re: Thank you - was [Re: Choosing among "Desktop Enviroments" and/or "Windows Managers"]

2013-09-18 Thread berenger . morel



Le 18.09.2013 22:28, Richard Owlett a écrit :

berenger.mo...@neutralite.org wrote:

[*SNIP*]
I guess this is why Debian meet my needs: you can tinker things
easily, and so learn a lot even if you have poor starting knowledge.
But do not worry, it seem that on that list, everyone is still
learning things regularly.



That "tinkering" is generally recognized as valid in Debian community
is part of my justification for using it.

That does NOT mean that my sanity is never questioned :>

I continue to say "Thank you" to the community.


The best way to show that you liked what we did, is to do the same :) 
(and that's a nice way to learn stuff too, when more experienced people 
fix what you said)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/24da8f5d767a99439e35dfbad8b40...@neutralite.org



Re: Don't want a desktop environment

2013-09-22 Thread berenger . morel

Le 21.09.2013 21:41, Fred a écrit :

but I understand I need to do more than that. Seems like I need some
display manager also?

I did try to do

aptitude install slim

which installs, but it's not like it magically autostarts when I boot
the system.

What I need is to config slim to autostart, which will run i3 upon
login?

Or do I need to set up the X server to autostart?

Thanks for any help!


You do not need to install any manager ( gdm, slim, kdm... ) to start 
an X session: simply install xinit ( if it is still not here, I think it 
is required by session managers? ) and start your X session with 
"startx".


Doing it by hand everytime is quite boring, of course, so you can add 
those lines to your ~/.bash_profile (create the file if it does not 
exists) :

###
if [ -z "$DISPLAY" ] && [ `tty` == /dev/tty1 ]
then
startx
fi
###
If you need a hint about what is the -z, take a look at "man test", 
since in bash "[ something ]" means "test something".


Since you are searching for the lightest solution, I think this is what 
you want. Of course, it means your login will not be a shiny graphical 
one, but if you really intend to use i3 (which I am using, it is really 
a nice wm) you probably do not aim at shiny stuff.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/ffd4270cb9ac25bd5c5fc38eb59e7...@neutralite.org



Re: Don't want a desktop environment

2013-09-22 Thread berenger . morel



Le 22.09.2013 01:14, Paul Cartwright a écrit :

On 09/21/2013 03:58 PM, Josef Bailey wrote:

i3 is a very good wm

I just installed & ran I3. I have a couple of issues.
1. it breaks Thunderbirds "n" to go to Next Message in another 
folder.

So you have to constantly use the mouse to move to the next folder
2. You can no longer click on web links in Thunderbird , nothing 
happens.

3. When I went back to Trinity WM and ran Thunderbird, it was in
fullscreen mode with no way to resize the window. I googled & found a
workaround, but I don't think I3 plays nice with Thunderbird, so  
that

won't work for me..

--
Paul Cartwright
Registered Linux User #367800 and new counter #561587


I am not a thunderbird user, but I have never noticed that problem. 
Maybe you messed yourself with your configuration? Can you show us the 
configuration file you used and describe exact steps to reproduce?


Anyway, I tried thunderbird (icedove in facts) to reproduce your bug, 
and indeed, I had bug. Network is no longer usable on the computer on 
which I did the try :) it's not really fun, I'll have to debug it now... 
I'll ask for help on another thread if I can not do it myself ( even 
localhost seems damaged btw, so at least it is not hardware failure ).



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4b63804d73d0ee46c94c593da198d...@neutralite.org



Re: Building computer

2013-09-28 Thread berenger . morel



Le 28.09.2013 13:33, Stan Hoeppner a écrit :

Hi Catherine,

I haven't caught up with the rest of the thread but just wanted to
address a couple points here.

On 9/26/2013 11:12 AM, Catherine Gramze wrote:


On Sep 26, 2013, at 1:05 AM, Stan Hoeppner  
wrote:




What desktop applications are you using that require 8GB, let alone
16GB, of RAM?  I'd think 4 would be plenty.  If you wish to over 
buy
DRAM, that's a personal choice.  It will likely not improve 
performance

in any meaningful way, for WOW in Wine, or anything else.



I will be running more than one app at a time. For example WoW, a 
browser, a Ventrilo client, and a chat client at minimum.


4GB is more than plenty, unless WOW has turned into a complete and 
total
memory hog.  Obviously it eats more running through Wine emulation.  
But
Win and WOW combined shouldn't eat more than 2GB, so you have 2GB 
left

to the rest, which is plenty.


I am only quickly reading the thread, and it's the 2nd time I see 
"wine" associated with "emulation".
As the name says, WINE Is Not An Emulator, it does not emulate a 
computer, it does not emulate the windows' kernel, it emulates nothing.


If WINE was an emulator, then GTK, Qt, GStreamer and almost everything 
on your computer would be emulators too. All those softwares are 
libraries, they provide functions to softwares and that's all.
So, saying that running an application with WINE will take more memory 
than the same application without it needs proofs. Run an OS which does 
not depend on windows' kernel, Debian for example, with wine and without 
your desktop environment and other linux only applications. Instead, 
replace them with windows' applications, and then, if you effectively 
have difference of memory compared to the same application set on 
windows, then ok.


What WINE does is translating application's requests when they have a 
different name, and implementing them when they does not exists on our 
systems.


Plus, about the memory consumption wine, or any software library should 
take against another one, it is almost nothing. What takes memory in a 
software are data: graphics and sounds, of course, but also logic data, 
in case of WoW: player's lists, with their names, hp, etc. And that last 
kind of data is almost nothing compared to graphical resources of most 
3D games.
Add to this the performance difference between those systems, and you 
might even need less memory with linux. Or more. It depends on what the 
application needs, on the options of the kernel and other obscure things 
that I do not understand, even if I am a programmer ( but a good one and 
not a WINE one, however. ).


For WoW, you will probably not need more memory than on windows, since 
it is a software used by a lot of people from a long time and so, 
probably have a good "support" from wine. And if you need more, then it 
will not be tons of MiB. It will not be measurable against the GiB that 
WoW probably needs for graphical resources.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/866d3396edcd37385680109ba03d2...@neutralite.org



Re: Building computer

2013-09-28 Thread berenger . morel



Le 28.09.2013 22:46, Stan Hoeppner a écrit :

On 9/28/2013 8:14 AM, berenger.mo...@neutralite.org wrote:



Le 28.09.2013 13:33, Stan Hoeppner a écrit :

Hi Catherine,

I haven't caught up with the rest of the thread but just wanted to
address a couple points here.

On 9/26/2013 11:12 AM, Catherine Gramze wrote:


On Sep 26, 2013, at 1:05 AM, Stan Hoeppner 


wrote:



What desktop applications are you using that require 8GB, let 
alone
16GB, of RAM?  I'd think 4 would be plenty.  If you wish to over 
buy
DRAM, that's a personal choice.  It will likely not improve 
performance

in any meaningful way, for WOW in Wine, or anything else.



I will be running more than one app at a time. For example WoW, a
browser, a Ventrilo client, and a chat client at minimum.


4GB is more than plenty, unless WOW has turned into a complete and 
total
memory hog.  Obviously it eats more running through Wine emulation. 
But
Win and WOW combined shouldn't eat more than 2GB, so you have 2GB 
left

to the rest, which is plenty.


I am only quickly reading the thread, and it's the 2nd time I see 
"wine"

associated with "emulation".
As the name says, WINE Is Not An Emulator, it does not emulate a
computer, it does not emulate the windows' kernel, it emulates 
nothing.




Whether you call it an emulator, translator, simulator, or Santa 
Claus,

it 'provides' the Windows APIs to the application, the DLLs, etc.
Providing this does require additional memory.  It's not a large 
amount
by today's standards, but it is non negligible.  I made that case 
above

and previously in the thread.


So, you do think that an emulator is the same as a dynamic library? If 
yes, well... why not. But then, since all DL are emulators, do not use 
that word to prove that they'll consume more resources, being CPU or 
memory.



So I'm unclear as to why you picked my reply for your rebuttal, given
we're on the same page.


The reason for which I replied, is that an emulator emulates a complete 
system and this have a huge overhead. WINE, as a dynamic library, could, 
theoretically ( I say theoretically because I did not ran any tests, 
I'll be honest on that point. Plus, it's impossible to have strictly the 
same costs ) have the same overhead than window's API. In practice, it 
will have a small CPU overhead, but to say it's not a small one, one 
should give some valgrind analysis.


About the reason to reply to your post and not another one, it was 
because it was the second one saying that, that I had read on the same 
thread at the moment I replied :)

It was not personal.


The problem with 3D game performance under Wine


I will not say it does not cost more than running on windows. I have no 
proofs. But it is not related to the fact it is an emulator, it can only 
be related to the fact it is a badder implementation, or one with more 
more layers.



is not memory
consumption, but the CPU overhead,


CPU for 3D stuff? You might be right and I am probably wrong, but could 
not it be because linux's 3D drivers are less good than windows' ones? 
This is a real question, not a troll, and the reason of that opinion of 
mine is quite easy to understand, and so, probably very simplistic: 
video games are mostly targeting windows' users, and so, there were more 
money and time spent on enhancements... on windows' side.


Well, again, I admit, I have no benchmark to prove my words.
Of course, on a more technical point, I can agree that one more layer 
for OpenGL related stuff might have a cost. But, that cost might also be 
removed at compile time.
I can learn that my opinion here is wrong, I have no problem with that. 
I'm wrong on a lot of things after all, and am always happy when I learn 
that I was wrong on something else. But give me any reason. Or a proof. 
Use a linux kernel, and a WINE base environment, then show be 
benchmarks. That would be sufficient. Or reasons for why wine should 
cost much ( so that I could do some searches on your words and my errors 
).



which I also made clear previously.


This is exactly why I admitted having only read quickly the thread. 
Sorry, but I did not noticed that part. From my memories, what I have 
read and that might not be your own words ( I have also read that you 
say that a lot of RAM is useless for most users, and I agree with that ) 
was that wine + wow would take at least 2GB. It may be true, but from 
what I remember to have read, there was an implicit affirmation that it 
was due to wine, when it was because of the whole system.


My reply was not against anyone, my apologies if it seems so. It was 
because an emulator is something, and an API is another different thing 
(but, with enough abstraction, we could say it's the same, since their 
uses is always to make a software running...) which have less costs.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/6c2fa4

boot problem: stuck at "Switching to clocksource tsc"

2013-10-05 Thread berenger . morel

Hi.
Since my last kernel update my desktop can not boot anymore, it is 
stuck at "Switching to clocksource tsc". Also, nothing at all reply, 
even the keyboard does just nothing.


Of course, I was stupid enough to remove the last kernel without 
testing it, and I have no idea about what is wrong.
The kernel currently installed on that computer is 3.10-3, which works 
perfectly on that computer ( a netbook ).
I thought it was a problem with my lilo.conf, but I checked it for the 
4th time ( which is not fun on recovery mode, since vim is just highly 
bugged in ansi mode ) and it seems fine.
Few searches on the web indicates that it could be a compilation 
problem, but I have used the Debian's kernel without any change.
I also tried to reinstall everything (just in case), but it does not 
changed anything (of course).



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5beaeb9d67317109f033517b4487c...@neutralite.org



Re: About debain kernel stable repo

2013-10-06 Thread berenger . morel



Le 06.10.2013 11:11, 陶治江 a écrit :

Hello!

 I am using debian wheezy, and now I want to focus on
 the kernel study, or other stuffs.
 I can install release code from linux-source-3.2.xx package,
 But I want to ask whether we can access or track the debian
 official kernel repos (of stable release), It means I can
 using git or svn to get the lastest stable kernel source and the
 kernel updating tracks,
 without installing the latest source package.

 Any info would appreciated.

 Nicol


This link might help you: https://alioth.debian.org/scm/?group_id=30428
Honestly, I am surprised, it seem they are using svn here, but I 
remember having seen git repo for other packages... Maybe you should ask 
such kind of things on the development mailing list, informations would 
be more precise and or updated.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/b476d5988751f24dbfba8a4b4e9f1...@neutralite.org



Recovery mode: impossible to run dpkg ( was Re: boot problem: stuck at "Switching to clocksource tsc" )

2013-10-06 Thread berenger . morel

Le 05.10.2013 14:54, berenger.mo...@neutralite.org a écrit :

Hi.
Since my last kernel update my desktop can not boot anymore, it is
stuck at "Switching to clocksource tsc". Also, nothing at all reply,
even the keyboard does just nothing.

Of course, I was stupid enough to remove the last kernel without
testing it, and I have no idea about what is wrong.
The kernel currently installed on that computer is 3.10-3, which
works perfectly on that computer ( a netbook ).
I thought it was a problem with my lilo.conf, but I checked it for
the 4th time ( which is not fun on recovery mode, since vim is just
highly bugged in ansi mode ) and it seems fine.
Few searches on the web indicates that it could be a compilation
problem, but I have used the Debian's kernel without any change.
I also tried to reinstall everything (just in case), but it does not
changed anything (of course).


I think I have no other choice than trying to downgrade, so I have used 
ssh to send the packages ( that I kept on that computer ) to the target.
Now, bterm which is used in recovery mode starts to be *very* annoying, 
and avoid dpkg to work!

Here is what the system says:
===
Running depmod.
Error opening terminal: bterm.
debconf: dialog output the above errors, giving up!
dpkg: erreur de traitement de linux-image-3.10-2-amd64 (--install) :
 le sous-processus script post-installation installé à retourné une 
erreur de sortie d'état 255

Des erreur ont été rencontrées pendant l'exécution :
 linux-image-3.10-2-amd64
===

dpkg uses french messages, but I think they are useless. What is 
important is that I can not install package in recovery mode, so I can 
not revert my changes, which is very annoying in recovery mode.


I checked if all files were correctly generated in /boot, just in case, 
but the initrd.img file is not. I suppose it is generated by depmod?


Any idea about how to be able to run terminal stuff correctly in 
recovery mode? Maybe changing the terminal (if so, how could I do 
that?)?


Honestly, recovery mode is easier than using the unusable busybox ( I 
stopped installing them when I understood that they allow to do nothing 
else than cd and ls. ), but having it unable to correctly run basic 
tools like text editors and dpkg is not nice. I could use a debian live 
or anything else, but I would like to be able to use the tools debian 
give us to repair damaged systems.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/cdae5ac3b2e4d42fa201d590c4810...@neutralite.org



Re: You can have any color you want - as long as it's Gnome?

2013-10-07 Thread berenger . morel

Le 07.10.2013 09:10, Oliver Fairhall a écrit :

Thanks for posting on this Jape. I'm new to Debian (preparing to
install soon) and will likely be caught by this. Hope it's OK to ask
some newby questions.


I do not remember having read that this list is for experts only ;)


On 07/10/13 02:31, Jape Person wrote:

From my perspective, it looks to me as though the problem is
network-manager-gnome's desire to install gnome-control-center. Xfce 
and LXDE
both want network-manager-gnome, so they also get 
gnome-control-center,

gnome-session, and just about everything else gnome-like.


Is it possible to not install network-manager-gnome when installing
Debian with XFCE?


Debian's installer is not so flexible to allow you to choose softwares 
one by one, but it can allow you to install "tasks" one by one, by 
example "portable computer", "ssh server", "desktop", etc.


If you want a minimal system and know 1) all uses you will have for 
your computer and 2) which softwares in Debian's repository to use for 
them, then I suggest you to uncheck all tasks' checkboxes.
Then, when the installation will be finished, you will login on the TTY 
and install manually packages. For that, I honestly think that the 
aptitude's ncurses interface is the better, but "apt-get install [long 
list of software]" will work too, of course.


( another advice I could give you if you do not know what softwares to 
use for each task, is to install debtags, that you can access in 
aptitude's ncurses' interface in "view". This will help you to choose 
softwares according to various parameters like technology they use )


Of course, that kind of install is not the recommended one for people 
who are not ready to spend time on their computer: you might forget 
something and wonder why it does not work as you expect, when it would 
work fine on a classic installation.



I've bypassed the network manager in Ubuntu in the
past, running on a desktop machine, and just configured network 
access

by text file anyway. Not sure if that would make things awkward on a
laptop connecting to different wireless sites.


I am editing /etc/network/interface too. It works perfectly, but of 
course, you will have to change the values and do some ifdown/ifup 
everytime you will change the target network.
I guess it is possible to build some scripts to change network 
automatically, IIRC there is some kind of events that can be programmed, 
so, maybe... but I never tried. I stopped myself after making scripts to 
generate the interface file depending on a parameter and then down/up 
the wlan.

Using tools made for that is probably easier.


Are all these Gnome packages real dependencies for
network-manager-gnome, or are they just selected by some other means?


What is strange here, is that IIRC, XFCE uses network-manager, and not 
network-manager-gnome.
Anyway, installing network-manager does not install the whole gnome DE 
here, at least not without installing recommended packages.
Keeping automatic installation of those recommended packages would 
effectively install a lot of crap, because network-manager-gnome 
recommends other packages which, themselves, recommends or depends on 
DEs like KDE.


I am only doing a short test here, so this reply should not be taken as 
the exact truth.



Is there an alternative network manager for XFCE, and can one be
selected during initial installation?


During installation? I guess no, as I said previously, you can not 
choose a single package in the installer, only tasks, which are not 
precise at all.
But post-installation, yes, network-manager-gnome depends on a package 
named network-manager. I did not noticed any GUI software depending on 
it, but I did not take a deep look. My bet is that there is one, I just 
do not know the name because I do not need it, being perfectly happy 
with my configuration edition (which would not be the right solution if 
I had a lot of places where I go frequently with my computer, of course)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/36fa5a4a42cc04729053e31447eff...@neutralite.org



Re: Recovery mode: impossible to run dpkg ( was Re: boot problem: stuck at "Switching to clocksource tsc" )

2013-10-07 Thread berenger . morel



Le 06.10.2013 18:34, Curt a écrit :

On 2013-10-06, berenger.mo...@neutralite.org
 wrote:


Any idea about how to be able to run terminal stuff correctly in
recovery mode? Maybe changing the terminal (if so, how could I do
that?)?



export TERM=linux

says google


Thanks. But, this line does obviously not change the terminal: it lure 
the softwares so that they can not check that the real running terminal 
is not the good one.
Remember the context: I am trying to save an installation from recovery 
mode, which is in TTY.


However, I've just tried it, in case it would work. The dialog box is 
very ugly, since the formating is completely destroyed, but since 
changing kernel only shows a dialog box to say that one could have to 
install firmwares, it's ok. So, thanks you.


Sadly, it does not resolve my original problem of clocksource... I 
guess I'll have to reinstall.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/36b8e96a3b4d8f802e18f77a2c148...@neutralite.org



Re: Compiz on Debian 7.1 problem

2013-10-07 Thread berenger . morel



Le 07.10.2013 12:44, Douglas Brito a écrit :

good day, please be possible to add repositories in compiz again,
many users here in Brazil, the viviaolinux.org need this package and
it is no longer possible to use the debian, some are switching
distributions for this problem, it is not possible to compile it, or
use the version 0.9.9 of ubuntu, any tips?

thank you


I do not know why it is not in stable, but I can see it in unstable:
http://packages.debian.org/search?lang=fr&searchon=names&keywords=compiz


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8dffc8b600169821140771f0c4db8...@neutralite.org



Re: You can have any color you want - as long as it's Gnome?

2013-10-07 Thread berenger . morel



Le 07.10.2013 15:12, Curt a écrit :

On 2013-10-07, Oliver Fairhall  wrote:


I want a light weight DE, and was thinking to use XFCE (been fine on 
my
other distros). I'm at best an intermediate level with Linux, so 
will
likely struggle with not installing Gnome (really don't want Gnome 
or KDE).




With the wheezy netinstaller you simply choose 'Advanced Options' on 
the

first page you're presented with, then 'Alternative desktop
environments," then 'Xfce.' No Gnome.

With this version of the netinstaller at least that's the way it's 
done:


debian-wheezy-DI-rc1-amd64-netinst.iso


The problem is that what you describe does not work as expected, 
according to the OP.


It seem that installing XFCE DE will also install 
network-manager-gnome, which recommends the gnome desktop. Because 
recommendations are automatically installed, this will result in having 
gnome installed, and Oliver wants "light weight DE" and "not installing 
Gnome".



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/cbc7f3883ca81c4b0a826faa1c3d1...@neutralite.org



Re: You can have any color you want - as long as it's Gnome?

2013-10-07 Thread berenger . morel



Le 07.10.2013 16:02, Curt a écrit :

On 2013-10-07, Curt  wrote:


With the wheezy netinstaller you simply choose 'Advanced Options' 
on

the
first page you're presented with, then 'Alternative desktop
environments," then 'Xfce.' No Gnome.

With this version of the netinstaller at least that's the way it's
done:

debian-wheezy-DI-rc1-amd64-netinst.iso


The problem is that what you describe does not work as expected,
according to the OP.


I followed the procedure described above, installing to a usb key, 
and when
I booted the usb key I was in the desktop environment of my choice 
and not in

Gnome.

But I can't remember whether I chose Xfce or LXDE (but he said for 
the

both of them he ended up in Gnome).

So what I described worked as expected for me (I think--I didn't 
check to see

whether all of Gnome got installed somehow behind my back).

Anyhoo, the plot sickens.



Actually, rereading his post, I don't see him saying he followed the
procedure I followed anywhere; rather he says he installed testing
without a DE, then tried to install a lightweight desktop with 
aptitude,

so this is something of time-waster, isn't it?


If so, yes, but I think that he first tried the procedure you 
described, because of those words:


I needed to do a fresh installation of Debian on two systems for 
friends this
weekend. I tried both stable (7.1) and the 10/02/2013 daily of 
testing -- both

of them the netinst image.

For both stable and testing I tried both LXDE and Xfce desktop 
environment
installations. But when the systems rebooted, I was at the Gnome 
desktop.

snip
Okay. So I installed Debian testing without a DE, and then tried to 
add
xfce-desktop via aptitude. I saw that I was still going to get Gnome 
and

canceled the operation.


I think that he meant that he first tried the normal way, and when 
seeing that it was not working tried testing with installation from 
aptitude. But I might have misunderstood something, would not be the 
first time and probably not the last too :)




apt-get --no-install-recommends

No?


I guess. Must admit that my installation process is not really the 
fastest one: installing no package at first, then on reboot disabling 
automatic installation of recommended packages and finally install what 
I know I'll use :)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/b0b6600534f2ca73551ff3cf79d74...@neutralite.org



Re: schroot

2013-10-07 Thread berenger . morel



Le 07.10.2013 18:38, shawn wilson a écrit :
This is at the top of every config file, but I can't find it 
documented:


. "$SETUP_DATA_DIR/common-data"
. "$SETUP_DATA_DIR/common-functions"
. "$SETUP_DATA_DIR/common-config"

Where is this being sourced from (ie, where is the 'common-data'
file?) and (more important) where is this documented?


Doing "echo $SETUP_DATA_DIR" should help you, I think. And for 
documentation, reading about shell will also help you.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/6c8fe63572c303ff7017f63dd8d19...@neutralite.org



Re: Debian wheezy boot problem - Boot sector not identified by board

2013-10-07 Thread berenger . morel



Le 07.10.2013 09:54, Roland RoLaNd a écrit :

All,

I have an intel dh77kc board. it previously had windows 7 installed
on it.
i tried installing debian wheezy net install. installation goes
perfectly fine up untill reboot.
once reboot is done, i get " Initializing and establishing link"
and immediately goes into network boot
I tried resetting bios settings to default, i even upgraded the bios
itself. and changed from AHCI to IDE
nothing is working so far.

Note: i thought it's not debian specific, by installing windows 7
again. and it worked fine..

Any hint on what might be going on ?


Are you sure the flag "boot" is set on the right partition?


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5d36dfb019ac15e1577048f18e5cf...@neutralite.org



Re: schroot

2013-10-07 Thread berenger . morel

Le 07.10.2013 18:59, Shawn Wilson a écrit :

berenger.mo...@neutralite.org wrote:



Le 07.10.2013 18:38, shawn wilson a écrit :

This is at the top of every config file, but I can't find it
documented:

. "$SETUP_DATA_DIR/common-data"
. "$SETUP_DATA_DIR/common-functions"
. "$SETUP_DATA_DIR/common-config"

Where is this being sourced from (ie, where is the 'common-data'
file?) and (more important) where is this documented?


Doing "echo $SETUP_DATA_DIR" should help you, I think. And for
documentation, reading about shell will also help you.



 I'm guessing this means it's exported by some schroot internal
mechanism inside the schroot? I'm not sure what I don't know about
bash that would help here? This doesn't seem to be an export bash
knows about?


I have no idea about what are the files you are speaking about, but the 
$ prefix usually indicates a variable in shell, and shell scripts are 
widely used in the system. To find what file could export that variable, 
try a grep -r SETUP_DATA_DIR, it might help you find which file uses 
that variable.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/c1e9972943119920c856a4233d103...@neutralite.org



Re: schroot

2013-10-07 Thread berenger . morel

Le 07.10.2013 19:50, shawn wilson a écrit :

Not a bad idea. However:
find / -type f -print0 | xargs -0 -i{} -P 10 grep -H 
'SETUP_DATA_DIR='

{} 2> /dev/null

found nothing.

On Mon, Oct 7, 2013 at 1:05 PM,   
wrote:

Le 07.10.2013 18:59, Shawn Wilson a écrit :


berenger.mo...@neutralite.org wrote:




Le 07.10.2013 18:38, shawn wilson a écrit :


This is at the top of every config file, but I can't find it
documented:

. "$SETUP_DATA_DIR/common-data"
. "$SETUP_DATA_DIR/common-functions"
. "$SETUP_DATA_DIR/common-config"

Where is this being sourced from (ie, where is the 'common-data'
file?) and (more important) where is this documented?



Doing "echo $SETUP_DATA_DIR" should help you, I think. And for
documentation, reading about shell will also help you.



 I'm guessing this means it's exported by some schroot internal
mechanism inside the schroot? I'm not sure what I don't know about
bash that would help here? This doesn't seem to be an export bash
knows about?



I have no idea about what are the files you are speaking about, but 
the $
prefix usually indicates a variable in shell, and shell scripts are 
widely
used in the system. To find what file could export that variable, 
try a grep
-r SETUP_DATA_DIR, it might help you find which file uses that 
variable.


Do not send me private mail for something like that, it could interest 
someone else on the list.
Honestly, I can not help you more that that, I do not use chroot very 
often, and do not know what is schroot. If the variable is not defined, 
then maybe it have a default value.
Maybe if you find other files containing simply the SETUP_DATA_DIR text 
(without '=' or '$') you could find more hints.


PS: do the giant line you posted above make the same thing as "grep -r 
'SETUP_DATA_DIR=' 2>/dev/null" ? If yes, it seems quite complex for what 
it does...



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/46bc5f77d3e4a1c01373b74d77769...@neutralite.org



Re: apt-get vs. aptitude

2013-10-08 Thread berenger . morel



Le 08.10.2013 13:33, Florian Lindner a écrit :

Hello,

Since I'm about to setup a new server using current stable wheezy, I 
want to

recheck some of debian knowledge.

What is the prefered tool for installing on the CLI? apt-get or
aptitude? Last
time I read about it, it was aptitude, due to better dependency
checking. What
is the current state? apt-get or aptitude? Does it matter? What about 
using

both?

Thanks!

Florian


Both are front-end for dpkg, so there will be no real difference on the 
system after using one, the other or both.

I use both without any problem.

The differences between them is that aptitude:
* is slower
* have more built-in features
* have a ncurse interface

The slowness of aptitude should not be noticeable on a fast computer 
like a server, so you can probably forgot about that.
Built-in features aptitude have that apt-get *might* (I am not sure) 
not have *built-in* (but you can add apt-* stuff to add them I think) 
are some tools to support debtags (if installed), research of solutions 
when something is broken, and... I do not know what else :)


The aptitude's ncurse GUI is nice, at least for an end-user like me. On 
a server, where I think you probably know what and why to 
install/update, I doubt it will be.
But for the end user, it allows to quickly search and find a package 
depending on it's name, sorted by either categories or by debtags.

The problem imho of that interface is that:
* if you have any broken package ( by a modification you put, but did 
not validated ) it will be slow as hell ( it will checks solutions at 
each move for nothing )
* debtags are only shown as a tree, there is no really good interface 
to manage them ( but it is better than nothing )
* if you need multiarch, you will have to brows each package multiplied 
by the number of archs you use. Sadly, the interface did not used a way 
like the one used for versions for that...


Conclusion:
On a production server, I would use apt-get: faster.
For testing needs, or R&D, I would go for aptitude and use it's 
interface to browse packages and find solutions depending on the current 
state of the system ( it is easier to check which dependencies are being 
installed with aptitude's ncurse GUI than on a command line ).


Hope it helps.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5129e9683e203ac0ad56c260d...@neutralite.org



Re: apt-get vs. aptitude

2013-10-08 Thread berenger . morel



Le 08.10.2013 15:28, Whit Hansell a écrit :

On 10/08/2013 08:57 AM, Morten Bo Johansen wrote:

On 2013-10-08 Florian Lindner wrote:


What is the prefered tool for installing on the CLI? apt-get or
aptitude? Last time I read about it, it was aptitude, due to better
dependency checking. What is the current state? apt-get or 
aptitude?

Does it matter? What about using both?
Both use libapt-pkg, so when used from the command line I don't 
think it

matters which you use.

   Morten



I had asked the same question a year ago or so and changed from
aptitude to apt-get with using dist-upgrade for upgrades on both. On
aptitude I was getting recommended packages no being installed but
they are all installed using apt-get.  I had always wondered about 
the

recommends being held back and have found no problem having them
installed using apt-get.

Hope that helps.
whit


This only depends on your configuration. Aptitude's default is to 
install recommended packages, as for apt-get I think.
For apt-get, there is a command-line option "--no-install-recommends" 
which allows you to not install those. Probably aptitude have something 
like this too ( I only know about the GUI's option ) and they probably 
both are able to store that choice in some configuration file.


Installing recommended packages will probably not give you problems, if 
you do not care about your computer performances or the bandwith.
Automatically adding recommended packages will add, and enable, 
services that you could never need, and on a server, adding services 
often means adding risks of bugs, and so of hacks.
On my own computer, where I do not mind about being hacked ( at least, 
not at a point that I want to reinforce everything ) it will cost me 
resources: system and applications will be slower to start, and might 
even saturate my memory ( at least on my netbook with it's 1GB or ram ). 
It can also lead to behavior that I do not want.


But, this will add features you could like, too. I see recommended 
packages as suggestions, and suggested packages as "see also". They are 
not evil, they simply need thinking before enabling all of them, if you 
want an efficient system ( lacking some of them will also need to 
inefficient systems ).


Little example from here:
Installing "only" network-manager-gnome and it's recommended packages 
leads to 345 packages automatically installed, 550MiB to download, and 
193MiB to install (I use a tiling window manager and am as careful as my 
skills allows to avoid useless -for me- stuff, so, no complete DE).
Installing it without recommended packages leads to 20 packages 
automatically installed, 28.2MiB to download, and 14.1MiB to install.
Sounds like a nice hint about what is wrong with automatic installation 
of recommended packages for advanced users ;) ( here, it means 40 min to 
spend in downloads, at least, and it uses Debian's mirrors' bandwidth 
for nothing, too )



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/61f8caba26eb72e4dc9ad27dd5290...@neutralite.org



Re: I have an Openoffice question for small business.

2013-10-08 Thread berenger . morel



Le 08.10.2013 14:10, Ezequiel a écrit :

Hi all:

I am Sysadmin at a small business. We have a complete mail-web-vpn
infrastructure and my boss is happy with it. I guess we are a
successful case of open software use in the "real world"

But -of course, there is always an issue- my users keep complaining
about OpenOffice migration to Libre Office. They even complain if I
change OO from 3.2 to 3.3. I believe there were major changes in that
version.

The question is: Is there any way of freezing OO version indefinetly?

I am currently using oldstable OO but I guess my time is going short.
What will happen when they release the new version of debian? I don't
know what to do...

Thanks in advance for any advices.

Zeke

PD: My native language is not English, I'm sorry for any mistakes in
my writing.


First, I do want to say that you should follow previous advice to try 
to convince your users :)


But since you did not ask "how to convince my users to upgrade", here 
is what you might want:

Use the /etc/apt/preferences file.

I used it some times ago but can not remember the exact syntax, but you 
should be able to quickly find some samples on debian's forums. Search 
for apt-pinning (the name of the technique iirc) and you should find 
nice examples in debian's forums. This technique is more often used to 
only use some packages from testing/unstable/experimental on stable, but 
you should be able to adapt it for your needs easily: simply give very 
low priorities to the packages you want.


But you should know that it also means that OO (or LO) dependencies 
will also need to be frozen, and this might avoid other other updates, 
in turn. Have fun :)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/f69b149088fbb64b728c572c1ee23...@neutralite.org



Re: I have an Openoffice question for small business.

2013-10-08 Thread berenger . morel

Le 08.10.2013 16:14, berenger.mo...@neutralite.org a écrit :

Le 08.10.2013 14:10, Ezequiel a écrit :

Hi all:

I am Sysadmin at a small business. We have a complete mail-web-vpn
infrastructure and my boss is happy with it. I guess we are a
successful case of open software use in the "real world"

But -of course, there is always an issue- my users keep complaining
about OpenOffice migration to Libre Office. They even complain if I
change OO from 3.2 to 3.3. I believe there were major changes in 
that

version.

The question is: Is there any way of freezing OO version 
indefinetly?


I am currently using oldstable OO but I guess my time is going 
short.
What will happen when they release the new version of debian? I 
don't

know what to do...

Thanks in advance for any advices.

Zeke

PD: My native language is not English, I'm sorry for any mistakes in
my writing.


First, I do want to say that you should follow previous advice to try
to convince your users :)

But since you did not ask "how to convince my users to upgrade", here
is what you might want:
Use the /etc/apt/preferences file.

I used it some times ago but can not remember the exact syntax, but
you should be able to quickly find some samples on debian's forums.
Search for apt-pinning (the name of the technique iirc) and you 
should

find nice examples in debian's forums. This technique is more often
used to only use some packages from testing/unstable/experimental on
stable, but you should be able to adapt it for your needs easily:
simply give very low priorities to the packages you want.

But you should know that it also means that OO (or LO) dependencies
will also need to be frozen, and this might avoid other other 
updates,

in turn. Have fun :)


Sorry for my self reply, but I just thought of that:
Another solution, not the easiest one but which would avoid freezing 
dependencies, would be download source of OO and compile it ( not on all 
computers of course, only on yours ) and then distributing the binary 
through a package.
To download source and install libraries needed, you can do something 
like:
aptitude build-dep openoffice ( will install development libraries OO 
will need )

apt-get source openoffice ( will download source code for openoffice )

apt-get source will download an archive with source code, so untar it, 
and then probably do the old "./configure && make".
Next steps is to build a package, but I can not help you on those, 
however a lot of people here can probably.
And the last one is to distribute it. For that, you might want to setup 
a local repository, add your OO package in it, and add that repo to the 
sources.list ( or sources.list.d/local_OO.list, if you prefer ) of the 
desktops.
That procedure is more complex than the one with file preferences, but 
can survive longer without giving you cascading version problems in 
future.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/91db1f8ed6f5b21c861a2956ef2ca...@neutralite.org



Re: apt-get vs. aptitude

2013-10-08 Thread berenger . morel



Le 08.10.2013 16:15, Richard Owlett a écrit :

berenger.mo...@neutralite.org wrote:



Le 08.10.2013 13:33, Florian Lindner a écrit :

Hello,

Since I'm about to setup a new server using current stable
wheezy, I want to
recheck some of debian knowledge.

What is the prefered tool for installing on the CLI? apt-get or
aptitude? Last
time I read about it, it was aptitude, due to better dependency
checking. What
is the current state? apt-get or aptitude? Does it matter? What
about using
both?

Thanks!

Florian


Both are front-end for dpkg, so there will be no real difference
on the system after using one, the other or both.
I use both without any problem.

The differences between them is that aptitude:
* is slower
* have more built-in features
* have a ncurse interface

The slowness of aptitude should not be noticeable on a fast
computer like a server, so you can probably forgot about that.
Built-in features aptitude have that apt-get *might* (I am not
sure) not have *built-in* (but you can add apt-* stuff to add
them I think) are some tools to support debtags (if installed),
research of solutions when something is broken, and... I do not
know what else :)

The aptitude's ncurse GUI is nice, at least for an end-user like
me. On a server, where I think you probably know what and why to
install/update, I doubt it will be.
But for the end user, it allows to quickly search and find a
package depending on it's name, sorted by either categories or by
debtags.
The problem imho of that interface is that:
* if you have any broken package ( by a modification you put, but
did not validated ) it will be slow as hell ( it will checks
solutions at each move for nothing )
* debtags are only shown as a tree, there is no really good
interface to manage them ( but it is better than nothing )
* if you need multiarch, you will have to brows each package
multiplied by the number of archs you use. Sadly, the interface
did not used a way like the one used for versions for that...

Conclusion:
On a production server, I would use apt-get: faster.
For testing needs, or R&D, I would go for aptitude and use it's
interface to browse packages and find solutions depending on the
current state of the system ( it is easier to check which
dependencies are being installed with aptitude's ncurse GUI than
on a command line ).

Hope it helps.



Don't know if it helps the OP, but it does help me. Thank you.


Nice :)


I'm experimenting with a very lean idiosyncratic install. It sounds
as aptitude will be appropriate for me. Off to read man pages etc ;)


Don't copy me! xD
More seriously, without aptitude, I would probably not be with debian ( 
probably I would have stayed with windows, that I known better some 
years ago ), and it is really that tool which allowed me to have fast as 
lighting computers built from low-price hardware ( but no one that I 
know could use any of my systems if I am far away ).


You might also be interested by dselect, I have read about it several 
times, but never took enough time to really discover it.


My first moves when installing a new system: uncheck all ( yes, 
including basic tools ) checkboxes while installing, booting on the new 
system, disabling in aptitude the automatic install of recommended 
stuff, and install only packages that I invoke by myself.


Sometimes I take some fun to also purge all packages ( yes, all of 
them: go to root entry of aptitude and then press '_' ) to add them back 
one by one in the preview, marking all packages I do not remove as 
automatically installed ( so that they'll go away when there will be no 
reason to keep them ). It's nice to see that Debian still install some 
tools which are not really needed when you uncheck everything at install 
time.


Be careful, that way to install a computer is the best one to install 
broken systems :) but I'll bet that you know that ( it's more a 
disclaimer for people who could fall on that mail )



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/93b69918bea6daefb0cdb0a57a8bb...@neutralite.org



apt-pinning, strange behavior

2013-10-08 Thread berenger . morel
Since I had to reinstall from my last kernel error, I decided to stay 
with stable on that computer, but I need some softwares in less outdated 
versions, like development libraries or i3 ( this one is not a need but 
a question of comfort, I admit ), so I want to use apt-pining.


I have set all packages from stable to a priority of 900 and testing 
packages with 500.
But tzdata wants to upgrade, for an unknown reason. Explicitly making 
it to a priority of 900 for stable fixes that, but I can not understand 
why it is needed?



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/28f8d241e1c1bcf2f24cb2073969d...@neutralite.org



Re: apt-pinning, strange behavior

2013-10-08 Thread berenger . morel



Le 08.10.2013 22:42, Sven Joachim a écrit :

On 2013-10-08 19:06 +0200, berenger.mo...@neutralite.org wrote:

Since I had to reinstall from my last kernel error, I decided to 
stay

with stable on that computer, but I need some softwares in less
outdated versions, like development libraries or i3 ( this one is 
not

a need but a question of comfort, I admit ), so I want to use
apt-pining.

I have set all packages from stable to a priority of 900 and testing
packages with 500.
But tzdata wants to upgrade, for an unknown reason. Explicitly 
making

it to a priority of 900 for stable fixes that, but I can not
understand why it is needed?


I don't know either, but "apt-cache policy tzdata" should explain it.

Cheers,
   Sven


Thanks for the hint, I had forgotten about apt-cache policy.
I finally understood, why the update was on the run:

tzdata:
  Installé : 2013d-0wheezy1
  Candidat : 2013d-1
 Table de version :
 2013d-1 0
500 http://ftp2.fr.debian.org/debian/ testing/main amd64 
Packages

 *** 2013d-0wheezy1 0
500 http://ftp2.fr.debian.org/debian/ stable-updates/main amd64 
Packages

100 /var/lib/dpkg/status
 2013c-0wheezy1 0
900 http://ftp2.fr.debian.org/debian/ stable/main amd64 
Packages


Version 2013d-0wheezy1 had the same priority as testing one so testing 
was installed because more recent.
Now, I wonder why I have 3 versions of that package listed when I only 
have 2 sources enabled? Could it be because of stable, stable/updates 
and stable-updates repositories? ( I am not used to stable, so I do not 
have the "updates" repos usually )
And also why I have a wheezy version with a priority of 500... I can 
not even find the 2013d-0wheezy1 in debian packages...



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/9dc2fe2bca18a98dd48abc1de5313...@neutralite.org



which MTA to choose for a simple client?

2013-10-08 Thread berenger . morel

Hi.
I finally decided myself to install a software to manage my mails.

So, I think I'll go for mutt: it appears quite often on the list ( so I 
might ask if I have problems, before trying another one when everything 
will be ok ) and runs into a terminal, which makes it perfect for me 
since I use a tiling window managers: it will never spawn ugly dialog in 
my face for a reason or another, and for the situations when I simply 
want to run a TTY without X, it will fit perfectly too.


But now, it seems ( says this article 
http://nongeekshandbook.blogspot.fr/2009/02/mutt-email-client-that-sucks-less.html 
) that it needs 2 other tools: one to fetch mails from server, and 
another one to send them.
For the fetcher, I am surprised that debian does not seems to recommend 
or suggest using one, so I will not spend time on that -for now at 
least- and will do as the article says, unless I discover something 
interesting in the process.
But for the tool to send mails, things are different: I can count 16 
alternatives. Some are obviously wrong for my use, like lsb-invalid-mta, 
postfix or exim ( those last ones are probably too big for my simple 
usage, they seems designed for big boxes where mailing is an important 
task ), but even after removing some obvious ones, I still have a lot of 
choice.


So, here is my question:
What would you use as a MTA on a Debian system made for an end-user?

Of course, I could take one of them at random and try to go with it, 
but I would like to take the occasion to learn basics about that, 
without installing a tool from which I will never use or understand 10% 
of the features...

So, I would like something which:
_ supports IMAP, POP3 and SMTP ( this does not sound excessive I think, 
but if there are other important protocols, I do not even know their 
existences or uses )
_ is not a daemon running constantly: why should I have a daemon 
running to send mail when I am not connected to Internet or not taking 
care of my mails? Something which is started by the client ( MUA it 
seem? ) is good enough for me and does not consume time when starting or 
shutting down my computers.
_ is lightweight, because I always aim to have a system which let all 
possible resources to my compilers, and which respect my batteries. I 
bet that if I can still survive 4H with wifi after 3 years of intensive 
use, it is partly because I do not use heavy softwares.
_ is configured by raw text in the good old UNIX way because I have 
learn so many from Debian's configuration files and their comments, 
which are very useful when you messed everything and can not even access 
Internet :)


Does it even exists? If not, what is, in your opinions, the better to 
fit those goals?

Thanks.

PS: sorry for the long description of my request, but I tried to be as 
complete as possible. Hopefully it makes things I aim for more clear...



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/971376ed67115d6ae8b5d28130021...@neutralite.org



Re: google-chrome-unstable apparently removes its executable

2013-10-08 Thread berenger . morel



Le 09.10.2013 02:43, Stephen Allen a écrit :

On Sun, Oct 06, 2013 at 11:42:15AM -0400, Patrick Wiseman wrote:

Updating two testing systems this morning, I found that my Chrome
suddenly became unable to find any web pages. Attempting to 
relaunch,

it wasn't there!

'which google-chrome' returned nothing. I tried 'sudo apt-get 
install

google-chrome-unstable', but it was already installed. Finally, on
both systems, I did 'sudo apt-get install google-chrome-stable' and
now have a working Chrome again.

I guess that's what I get for being willing to live with unstable!


On the other hand, which will only show you the command with the exact 
name you provided it, right?
You would have better informations if you simply list the content of 
the package and grep it with /usr/bin, so that it would show the 
binaries.
Something like "$apt-file list google-chrome-unstable |grep '/usr/bin'" 
should do the trick, and says if the command you are searching from has 
not simply be renamed.


( sorry Stephen to reply to you but it seem I have deleted the OP's 
post )



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/580087039fc63d642e876e55215aa...@neutralite.org



Re: apt-pinning, strange behavior

2013-10-09 Thread berenger . morel



Le 09.10.2013 11:17, Marko Randjelovic a écrit :

On Wed, 09 Oct 2013 00:12:46 +0200
berenger.mo...@neutralite.org wrote:




Le 08.10.2013 22:42, Sven Joachim a écrit :
> On 2013-10-08 19:06 +0200, berenger.mo...@neutralite.org wrote:
>
>> Since I had to reinstall from my last kernel error, I decided to
>> stay
>> with stable on that computer, but I need some softwares in less
>> outdated versions, like development libraries or i3 ( this one is
>> not
>> a need but a question of comfort, I admit ), so I want to use
>> apt-pining.
>>
>> I have set all packages from stable to a priority of 900 and 
testing

>> packages with 500.
>> But tzdata wants to upgrade, for an unknown reason. Explicitly
>> making
>> it to a priority of 900 for stable fixes that, but I can not
>> understand why it is needed?
>
> I don't know either, but "apt-cache policy tzdata" should explain 
it.

>
> Cheers,
>Sven

Thanks for the hint, I had forgotten about apt-cache policy.
I finally understood, why the update was on the run:

tzdata:
   Installé : 2013d-0wheezy1
   Candidat : 2013d-1
  Table de version :
  2013d-1 0
 500 http://ftp2.fr.debian.org/debian/ testing/main amd64
Packages
  *** 2013d-0wheezy1 0
 500 http://ftp2.fr.debian.org/debian/ stable-updates/main 
amd64

Packages
 100 /var/lib/dpkg/status
  2013c-0wheezy1 0
 900 http://ftp2.fr.debian.org/debian/ stable/main amd64
Packages

Version 2013d-0wheezy1 had the same priority as testing one so 
testing

was installed because more recent.
Now, I wonder why I have 3 versions of that package listed when I 
only
have 2 sources enabled? Could it be because of stable, 
stable/updates
and stable-updates repositories? ( I am not used to stable, so I do 
not

have the "updates" repos usually )
And also why I have a wheezy version with a priority of 500... I can
not even find the 2013d-0wheezy1 in debian packages...




The answer to your question is in files /var/lib/apt/lists/*_Release.
You can notice the difference:

Suite: stable
Suite: stable-updates


Thanks. I did not thought that it would be considered as a different 
repo, which is absent from http://packages.debian.org but it explains 
the behavior I have.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1a5f89a7d2f551b1158e86eecc6df...@neutralite.org



Re: which MTA to choose for a simple client?

2013-10-09 Thread berenger . morel



Le 09.10.2013 04:48, Jerry Stuckle a écrit :

On 10/8/2013 9:13 PM, berenger.mo...@neutralite.org wrote:

Hi.
I finally decided myself to install a software to manage my mails.

So, I think I'll go for mutt: it appears quite often on the list ( 
so I
might ask if I have problems, before trying another one when 
everything

will be ok ) and runs into a terminal, which makes it perfect for me
since I use a tiling window managers: it will never spawn ugly 
dialog in
my face for a reason or another, and for the situations when I 
simply

want to run a TTY without X, it will fit perfectly too.

But now, it seems ( says this article

http://nongeekshandbook.blogspot.fr/2009/02/mutt-email-client-that-sucks-less.html
) that it needs 2 other tools: one to fetch mails from server, and
another one to send them.
For the fetcher, I am surprised that debian does not seems to 
recommend

or suggest using one, so I will not spend time on that -for now at
least- and will do as the article says, unless I discover something
interesting in the process.
But for the tool to send mails, things are different: I can count 16
alternatives. Some are obviously wrong for my use, like 
lsb-invalid-mta,

postfix or exim ( those last ones are probably too big for my simple
usage, they seems designed for big boxes where mailing is an 
important
task ), but even after removing some obvious ones, I still have a 
lot of

choice.

So, here is my question:
What would you use as a MTA on a Debian system made for an end-user?

Of course, I could take one of them at random and try to go with it, 
but
I would like to take the occasion to learn basics about that, 
without
installing a tool from which I will never use or understand 10% of 
the

features...
So, I would like something which:
_ supports IMAP, POP3 and SMTP ( this does not sound excessive I 
think,

but if there are other important protocols, I do not even know their
existences or uses )
_ is not a daemon running constantly: why should I have a daemon 
running
to send mail when I am not connected to Internet or not taking care 
of
my mails? Something which is started by the client ( MUA it seem? ) 
is
good enough for me and does not consume time when starting or 
shutting

down my computers.
_ is lightweight, because I always aim to have a system which let 
all
possible resources to my compilers, and which respect my batteries. 
I
bet that if I can still survive 4H with wifi after 3 years of 
intensive

use, it is partly because I do not use heavy softwares.
_ is configured by raw text in the good old UNIX way because I have
learn so many from Debian's configuration files and their comments,
which are very useful when you messed everything and can not even 
access

Internet :)

Does it even exists? If not, what is, in your opinions, the better 
to

fit those goals?
Thanks.

PS: sorry for the long description of my request, but I tried to be 
as
complete as possible. Hopefully it makes things I aim for more 
clear...





I use Exim on my servers - and they don't handle much mail, and have
ever since I started with Debian.  It can be a real pain to configure
due to its flexibility, but the default configuration will probably
work for you.  And SpamAssassin installs nicely into the Exim4
configuration.


I guess it will probably works out of the box, yes. I trust Debian a 
lot for that, but I am the tinkerer kind of users, so it will quickly be 
out-of-order because of my changes in the search of the optimal system 
;)



I've done a fair amount of custom configuration, especially to keep
spam down.  But that's also been done over a period of years, not all
at one time.


Since my intent ( I forgot to explicitly mention that ) is to simply 
get my mails from an Internet mailbox ( the one I'm currently using ) I 
do not really feel the need to redo the job the admins made. I'll prefer 
to move them to the dedicated section of the distant MTA where a bot 
will learn from them, which will contribute to overall quality of the 
service instead of only my computer.



One thing you do need to be careful with, no matter which MTA you
use. Don't make it an open relay - you'll soon become a source of
SPAM.  And you should take steps to prevent bots from guessing your
userid and password (I use fail2ban).

Jerry


Should I guess from this that I'll have to configure my router for some 
port forwarding? It's not something hard, but it will not be possible 
everywhere I could go with the netbook... But I'll keep that in mind 
anyway. It must not be so hard to only relay mail from localhost...


Thanks for your reply


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/49270eaeb7dd8c2695ea11165a1b3...@neutralite.org



Re: which MTA to choose for a simple client?

2013-10-09 Thread berenger . morel



Le 09.10.2013 04:57, Celejar a écrit :

What would you use as a MTA on a Debian system made for an end-user?


I've used Exim, basically because it's (was?) the Debian default.


I do not want a default software just because it is the default. 
Otherwise I would have be perfectly happy with windows, and then on 
Debian with gnome and then... well. I'm not.


Exim is probably a very good and complete - and so complex - tool, I do 
not doubt it, but I really feel like it is over-engineering to use it on 
a computer which does does not have any mail server task (being for a 
gigantic enterprise, a small one or a family does not change the idea of 
mail server in my mind. I have a simplistic vision of admin tasks, I 
know).
And I try to avoid OE on my computers and softwares. For planet's 
ecology and my economy's sake, OE is bad :p


Plus, I want to continue my linux road with more and more understanding 
of my system. Too complex tools is a problem for that when it's not your 
main task to maintain them.



_ is not a daemon running constantly: why should I have a daemon
running to send mail when I am not connected to Internet or not 
taking

care of my mails? Something which is started by the client ( MUA it
seem? ) is good enough for me and does not consume time when 
starting or

shutting down my computers.
_ is lightweight, because I always aim to have a system which let 
all


Assuming you're using a smarthost (relay host), you
can use a relay server such as ssmpt, msmtp or nullmailer which I
believe all meet these two conditions.


By relay host, you mean the server from which I am sending this mail ( 
through a web interface )? If so, yes, I only want to discuss with it, 
except if there is some advantage ( for me or that server ) to directly 
send mails to their final server.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/023b440d9be6811932809326b7a27...@neutralite.org



Re: which MTA to choose for a simple client?

2013-10-09 Thread berenger . morel

Le 09.10.2013 11:08, Jochen Spieker a écrit :

berenger.mo...@neutralite.org:


I finally decided myself to install a software to manage my mails.


Good luck! My impression is that this is one of the few things that 
have

not become considerably easier on Linux in the last ten years.

Mutt is still a good choice today if you can live with the limits of 
a

terminal. I use it myself, but not completely exclusively.


I can live enough with terminal's limits to no longer using IDE for 
programming, no longer using graphical file-manager to manage my files, 
and even to start user installed applications.
In fact, I simply can not support the limits of graphical GUIs: they 
force me to grab my mouse for minor tasks that I should be able to do 
with keyboard.
I still have the web browser to adapt on my todo-list, opera is nice 
but still have failures in this regard.
Plus, GUIs are rarely efficient when you use dark background, and white 
background hurts less my eyes at night when switching from the dark ones 
that I like to use for terminals.



You need to address at least these questions:

- How do I plan to access mails using mutt (IMAP or local storage?)


IMAP. Local storage is probably nice for some uses, but I'll be honest: 
I can not see the interest of reading all my mails without Internet 
access. I do not say there are no uses for that, simply, they are not my 
uses.


that part completely if you mail is already on an IMAP-capable 
server.

Mutt can access that as well. Works fine for me. Some people complain
about speed, but that doesn't apply to me. As long as you have header
caching enabled in mutt and your mailboxes are smaller than, say, 
10,000

mails, it should be fast enough.


So it seems that it is perfect for me. I often clean my mails so have 
less than 200, and the server is IMAP-capable.



This is handy if you use several different mail
providers


Few months ago, I had something like 4 or 5 addresses. It was a ugly 
and boring mess, so I deleted most of them and only kept this one. No 
ads, associative work, no headaches because standard features of server 
are configured in ways to only work by some software of the same 
company's defaults, etc.
I no longer have need for multiple addresses, now when I need a garbage 
mail, I ask jetable.org to give me a temporary one.



or if several people use the mail server.


It is for a personal use. I want an advanced end-user desktop, not a 
universal computer able to make tea and coffee.



For the fetcher, I am surprised that debian does not seems to
recommend or suggest using one, so I will not spend time on that
-for now at least- and will do as the article says, unless I
discover something interesting in the process.


Setups like yours have come out of fashion in the last few years. 
People

either use GUI clients or web interfaces these days. :-/


I use web interfaces for now. You can not be surprised of that when you 
see the difference of time needed between a web-client or a heavy GUI 
client ( thunderbird, MS outlook... I do not mind ) versus client which 
only does one thing but does it correctly, after X hours of 
configuration and choice. Having tools which does their stuff correctly 
imply that you have to find and understand them. That's what I want to 
end-up for my mail client.



_ is not a daemon running constantly: why should I have a daemon
running to send mail when I am not connected to Internet or not
taking care of my mails?


Don't be bothered about that. My central Postfix server in my LAN 
uses
less than 5MB of RAM. It really doesn't matter. Total CPU usage in 
100

days of uptime is about 36 seconds.


5MiB are nothing, true. It is only 0.5% of the capacity of my netbook 
and you can divide that by 4 for my desktop: 1GB and 4GB or ram. But I 
also have a very old computer with less than 200MB which is surprisingly 
not so slow with a Debian on it. Here, it is not so negligible.
Plus, if for every single task I want to do frequently, I think '5MB 
are nothing', then I'll quickly come to have, say, 30*5 = 150MB, which 
is not nothing. Of course, on my system I guess that it won't take 5MB, 
since I have less mails to process, and my systems are often restarted. 
My average uptime is probably close to 8H for all my computers.



Something which is started by the client (
MUA it seem? ) is good enough for me and does not consume time when
starting or shutting down my computers.

_ is lightweight, because I always aim to have a system which let
all possible resources to my compilers, and which respect my
batteries. I bet that if I can still survive 4H with wifi after 3
years of intensive use, it is partly because I do not use heavy
softwares.


Ok, I see where you are coming from. But believe my when I say that a
simple full-fledged MTA does not use any considerable amount of
resources (unless you constantly bombard it with more than ten mails 
per

second).

Startup time is also negligible.


_ is configured by 

Re: which MTA to choose for a simple client?

2013-10-09 Thread berenger . morel

Le 09.10.2013 14:05, Curt a écrit :

What use do you find for an MTA if you're using IMAP?

I do not know, really.
I simply have read here and there that it was needed, and since I have 
noticed so much choice in aptitude, I asked here to understand what 
solution would be the best.


If it is to not use any stuff like that, then I am really happy.


Of course, there's the question of local daemons (etc.) sending
system messages to
root (there are small, simple solutions for that).


I am happy without that, since at least 2 years. Having a desktop 
computer, and not a server, I have removed most daemons which are 
installed by default.
I still have near 100 processes running in my back, of course, and I 
would like to know what are their roles, but I think they are not 
daemons: /usr/sbin/service --status-all gives me less than 40 lines ( on 
the fresh install of my desktop, on which there things that I do not use 
like cron for example ) and not all of them are running ( if I interpret 
correctly the ouput, which is not certain ).


As far as having multiple smart hosts, alpine (my mail client) 
handles
that with "roles" easily enough.  I'm sure (well, I'm optimistic) 
Mutt

can do the same thing.


I am trying to put myself into that world of local mail clients, so I 
can not say if it does or not :)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/617e24811f5419695452562837352...@neutralite.org



Re: apt-get vs. aptitude

2013-10-09 Thread berenger . morel



Le 09.10.2013 15:39, Richard Owlett a écrit :

berenger.mo...@neutralite.org wrote:



Le 08.10.2013 16:15, Richard Owlett a écrit :
[snip]


I'm experimenting with a very lean idiosyncratic install. It
sounds
as aptitude will be appropriate for me. Off to read man pages
etc ;)


Don't copy me! xD
More seriously, without aptitude, I would probably not be with
debian ( probably I would have stayed with windows, that I known
better some years ago ), and it is really that tool which allowed
me to have fast as lighting computers built from low-price
hardware ( but no one that I know could use any of my systems if
I am far away ).

You might also be interested by dselect, I have read about it
several times, but never took enough time to really discover it.

My first moves when installing a new system: uncheck all ( yes,
including basic tools ) checkboxes while installing, booting on
the new system, disabling in aptitude the automatic install of
recommended stuff, and install only packages that I invoke by
myself.

Sometimes I take some fun to also purge all packages ( yes, all
of them: go to root entry of aptitude and then press '_' ) to add
them back one by one in the preview, marking all packages I do
not remove as automatically installed ( so that they'll go away
when there will be no reason to keep them ). It's nice to see
that Debian still install some tools which are not really needed
when you uncheck everything at install time.

Be careful, that way to install a computer is the best one to
install broken systems :) but I'll bet that you know that ( it's
more a disclaimer for people who could fall on that mail )



AMEN to last paragraph. But those broken systems can be educational.
I'm a newbie whose learning style is very hands on, with a laptop
devoted ONLY to experimenting with installs. Yesterday's education 
was

titled "How small can a NEWBIE make a working XFCE system?".

Your post, among others, encouraged me to try leaving "basic tools"
unchecked.
I followed with "apt-get --no-install-recommends lightdm xfce4". On
reboot I got a blinking cursor. ONLY a blinking cursor - couldn't
discover a way to do anything. Got to a terminal, purged lightdm and
xfce4 and reinstalled.


You should try to not install display managers, and then starting your 
XFCE installation with startx, on a TTY. Without recommended packages, 
you should have a working DE like this ( only used GDM and XDM IIRC, but 
using none since many months now).
Also, since I also started on XFCE I think I can say that without too 
much errors, you could have some fun installing XFCE's packages one by 
one, instead of relying on the meta-packages. The first time I did this 
was to remove orage, since I had no use for it and it was a dependency, 
where I would have loved it to be a simple recommendation.


And, purging them will not remove all packages they might have 
installed through recommendations, since those recommendations are often 
shared with other tools you have.
Purging stuff only removes configuration files in /etc after ( or not 
after, I do not know nor mind ) having removed the software, nothing 
more, nothing less.



The lesson learned? That whether or not something is a dependency is
in eyes of beholder. LOL


I apologize, but I have no idea about what does means "eye of the 
beholder".



Beneficial side effect - guided reading of the
description/recommends/suggests fields of packages.


To know what a program can do, reading them is truly very useful.
But sometimes, it is interesting to use the reverse of that feature: 
reading packages which are in need of a target, for example, Qt, Gtk, 
python... in order to purge your system of technologies you do not want 
for a reason or another. Doing this with perl is especially fun, same 
for all "mandatory" packages. You won't become a sysadmin like that, but 
you might learn what is really useful or not.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8f0a53ece783ae3b383eb190eb70d...@neutralite.org



Re: Tool for investigating dependency chains?

2013-10-09 Thread berenger . morel



Le 09.10.2013 17:22, Richard Owlett a écrit :
I'm in the process of doing some idiosyncratic minimalistic installs 
using

the "--no-install-recommends" option of apt-get.

What I would like to do is enter the package name. The tool's
response would be a list of the recommended packages and their
associated description from packages.gz. At the moment the referenced
repository would be a distribution DVD.

It's doable completely manually, but ;/

TIA


This is one of the reasons I like a lot aptitude.
Run it in GUI mode (no argument, it can be used with any user, not only 
root), search for the package you want (press '/' and enter a regex 
describing the package's name you want), and press enter.
You will have description, debtags, lot of various informations, and at 
the end, just before all versions you can install:

_ dependencies it have ( what it depends, recommends and suggests)
_ which packages depends on it ( being a conflict, a dependency, a 
recommendation or a simple suggestion)
_ which packages does it provides ( for virtual packages, like 
x-window-manager, x-display-manager, editor, x-terminal-emulator and 
much other. Those are only the probably most important ones for most 
users. )


Those are trees, with "depends on" or "depended by", then package name, 
and then complete name of package, including versions.
Sometimes you have choice, more than one package name can fulfill a 
single dependency, and/or a meta-package is here. In those situations, 
each possible package is simply shown in the complete package list.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1ae79e1bfed2f925b56774e4c1131...@neutralite.org



Re: apt-get vs. aptitude

2013-10-09 Thread berenger . morel



Le 09.10.2013 19:11, Dmitrii Kashin a écrit :

I met situations when aptitude completly broke functioning of
APT. APT utilities in their turn are simpler, and they are more
preferred to manage packages.


Just curious here, how could aptitude break apt? AFAIK, both uses dpkg 
for packages' installations, the only things they do, imho, are 
selection and downloads of packages and their dependencies, so I am 
really curious about what could be broken there.
I have read that some years ago, things were not as simple as that and 
that apt and aptitude would have broken each other because aptitude 
maintained it's own database of what is installed, but it really was 
years ago: I have always used a mix depending on my simple feeling of 
the moment to install/remove package and never had problems.
I only know that because of some ancient readings, so having some more 
info on that subject interests me, especially if it is still actual.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/31d5d794d9768d125f3a9dcd0457f...@neutralite.org



Re: Tool for investigating dependency chains?

2013-10-09 Thread berenger . morel

Le 09.10.2013 19:49, Richard Owlett a écrit :

If I can correctly read and follow instructions (friends and family
tend to doubt), the above tells me too much about what is already
installed and not enough about what is not installed :(


Hehe, I can understand your friends, or maybe I was not clear enough. 
;)


You can select installed package, or not: aptitude shows all packages 
you could install, not only the ones you have actually installed.
Plus, there are here another informations that I did not spoke about: 
it will show you the current state of the package, aka: installed, 
automatically installed, removed or absent from the system ( it means 
never installed or purged ). Those informations are shown with the 
characters before the package full name ( name + version ). "i" means 
installed, "iA" means automatically installed, "p" is for absent from 
the system, and "c" is for removed ( some configuration are still 
system-wide installed ). Oh, and, "B" means there is a broken 
dependency, but you will guess that with the red color more than with 
the letter I think.
Those characters also indicate the actions aptitude will do, but I 
think you will understand that quickly when you will have played a 
little with that tool.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/3e365d6d7146c3dd0545a99b48ae1...@neutralite.org



Re: I have an Openoffice question for small business.

2013-10-09 Thread berenger . morel

Le 09.10.2013 23:04, Joe a écrit :

On Wed, 9 Oct 2013 12:52:43 -0300
Ezequiel  wrote:


Thanks to all for your replies. I am actually pinning OO in order to
use oldstable versions. I guess I will try to compile it an make my
own repos. You were very helpfull.



Bear in mind that OOo and LO use Java for various purposes, and Java 
is

under continuous siege from the bad guys.


But Java is not a dependency, it is only recommended. Those tools works 
fine without java. I would not even have thought half a minute to 
install them otherwise, and not because of security breaches in java ( 
which will more often be used via internet browsers, not for something 
like an office suite ).



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/54c30ebb9805569f919b810e93629...@neutralite.org



Re: apt-pinning, strange behavior

2013-10-10 Thread berenger . morel

Le 09.10.2013 19:28, Dmitrii Kashin a écrit :

berenger.mo...@neutralite.org writes:

Since I had to reinstall from my last kernel error, I decided to 
stay

with stable on that computer, but I need some softwares in less
outdated versions, like development libraries or i3 ( this one is 
not

a need but a question of comfort, I admit ), so I want to use
apt-pining.

I have set all packages from stable to a priority of 900 and testing
packages with 500.
But tzdata wants to upgrade, for an unknown reason.


Certainly it wants. According to apt_preferences(5):


Yes, it wants, because I did not specified the priority for the release 
stable-updates. This is what apt-cache policy pointed, and once fixed, 
my problem disappeared, and I finally understood that obvious issue.



You should use priority of >=990 for the target release.


In the same priority range, the package which will be installed is the 
one with the highest priority, so it is fine to have one set of package 
with 500 ( or I could take 600 or any other value ) for low priority, 
and the other at 900 ( or 800 or... ), so that the version with 900 will 
be installed against the lower one, even if the lower one is more 
recent.


Explicitly making it to a priority of 900 for stable fixes that, but 
I

can not understand why it is needed?


You have just set this priority to the whole stable repository. This
should not work at all.

Maybe it will be sane to show us how you set pins?


Here are the 2 versions, first the one which really works ( in the hope 
I did not forget something, since I have enabled stable-updates, 
stable/updates, stable and testing. So I am not sure about the 
stable/updates one. Not used to stable.):



Package: *
Pin: release a=stable
Pin-Priority: 800

Package: *
Pin: release a=stable-updates
Pin-Priority: 900

Package: *
Pin: release a=testing
Pin-Priority: 500

Package: i3-wm i3
Pin: release a=testing
Pin-Priority: 900

#i3 dependencies
Package: gir1.2-glib-2.0 libc-dev-bin libc6 libc6-dev 
libgirepository-1.0-1 libglib2.0-0 libpango1.0-0 locales python-gi

Pin: release a=testing
Pin-Priority: 900

Package: clang
Pin: release a=testing
Pin-Priority: 900

#clang dependencies
Package: libclang-common-dev libgcc1 libgomp1 libitm1 libquadmath0 
libstdc++6 libobjc4

Pin: release a=testing
Pin-Priority: 900


The one which with tzdata updated to testing:

Package: *
Pin: release a=stable
Pin-Priority: 900

Package: *
Pin: release a=testing
Pin-Priority: 500

Package: i3-wm i3
Pin: release a=testing
Pin-Priority: 900

#i3 dependencies
Package: gir1.2-glib-2.0 libc-dev-bin libc6 libc6-dev 
libgirepository-1.0-1 libglib2.0-0 libpango1.0-0 locales python-gi

Pin: release a=testing
Pin-Priority: 900

Package: clang
Pin: release a=testing
Pin-Priority: 900

#clang dependencies
Package: libclang-common-dev libgcc1 libgomp1 libitm1 libquadmath0 
libstdc++6 libobjc4

Pin: release a=testing
Pin-Priority: 900


PS: I think I should probably send the package-specific priorities and 
their dependencies into specific files in preferences.d/ but I'll do 
that when I'll have a real lot of packages that I need updated. 2 
packages ( 3 in fact, I do not show opera here, to stay with the 
original situation ) can be kept in one file without making it 
unmaintainable.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4dff77b2e59cff8911bd3f6c405f5...@neutralite.org



Re: I have an Openoffice question for small business.

2013-10-10 Thread berenger . morel



Le 10.10.2013 10:24, Joe a écrit :

On Wed, 09 Oct 2013 23:21:29 +0200
berenger.mo...@neutralite.org wrote:


Le 09.10.2013 23:04, Joe a écrit :
> On Wed, 9 Oct 2013 12:52:43 -0300
> Ezequiel  wrote:
>
>> Thanks to all for your replies. I am actually pinning OO in order
>> to use oldstable versions. I guess I will try to compile it an
>> make my own repos. You were very helpfull.
>>
>
> Bear in mind that OOo and LO use Java for various purposes, and
> Java is
> under continuous siege from the bad guys.

But Java is not a dependency, it is only recommended. Those tools
works fine without java. I would not even have thought half a minute
to install them otherwise, and not because of security breaches in
java ( which will more often be used via internet browsers, not for
something like an office suite ).




As I said, I'm not a power user of most of LO, perhaps it isn't used
elsewhere, but it's a dependency of Base. Despite this, I had Base
installed without Java, but was unable to do much by way of 
connecting

to data without it.

I have a vague recollection that originally, OOo was written in Java
which years ago was popular for writing cross-platform applications. 
If
so, clearly LO is removing it progressively, and it may disappear 
from

Base at some point. I do know attitude of the US Dept. of Homeland
Security towards Java, and its days anywhere may be numbered.

--
Joe


Ah, no, I apologize, you were right. Base seems to have a hard 
dependency on Java, so I was really wrong.
I have no idea why it depends on Java, but it is written in C++, as the 
debtags shows, as the rest of Libre/Open Office.


To be honest, I also thought that it was written in Java until recently 
( well, I think I discovered that in the beginning of the year ), but 
someday I said that on a forum and was instantly replied that it was 
written in C++.


About it having be rewritten in C++ instead of Java, I do not know, I 
did not made any searches about that, but I strongly doubt it. C++ is as 
portable as Java ( just it needs to be recompiled and lacks standard 
portable GUI - and other features - that Java provides. But Java's 
standard GUI features are not the most used, as some Java dev said me. 
So... ). Plus, rewritten a huge... a very huge in facts, software in 
another language is really hard, especially if you come from a language 
where memory is managed by some obscure automatic mechanisms to come to 
language for which most of the power comes to RAII.

But my doubts can be wrong ;)

About Java's security problems... honestly, the only one thing which 
makes it true is that it is a popular and traditional language to write 
portable web applets. Like windows being the main target of hackers, in 
fact. I do not like Java, but I have learn that performance and security 
issues are not a programming language's problem, but a programmer's 
problem. Take a look in aptitude, try the games. Some are written in C 
or C++ and are as "beautiful" as 90's games, and they can not be run on 
my modern netbook. On the other side, some Java's one works perfectly. 
Debian really changed my mind on a lot of wrong ideas about computer 
sciences :)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/034dd8fbc2a22fe633b272e341588...@neutralite.org



Re: Tool for investigating dependency chains?

2013-10-10 Thread berenger . morel

Le 10.10.2013 13:36, Richard Owlett a écrit :

It does not tell me:
  A. the names of packages it recommends be also installed
  B. the description of those packages


It does. Go on that package, and press "enter". This will open a new 
tab with the same info, but with the targeted package being the main 
one. Not sure if I am clear... I should take a look to sofwares to make 
screenshots of ncurses applications :p



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8319286284dfb63e09572c1363049...@neutralite.org



Re: apt-get vs. aptitude

2013-10-10 Thread berenger . morel



Le 10.10.2013 14:30, Erwan David a écrit :

On Thu, Oct 10, 2013 at 12:47:08PM CEST, "David L. Craig"
 said:

On 13Oct09:2153+0100, Joe wrote:

> On Wed, 09 Oct 2013 13:24:57 -0500
> Richard Owlett  wrote:
>
> >
> > Being retired, I've no aspirations of being a sysadmin.
> >
> >
> >
> If you run Linux, you already are. You don't get to choose.

Probably.  There have been reports of parents set up on
Linux platforms by children who remotely attend to the
care and feeding of the box.  Hopefully these parents are
not deemed to be lusers by their progeny.


Just say "if you have your own compuyter (whatever the OS) that
nobody manages for you, you already are a sysadmin.


Guys, I must disagree with that. It would mean that any linux distro is 
hard to maintain, and that's wrong. Plus, sysadmin have a lot more 
knowledge than simple users and power users.


Imagine that it would mean that any people making some shitty excel 
formulas would be a programmer, anyone able to change a wheel on his car 
a mechanic, anyone able to grow something a biologist, etc... sounds 
fun, but not realistic at all :)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/6064429a723cd92e4d8a3ac26b13e...@neutralite.org



Re: Tool for investigating dependency chains?

2013-10-10 Thread berenger . morel



Le 10.10.2013 15:02, Richard Owlett a écrit :

berenger.mo...@neutralite.org wrote:

Le 10.10.2013 13:36, Richard Owlett a écrit :

It does not tell me:
  A. the names of packages it recommends be also installed
  B. the description of those packages


It does. Go on that package, and press "enter". This will open a
new tab with the same info, but with the targeted package being
the main one. Not sure if I am clear... I should take a look to
sofwares to make screenshots of ncurses applications :p



Ah-ha it does do what I want - just not the way I expected.
Have got to do detailed read of the documentation.
Thanks


You're welcome


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/60c2fb220b38750dac246ef1a9a06...@neutralite.org



Re: reinstalling Nginx & others usin apt from command line

2013-10-10 Thread berenger . morel



Le 10.10.2013 19:54, Gregory Nowak a écrit :

On Thu, Oct 10, 2013 at 06:21:55PM +0100, Brad Rogers wrote:

Hello Brad,

Not entirely true;  Any package that has configuration 
files/directories
in user space will have those left even after an apt-get(1) purge.  
So,

to be sure, one has to delete those as well.



$ man 1 apt-get
No manual entry for apt-get in section 1
See 'man 7 undocumented' for help when manual pages are not 
available.


The apt-get(8) man page says:

"purge
   purge is identical to remove except that packages are
   removed and
   purged (any configuration files are deleted too)."

Greg


Then the man is not precise enough. Try to install a random package 
with userspace settings, like, say, a game. Change it's settings, purge 
it from the system, and reinstall it. The settings will not change.


Purge only removes configuration files contained in the package, so, 
/etc ones.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/131a6e43114f5511a1ac7f0c1921c...@neutralite.org



Re: reinstalling Nginx & others usin apt from command line

2013-10-10 Thread berenger . morel



Le 10.10.2013 20:20, Brad Rogers a écrit :

On Thu, 10 Oct 2013 10:54:57 -0700
Gregory Nowak  wrote:

Hello Gregory,


"purge
  purge is identical to remove except that packages are
  removed and
  purged (any configuration files are deleted too)."


As has been said, the man page is imprecise.


I wonder if that could justify a bug report? Speaking about that, I 
have no idea about what exactly does the reinstall option: simply remove 
and then install, or purge and then install? If, as I think, it's the 
first option, what is the use for that, since packages should not lose 
files magically?



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/bc2c560e738971144df174cf36bb7...@neutralite.org



Re: which file should I download

2013-10-10 Thread berenger . morel



Le 10.10.2013 22:06, Anjan Mitra a écrit :

debian-7.1.0-amd64-DVD-1.iso [1] 2013-06-16 01:39 3.7G

debian-7.1.0-amd64-DVD-2.iso [2] 2013-06-16 01:39 4.4G

debian-7.1.0-amd64-DVD-3.iso [3] 2013-06-16 01:39 4.4G

debian-update-7.1.0-amd64-DVD-1.iso [4] 2013-06-16 05:34 2.9G


Links:
--
[1]

http://cdimage.debian.org/debian-cd/7.1.0/amd64/iso-dvd/debian-7.1.0-amd64-DVD-1.iso
[2]

http://cdimage.debian.org/debian-cd/7.1.0/amd64/iso-dvd/debian-7.1.0-amd64-DVD-2.iso
[3]

http://cdimage.debian.org/debian-cd/7.1.0/amd64/iso-dvd/debian-7.1.0-amd64-DVD-3.iso
[4]

http://cdimage.debian.org/debian-cd/7.1.0/amd64/iso-dvd/debian-update-7.1.0-amd64-DVD-1.iso


Not a lot of description of your problem, but I'll assume you want to 
download an installation image.

If so, then in all those files it would be debian-7.1.0-amd64-DVD-1.iso


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/d091bd6c546b6c81bbaabb3dce06e...@neutralite.org



Re: reinstalling Nginx & others usin apt from command line

2013-10-10 Thread berenger . morel



Le 10.10.2013 21:59, Jerry Stuckle a écrit :

On 10/10/2013 3:08 PM, berenger.mo...@neutralite.org wrote:



Le 10.10.2013 20:20, Brad Rogers a écrit :

On Thu, 10 Oct 2013 10:54:57 -0700
Gregory Nowak  wrote:

Hello Gregory,


"purge
  purge is identical to remove except that packages are
  removed and
  purged (any configuration files are deleted too)."


As has been said, the man page is imprecise.


I wonder if that could justify a bug report? Speaking about that, I 
have
no idea about what exactly does the reinstall option: simply remove 
and
then install, or purge and then install? If, as I think, it's the 
first
option, what is the use for that, since packages should not lose 
files

magically?




I'm not sure what the confusion is here.  Sure, it deletes all
configuration files it knows about.  But then I would NOT expect it 
to

delete any user-generated files, configuration or otherwise.  I would
only expect it to delete what it can restore.

Jerry


Maybe you, since you know how it works. But, someone with no knowledge 
would think that it also removes userland configuration files, since it 
claims to remove all configuration files.


To be honest, sometimes I would really like it to be able to do so :) 
and I could imagine ways to do that, too.
For example, imagine if all softwares would respect XDG Base Directory 
Specification ( oh, wait, they won't, lennart participate ;) but 
honestly, I really would like it to be followed ) to install 
configuration files in $XDG_CONFIG_HOME. dpkg would simply have to check 
for all users with a home directory what is their XDG_CONFIG_HOME and 
search here the repertory to remove, in case of a purge. That does not 
sounds so impossible, right?
Of course, there is the problem of determining where the hell that 
variable is defined for each users, and it would also imply that all 
softwares respect it, which will probably never happen.
But, in my opinion, having a tool able to remove configuration files 
from user directories would help. A lot. Because it happens that I try 
some softwares, and then purge them, but I still have to find every 
single file they could have installed by hand. And this is *really* 
boring.
I know that it would have it's own problems, of course, for example 
when you install multiple versions of the same program, or if some users 
want to keep their files anyway (this issue could be fixed by not 
effectively removing those files, but by installing a script which would 
ask to users if they want to keep or remove files, script which would be 
automatically run at their next session ).


I know that it would not be so easy ( I can think about other problems 
that I did not mention, but I do not want to write a novel ) , but it 
would be doable, by someone with enough knowledge of package management 
and global system, and, again, if most softwares were willing to respect 
XDG specs. Vim, bash, xpaint, firefox, those are few examples of widely 
known softwares, which does not respect that spec. And for which reason? 
I guess there are none. And there are tons of other softwares which does 
not.


My homedir is far more messy on Debian than on windows finally, and I 
clean it frequently, but I use it far more often on Debian than on 
Windows ( where the usages are more to use desktop, other partitions or 
"my documents"). Don't you think that something is wrong, here?


The problem is not technical, because it's easy to do, even in C:
==
#define HOME_CONFIG "XDG_CONFIG_DIR" // can be used to keep windows 
compatibility without changing any line of code

#define PROGRAM_OWN_CONFIG_DIR "/some/strange/config/dir/name/"
char* user_config_dir = getenv(HOME_CONFIG);
if( !user_config_dir )
  exit( -1 );
unsigned short mcd_len = strlen( user_config_dir ) + strlen( 
LEN_OF_PROGRAM_OWN_CONFIG_DIR ) +1;

char* my_config_dir = malloc( mcd_len );
if( !my_config_dir )
  exit( -2 );
if( 0 == strlen( user_config_dir ) )
  strncpy( my_config_dir, "~", 1 );
else
  strncpy( my_config_dir, user_config_dir, strlen( user_config_dir ) );
strncat( my_config_dir + strlen( my_config_dir ), 
PROGRAM_OWN_CONFIG_DIR, LEN_OF_PROGRAM_OWN_CONFIG_DIR );

==
That's portable, standard, written in less than 5min and relatively 
secure ( there are still flaws there, but I am more used to C++, and 
it's a simple demo ). So, why not doing it? Less than 20 lines in C, 
less than 5 lines in C++, and probably not more in other languages.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/e823575541db7233c2f9fe85c5b8d...@neutralite.org



Re: apt-pinning, strange behavior

2013-10-10 Thread berenger . morel



Le 10.10.2013 23:06, Dmitrii Kashin a écrit :

berenger.mo...@neutralite.org writes:

In the same priority range, the package which will be installed is 
the

one with the highest priority, so it is fine to have one set of
package with 500 ( or I could take 600 or any other value ) for low
priority, and the other at 900 ( or 800 or... ), so that the version
with 900 will be installed against the lower one, even if the lower
one is more recent.


Oh... Truely? I thought differently and was sure I am right.


Maybe you are right, but in that case, how would you explain the 
behavior I had if a package of a priority of 500 is considered to have 
the same priority as a package with 900 ? ;)


I just skimmed again through apt_preferences man page, but did not 
find

such examples or explanations. Where's it documented?


I must admit, that I only base my words on old readings and 
experimentations. It also seems logic: what would be the interest to 
have so wide ranges of numbers oterwise?
Maybe I'm wrong, but what I have seen those days tends to prove that I 
am not.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/85a4a2725fc50736dcf02195d33b1...@neutralite.org



Re: reinstalling Nginx & others usin apt from command line

2013-10-10 Thread berenger . morel



Le 11.10.2013 00:49, Gregory Nowak a écrit :

On Thu, Oct 10, 2013 at 11:23:50PM +0200,
berenger.mo...@neutralite.org wrote:

Maybe you, since you know how it works. But, someone with no
knowledge would think that it also removes userland configuration
files, since it claims to remove all configuration files.


Ok, since the man page is saying one thing, and yourself and Brad are
saying another, I decided to try this out. Here's what I did:
[snip]


Yes, I know that this is the normal behavior. It purged system-wide 
configuration. But, the point is, that for some people, when they read 
the man page, they think that it removes *all* configuration files, 
including the ones the application have created in userspace. Which it 
does not, we all agree about that.


Now, the question I asked was, should the manpage be fixed to avoid any 
user to think that dpkg can remove all configuration files, including 
those which were not created by dpkg, since some applications create 
some config files themselves ( so, the user does not create them not 
manipulate them without that application's dedicated interface ) which 
can makes things confusing.
But, yes, definitely, people with knowledge about dpkg, even as small 
as mine, understand that it does only remove files it have created.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/d3b78415d892ee8be81511c46...@neutralite.org



Re: reinstalling Nginx & others usin apt from command line

2013-10-10 Thread berenger . morel



Le 11.10.2013 01:20, Gregory Nowak a écrit :

On Fri, Oct 11, 2013 at 12:59:32AM +0200,
berenger.mo...@neutralite.org wrote:

Yes, I know that this is the normal behavior. It purged system-wide
configuration. But, the point is, that for some people, when they
read the man page, they think that it removes *all* configuration
files, including the ones the application have created in userspace.
Which it does not, we all agree about that.


Ok. What seems like common sense to some of us might not be viewed
that way by others.


When things becomes usual for someone, he sees them as common sense. On 
other subjects, some people asked me how I can not see the common sense 
of their actions, too, when that needed a global understanding of 
something I never used before. Common sense is subjective I think.


I remember a show from French humorist where they were playing the role 
of people doing marketing. The conclusion of it was (rough translation):

Do not take people as if they were stupid, but do not forget they are.
It was to make people laugh, and it is quite harsh, but, it is not 
completely wrong, we are all unable to understand things other will. 
(it's hard for me to explain what I mean in English, so do not take this 
too seriously and be indulgent please)


Finally, that's close to what I try to apply when I create a software. 
I try to consider the more dumb actions of the user possible, and do 
what I can to avoid them to make everything explode. It includes 
checking if the user entered characters on a field where you ask the age 
for example :)



Now, the question I asked was, should the manpage be fixed to avoid
any user to think that dpkg can remove all configuration files,
including those which were not created by dpkg, since some
applications create some config files themselves ( so, the user does
not create them not manipulate them without that application's
dedicated interface ) which can makes things confusing.


Yes, the man page should probably be fixed since it could be
misinterpreted. How about changing "purge is identical to remove
except that packages are removed and
   purged (any configuration files are deleted too)." to say
   "purge is identical to remove except that packages are
   removed and
   purged (any configuration files created by the package at
   install time are deleted too).



I think this would enhance, yes. Could be even better with "( any and 
only configuration files created by the package at install time are 
deleted)" so that I think it is not possible to be mistaken. But my 
English is not really good, so it may probably be said better.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8d2e481bad12cbcf7c106f2e3185a...@neutralite.org



Re: reinstalling Nginx & others usin apt from command line

2013-10-11 Thread berenger . morel

Le 11.10.2013 06:10, John W. Foster a écrit :

I stripped every reference to evolution from the entire file system,


Including hidden folders?

Try this:

$cd ~
$find -iname '*evolution*'

At that point, if you found something, then you forgot hidden folders, 
so do


$rm -r `find -iname '*evolution*'`

to remove them. Just be careful when checking files after the 1st find, 
some applications could use a close name.
And, of course, this only works if and only if you application used 
it's own name as part of folder's names where it stores stuff. Usually, 
they does. Usually, but I can think of iceweasel which uses .mozilla for 
example, and I had many examples of the same problem when I was using 
windows and removing things by hands. Checking the provider's name of 
the software is never a bad idea.



purged the application as you suggested in your references, and
reinstalled it.


This will only remove files in /etc. We can like it or not ( I do not, 
but the problem is not simple ) but it is like that currently.




--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/e3c0dd6143b5145930630e102aeed...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-11 Thread berenger . morel

Le 11.10.2013 13:03, Joel Rees a écrit :

On the converse, I think it is a crime to promise (as makers of
certain popular OSses do) that you can properly use a computer or
other computer-based communication device without administering or
managing the computer system.


It is not. I can use rapid transit without taking care or understanding 
anything behind it. I do not even have to know that it have wheels.
Same for computers. In some places, you can have access to public 
computers and use them for a set of operations, and people does not have 
to know how to do anything else on those.
Another, and last one, example. Take an electric saw (I do not know the 
English word for that). If it is your own, you will want to avoid buying 
a new one when the blade is too old, but it is a choice. You could just 
want to use it, buy a new one when it is too old. If so, you only know 
how to start it and how to apply it on what you want to cut. No need to 
know internals. And even if you know how to change the blade, it will 
not mean you are an electro-mechanic guy. Eventually, it can makes you 
an advanced user, but nothing more (because it is better to know which 
kind of blade to use to cut metal, or wood, or stone).
People who manage their computers themselves are not sysadmin, they are 
advanced users. Some of them are sysadmin, of course, but not because 
they maintain their own system.


In fact, the problem here is: "that you can properly *use* a computer". 
*Use*. Which use exactly have you in mind?



We're a long way from being able to build internet terminals that
people can use as simply as they use a phone, and it's quite possible
that it can't really be done.


Terminals to Internet? Probably it is not possible, since internet is 
not only the http section of the www. But building terminals for people 
to access websites is not so hard: take a keyboard, a pointing hardware 
( it can even be emulated by the keyboard ), a screen, a CPU, some RAM, 
a modem.
Build an OS around that, add to that OS a default configuration which 
works by default plus a website browser, and remove the ability for user 
to do anything than using the browser. No window manager, no software 
center, etc.

You have it.

Of course, for me this is a nightmare, not a dream. But it is doable. 
You simply have to sold the item pre-configured, and you know what? ISP 
are able to do that. They only do not do it, because people hopefully 
know that Internet is a tool to enhance their computer's features, like 
managing files in a lot of different ways: modifying images, writing 
texts, etc.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/45b0ccc8f75794125500b1c294dd6...@neutralite.org



Re: Installing Cinnamon 2.0

2013-10-11 Thread berenger . morel



Le 11.10.2013 19:42, Tom H a écrit :

AIUI, MATE is a redo of GNOME 2 using GTK2 and Cinnamon a redo of
GNOME 2 using GTK3. Will they merge at some point? Or have their
interpretation and evolution of the GNOME 2 UI diverged enough for
them to keep on co-existing?


AFAIK, they are fork of different softwares: gnome 2 for one, gnome 3 
for the other. Plus, I do not think there is any sign of will to merge.

So, I doubt they'll do, because of will and or technical problems.

But, to be honest, I do not know, I do not use any DE anymore, and if 
someday I would have to advice the use of DE to someone, that would be 
XFCE or LXDE. I have used them, and they offer enough features in my 
opinion. Plus, they are lightweight, their age can be used as a proof 
that they'll continue on their original line. Exactly what gnome seems 
to not have made ;) so when I consider it's forks... 2 forks, for 1 
reason, I allow me to smile and have hard doubts about their future.
And well, I do not mind at all, I'm no longer a DE user and probably 
will never become one anew. Except if someone make a DE built around 
TWM, very lightweight softwares, and no feature duplication ( it means: 
no other tool than the window manager would manage windows. We are far 
from that, even in terminals. )


My 2 cents.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/00c8dd9faac4e83279eca142d1b21...@neutralite.org



Re: Installing Cinnamon 2.0

2013-10-11 Thread berenger . morel



Le 11.10.2013 19:43, Alex Moonshine a écrit :

> > On Fri, 11 Oct 2013 16:50:01 +0200
> > Ralf Mardorf  wrote:
I'm speaking about upstream and here is one conflict:

[rocketmouse@archlinux ~]$ sudo pacman -S mate-file-archiver
resolving dependencies...
looking for inter-conflicts...
:: mate-file-archiver and file-roller are in conflict. Remove
file-roller? [y/N]



Sorry, but why would one need mate-file-archiver (Engrampa) and
file-roller on the same machine? The former is a fork of the latter
(and to all practical means just the same package under different
name). Same goes for nautilus/caja, gedit/pluma, etc.
I think that either you use MATE DE and Engrampa replaces file-roller
for you, or, if you want to install MATE on a system that already has
some other DE with file-roller installed, that you want to keep, you
install mate-base package, which (I believe) includes none of extra
applications (and you can go on using file-roller under MATE).


Or maybe they should all depend on a meta package named, say, 
x-archive-manager? That would indirectly fix the problem, and give users 
a choice :)


Otherwise, about reasons... some people uses more than 1 DE. I guess we 
could use the same reason, here.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/9d7ddc1cd775dfd16129a9647e1d9...@neutralite.org



Re: Installing Cinnamon 2.0

2013-10-11 Thread berenger . morel



Le 11.10.2013 20:44, Tom H a écrit :

On Fri, Oct 11, 2013 at 12:42 PM, Ralf Mardorf
 wrote:



Have you considered NAS?


No way!


What's wrong with NAS to elicit such a rejection?



GVFS is absolutely optional software.


You only ever consider things from your limited use-case but upstream
developers and distribution maintainers have to take into account the
majority of users.


The majority of users do not have NAS :)
Especially to use them as portable hardware, I think.


You've also complained in this thread and previously about compliance
with green drives. I could've replied: "I only ever connect an
external drive to my computer long enough to copy in and/or copy out
files.


Most of the people I know which uses external drives, and I speak here 
about drives, no small data containers, is to use, for example, music, 
films, and such stored on that harddrive.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/246d92331d9352f4f8265dc32a59c...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-11 Thread berenger . morel

Le 11.10.2013 23:06, Brian a écrit :

"are you root?"


It does only means you own the system. Not that you can claim to be a 
sysadmin. I own my car. I am not a mechanic, but I anyway have the 
*authorizations* to tinker it. It's what root, or to be more precise, 
uid=0 means in linux OSes.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/0e7350cb30a02dbf6748b0cc73f5a...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-11 Thread berenger . morel



Le 11.10.2013 23:44, Brian a écrit :
On Fri 11 Oct 2013 at 23:21:07 +0200, berenger.mo...@neutralite.org 
wrote:



Le 11.10.2013 23:06, Brian a écrit :
>"are you root?"

It does only means you own the system. Not that you can claim to be
a sysadmin. I own my car. I am not a mechanic, but I anyway have the
*authorizations* to tinker it. It's what root, or to be more
precise, uid=0 means in linux OSes.


If you own a system you control it and can do whatever you like with 
it.
You can give yourself whatever label you want (sysadmin, superuser, 
top

dog etc) - it matters not. How about "Debian Despot"?


Oh, of course, if you speak about giving yourself a label, then, fine. 
Take the one you want. But, it does not mean that you can claim to be a 
professional, or that you can say someone is a professional.


Take the label you want. But if you take the label of "programmer" 
because you can only write a hello world, and will own the source code. 
But then do not be surprised if other people gives you the label of 
"liar".
It is the same with sysadmin. You can own your computer, be only able 
to install softwares and use those excuses to label yourself a sysadmin. 
But then, other people are also free to give you the label of liar.


I guess you will not like my words, but if you think that people are 
free to give themselves the label they want, so you must accept that 
other are also free to give the labels they want.
Long time ago, I studied the "dark side of computer sciences", and the 
first things I have learn are that you can not claim to be a hacker, or 
elite, or... If you do so, then people will name you lamer. You are a 
hacker if other people recognize you as such.
The truth here is simple: you are not what you want, only other people 
can define who you really are.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/706e569fe1ce9f5b26d2f80f8476b...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-11 Thread berenger . morel



Le 12.10.2013 00:01, Paul Cartwright a écrit :

On 10/11/2013 05:21 PM, berenger.mo...@neutralite.org [1] wrote:


It does only means you own the system. Not that you can claim to be
a sysadmin. I own my car. I am not a mechanic, but I anyway have

the

*AUTHORIZATIONS* to tinker it. It's what root, or to be more
precise, uid=0 means in linux OSes.

 the difference is, you don't have a "mechanic" for Debian, you ARE
the mechanic. YOU make it better ( or worse), so yes you ARE the
sysadmin. your other users cannot maintain the system, they cannot
update the system, they cannot destroy the system. You can update it,
maintain it, destroy it, all by yourself.


I think we simply disagree on what a sysadmin is.
So, I'll now use a definition that I does not have created, from a 
source which is often considered as quite good, wikipedia:


===
A system administrator, or sysadmin, is a person who is responsible for 
the upkeep, configuration, and reliable operation of computer systems; 
especially multi-user computers, such as servers.

===

The keyword here is "multi-user", with the constraints it apply. If you 
manage your own computer, then, ok, you are the admin of your computer. 
You are not a sysadmin.
That's my opinion, and I accept to discuss about it, but unfortunately, 
I must admit that your argument is not strong enough to convince me, and 
that I can not find an argument that I think you could hear. Except the 
authority argument I just used, which I consider as many other people as 
a weak one ( at least I was able to quote, and not simply say "I know 
someone which...". Better than usual in that kind of arguments ).



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/c12e6f3aa2e6a3e6a814dbf81e3c8...@neutralite.org



Re: Installing Cinnamon 2.0

2013-10-11 Thread berenger . morel



Le 12.10.2013 00:46, Tom H a écrit :
On Fri, Oct 11, 2013 at 3:14 PM,   
wrote:

Le 11.10.2013 20:44, Tom H a écrit :

On Fri, Oct 11, 2013 at 12:42 PM, Ralf Mardorf
 wrote:

You've also complained in this thread and previously about 
compliance

with green drives. I could've replied: "I only ever connect an
external drive to my computer long enough to copy in and/or copy 
out

files.


Most of the people I know which uses external drives, and I speak 
here about
drives, no small data containers, is to use, for example, music, 
films, and

such stored on that harddrive.


Did you notice the "could've"?


Ermm... no, sorry :) did not noticed :/


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/246a9662066fa5c8e25f9b2224b12...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-11 Thread berenger . morel



Le 12.10.2013 02:24, Jerry Stuckle a écrit :

On 10/11/2013 6:22 PM, berenger.mo...@neutralite.org wrote:



Le 12.10.2013 00:01, Paul Cartwright a écrit :

On 10/11/2013 05:21 PM, berenger.mo...@neutralite.org [1] wrote:

It does only means you own the system. Not that you can claim to 
be

a sysadmin. I own my car. I am not a mechanic, but I anyway have

the

*AUTHORIZATIONS* to tinker it. It's what root, or to be more
precise, uid=0 means in linux OSes.

 the difference is, you don't have a "mechanic" for Debian, you ARE
the mechanic. YOU make it better ( or worse), so yes you ARE the
sysadmin. your other users cannot maintain the system, they cannot
update the system, they cannot destroy the system. You can update 
it,

maintain it, destroy it, all by yourself.


I think we simply disagree on what a sysadmin is.
So, I'll now use a definition that I does not have created, from a
source which is often considered as quite good, wikipedia:

===
A system administrator, or sysadmin, is a person who is responsible 
for
the upkeep, configuration, and reliable operation of computer 
systems;

especially multi-user computers, such as servers.
===

The keyword here is "multi-user", with the constraints it apply. If 
you
manage your own computer, then, ok, you are the admin of your 
computer.

You are not a sysadmin.
That's my opinion, and I accept to discuss about it, but 
unfortunately,
I must admit that your argument is not strong enough to convince me, 
and
that I can not find an argument that I think you could hear. Except 
the
authority argument I just used, which I consider as many other 
people as
a weak one ( at least I was able to quote, and not simply say "I 
know

someone which...". Better than usual in that kind of arguments ).




Wikipedia is hardly what I would call "reliable".


Did not I said "often" ?


If all you have for an "authority" is Wikipedia, then your "argument"
is not strong enough to convince me.


And I admitted it myself, that it was a weak one. I think we simply 
disagree on the meaning of the word, and maybe because of the fact that 
I am not a native or even good English speaker, I reached my limits on 
this discussion. That was the most important thing I said in my previous 
mail.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/fc1e19313b89cf06cc4def47aea41...@neutralite.org



gnome depends on adblock-plus

2013-10-11 Thread berenger . morel
Well, as I claimed more than once here, I am not a DE user, but I have 
read today something ( on linuxfr ) which surprise me a lot, and it seem 
that reading is true, from : gnome depends on xul-ext-adblock-plus.


I always thought that dependencies were here for technical reasons, and 
I can not imagine how a complete DE could depend on a web browser 
extension, and it becomes worse when the extension in question have 
side-effects on important subjects like neutrality.


So, I would like to know if someone knows why this dependency exists, 
instead of a recommendation or suggestion.
Without the problem of ideology, on which I do not want to go, so I'll 
stay on technical point.


Having gnome depending on an iceweasel extension, which in turn depends 
on iceweasel does simply seems stupid, for more than one reason:
_ gnome, or even iceweasel should only suggest, or recommend, 
extensions. The web browser does not need it to work fine.
_ gnome should depends on iceweasel, which in turn should be linked to 
some extensions.


There is a bug report, here ( 
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=689858 ) which is one 
year old.
I am not familiar with the debian bug tracking system, and am only 
reporting here a link provided to me through a forum, but I have seen no 
reply of gnome's maintainer, so I am asking here to know if there is a 
technical reason for that dependency, of something which makes gnome 
still depending hardly on that firefox's only extension, still being 
here after a year (the messages in the bug report are here since 2012).
Someone mentioned that someone should report that to the technical 
comity, but... I do not really feel like I am well placed to do this, so 
I am asking to userlist if someone knows the reason behind the 
non-fixing of the issue.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/91ce1967d913a10374a1943a0c67d...@neutralite.org



Re: gnome depends on adblock-plus

2013-10-12 Thread berenger . morel



Le 12.10.2013 10:58, Marko Randjelovic a écrit :

"One Ring to rule them all, One Ring to find them, One Ring to bring
them all and in the darkness bind them In the Land of Mordor where 
the

Shadows lie." :)

apt-get install gnome-applets gnome-screenshot gnome-session
gnome-panel gnome-terminal gedit gdm3 file-roller gnome-icon-theme


As I said, I am no longer a DE user, but before leaving the wonderful 
world of automatically installed DE through one package, I was using 
XFCE, and at a moment I noticed that I never used some applications, so 
I found a way to remove them: do not installing meta-package(1) was the 
solution.


But it does not change that a meta-package have a strange dependency.

More important, it is the gnome package. The one which is automatically 
installed by a default installation of Debian. Here, anew, I am not 
directly concerned, since I build my Debian from only the strict minimum 
number of packages(2)...



Notes:
1: In fact, I think that meta-packages should only have recommendations 
and suggestions, since they are automatically installed by a default 
Debian configuration. This would not change the default behavior, while 
it would provide tinkerers more freedom and flexibility. After all, by 
not being real softwares or features, they should not be able to have 
hard dependencies, because there is no *technical reality* behind that. 
Only political choices, and I do not think that a package management 
should take care of the politic side of things. This should be the 
admin's problem ( I did not said sysadmin ;) ).


2: plus some packages which are not vital at all, like cron, which I 
usually remove. I do not think it should be installed when you choose to 
have the strict minimum system. But this is a minor issue which is not 
the one I want to discuss here.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/3adacd1e405c7f44ea2f8e9d2d7b4...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-12 Thread berenger . morel



Le 12.10.2013 13:50, Ralf Mardorf a écrit :

On Sat, 2013-10-12 at 06:19 -0400, Paul Cartwright wrote:

don't you think most home use "single-user" systems have 2 people ??


Germany has become a society of singles, solo parents and of people 
who

live in relationships, but with separated households.

Even in
Europe it's likely different for meridional nations.


True. We are neighbors (in term of country, I mean) and I have some 
difficulties to imagine a family where people have each their own house. 
Thanks for info, it is a quite interesting thing to know. In France, 
when two persons form a couple, they usually start at a moment (duration 
varies of course) or another to live in the same house and start sharing 
their material possessions as well (this is only to confirm that it 
differs greatly in Europe, it seems to be the opposite as Germany and we 
share frontiers and history).


About the number of phones being greater than the number of people, I 
could have said that it could be made by the fact than people needs 
professional and private phones, but it is not true for all jobs. Could 
you explain me the use of more than 1 phone per person? (just pure 
curiosity, I try to keep my mind open, as much as possible. I feel like 
it becomes harder with the time.)



ok, then it becomes a multi-user-network, however, I suspect
that here most "single-user" systems do _not_ have 2 people.


It is a particular kind multi-user-network, at least how I understood 
what you described, where everyone have the same role. LAN parties (1) 
are usually made without a real DHCP server, with "hard-coded" IP 
addressees.
There could also be a configuration of a couple both using their own 
computer in the same room, but it seem to conflicts with previous 
statements.



1: of course, a DHCP can helps, and I have seen those made by an 
association starting to buy and configure dedicated hardware. But even 
then, the last time I went there they had not setup a DHCP. Windows is 
able to build a network without IP conflicts very well, in default 
configurations. I strongly doubt that Debian would be, but, of course, 
there is no DHCP daemon in default installations of end-user Debian 
systems. Plus, even if everyone had a DHCP, some magic would have to be 
made to make them able to discuss about which computer should take which 
address. Magic which is already present in Windows XP (not sure about 
more recent ones)


PS: I have split your comment to extract a part of it, I hope I not 
have moved things out of context.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/65274fb96f2258502aa74a48a8338...@neutralite.org



RE: Installing Cinnamon 2.0

2013-10-12 Thread berenger . morel

Le 12.10.2013 14:04, Mark Allums a écrit :

I'm not nagging about you. My concern is to nag about recommending a
third party repository, that is known to cause issues. If the OP 
does

like Cinnamon, then IMO the OP should use the official packages to
ensure that it's as stable as possible and not to switch to Mate 
from a
third party repository. If the OP needs a bug fix or a new feature 
of

Cinnamon 2.0, it might be worse the effort to build it from source.


The point of mentioning MATE was in context with adding MATE to
Debian.  Our point was, there are only minor problems with that
happening, and they should be solvable fairly easily within a short
period of time



(i.e., in the time between releases of Stable).


This problem is not MATE specific. All DEs in Debian have it: in Debian 
stable, the one that I am now running on the computer with which I now 
use to work, XFCE is only at version 4.8, but IIRC XFCE4.10 was released 
before Debian started the freeze.
MATE could go in Debian, if one of it's users accept to become it's 
maintainer. If it was so easy, then I can only guess that MATE's users 
finally do not care that much to have it into official repo, and should 
accept that other people uses that as an argument that it *might* give 
problems. Because, yes, 3rd party repo are possible, but are not 
supported by Debian.


And, things which *might* give problems which are not supported by the 
OS's maintainers made things harder for people not willing to deal with 
problems. When you start to have more than one team responsible for the 
state of a system, and when a problem appear, you usually have annoying 
games of ping-pong.


In short: if you are so sure that MATE is fine, and want to advertise 
it a lot, stop whining about Debian which refuses it, and start the way 
to integrate it into Debian. I think the dev team will be happy to have 
more manpower and provide more DE choice.



MATE
is usable NOW, and it's mainly an irrational attachment to Gnome that
prevents it, not any real problems with MATE.

People are abandoning Gnome 3 in droves.  Most are turning to XFCE, I
think largely because it is already in Debian.  If MATE were in
Debian, a large contingent of those people would use MATE if it were
available, I think, because most of them were using GNOME 2 before.


And what about those people being ancient gnome users because gnome is 
the default, and not because gnome is better (1) ? Maybe the problems 
they had with gnome encouraged them to try other DEs, like KDE, XFCE and 
LXDE present in Debian, and from that base, choosing the one which fit 
best their needs.
Perhaps they would not return to gnome2 know, because they finally 
think that it is inferior for their uses to XFCE?


Notice that, as you, I made hypothetic assertions based on the wind to 
make other's choices and opinions fitting my needs without asking them. 
Aka, that XFCE is better than gnome, which I do not truly think. They're 
simply different.


1: I was never able to understand what people can find to gnome, even 
the gnome2 DE. Matter of taste, of course. KDE was better suited for me, 
but shared a problem: too monolithic. So only XFCE and LXDE in Debian 
were able to satisfy me at start. Now, even them are not correct enough 
for me (but I still use some LXDE apps).



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/600adc2cc45525b11320812969d2d...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-12 Thread berenger . morel



Le 12.10.2013 15:16, Jerry Stuckle a écrit :

On 10/12/2013 7:50 AM, Ralf Mardorf wrote:

On Sat, 2013-10-12 at 06:19 -0400, Paul Cartwright wrote:
don't you think most home use "single-user" systems have 2 people 
??


Germany has become a society of singles, solo parents and of people 
who
live in relationships, but with separated households. Even food 
industry
makes huge profit by selling convenience foods for singles. Here you 
can
buy 3 packed salami slices. Perhaps we have more single persons who 
own
several computers, than several people sharing one computer. It 
seems to

differ from nation to nation. Some are "single nations" others are
"family nations". I don't know statistics, but IIRC in Germany we've 
got

more smart phones, than citizens. I've got the impression that most
people in Germany have a smart phone, tablet PC, netbook or laptop 
and

they don't share it with somebody else. Sharing a laptop sometimes
happens, sharing smart phones, tablet PCs and netbooks not that 
often.
Desktop PCs are often used for a special purpose, e.g. for gaming 
and I
know people who play games with other people, but even if they sit 
in
the same room, they don't use one computer, but connect their 
computers
instead ... ok, then it becomes a multi-user-network, however, I 
suspect

that here most "single-user" systems do _not_ have 2 people. Even in
Europe it's likely different for meridional nations.

Regards,
Ralf




Ralf,

I've seen you give similar arguments before.  Let me give you a hint
here.  Most of the people in the world don't care about what Germans
do in their own homes.  Germany isn't the world; in fact, it isn't
even in the top 10 population wise.  Your comments about what they do
is completely immaterial.

That's not to say I don't appreciate *MOST* Germans; I've visited
there three times and enjoyed every visit immensely.

Jerry


I think that the point in his message was to show that it is not 
because the usage in your, or mine, or his, country is something that 
this use is the usage of the majority.

Of course, I may be wrong.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8904771e0add3d822171583685771...@neutralite.org



RE: Installing Cinnamon 2.0

2013-10-12 Thread berenger . morel



Le 12.10.2013 15:42, Mark Allums a écrit :

I might could change the typos in docs to become different typos in
different places.  I can't develop for MATE, it's not part of my 
skill

set. I will encourage other people to get involved, that much, I can
do.


Package maintenance is not really a programming task AFAIK. It, of 
course, need you to be able to compile stuff, but not to write source 
code :)


> People are abandoning Gnome 3 in droves.  Most are turning to 
XFCE, I

> think largely because it is already in Debian.  If MATE were in
> Debian, a large contingent of those people would use MATE if it 
were
> available, I think, because most of them were using GNOME 2 
before.


And what about those people being ancient gnome users because gnome 
is
the default, and not because gnome is better (1) ? Maybe the 
problems
they had with gnome encouraged them to try other DEs, like KDE, XFCE 
and
LXDE present in Debian, and from that base, choosing the one which 
fit

best their needs.
Perhaps they would not return to gnome2 know, because they finally


That's possible. The point is:

1. Gnome 3 isn't desirable to many,


Yes, someone involved in computer news can not ignore that, even if he 
does not use a DE, same for windows 8 :)
In fact, I like this, because it always make me laugh. By trying to run 
on the last fashion (not sure about the wording) things are just 
collapsing. I am really happy to have had the curiosity to search 
alternatives, instead of having to be stuck with disgusting stuff.



2. Many have stated a preference for Gnome 2, and all aren't
completely satisfied with the alternatives offered.  XFCE is regarded
by many as the best of a bad lot. This wasn't stated implicitly; I
assumed it was implied by context, but that was a mistake.


But MATE is a part of that bad lot, is not it?

And honestly, of course you will never find gnome in XFCE. It's why 
XFCE is XFCE, not a 3rd clone of gnome. (or a 1st considering it's age)
Goals are not the same, they are to keep something as lightweight as 
possible, but deeply integrated with other features of the desktop, 
unlike LXDE which wants to have only independent softwares, which can 
interact easily between them, but does not have a dependency. Different 
philosophy, where the gnome one seems to be to always provide everything 
possible, but only if it follows dev's opinion of what is nice.
But my point was, that maybe so many of those people would have be 
happy with XFCE's philosophy at start? People on windows stayed with XP 
for a long time after all, with it's "poor" but "efficient" GUI. And 
they now refuse to move to w8. (I know, it's not windows the subject 
here, but the analogy is not unrelated I think)


And those dev, I can not blame them. It's an open source software, 
where people makes stuff for themselves (not paid, AFAIK), which can 
possibly be useful for other people. That's the danger of things built 
by volunteers, and that's why it's so sad that so many people forgot the 
UNIX philosophy. Ancient gnome's users now have to maintain multiple 
forks for all the software suite.



That's possible.  I never said all XFCE users wanted MATE.


Good to read it explicitly.



My main goal is to get the point across that, like others have said,
the prejudice against MATE is strong.  I believe it's mainly
religious, and that the practical reasons against it are all 
solvable.


I do not feel as if I were religious. If DEs were a religion, then I 
would be an apostate. (heh... sounds like a nice DE's name if someday I 
wanted to start my own)
My interventions are more to say that, in short, I do not think that 
people should advertise that much about Mate on debian's user-list while 
those issues are not fixed:
_ absent from official repos (with all the problems it *might* provide, 
considering installations which *could* be more complex than it seems)
_ OP did not asked about which DE to use (people which want to use 
cinnamon often comes from gnome, and so must know about mate)


I hope I made my point more clear.


It's too big,
and it's just not something that appeals to me.


Agree. I think the same about mainstream DEs, in general. And I do not 
consider XFCE as a mainstream DE, even if it seem it could become thanks 
to gnome3. If a different paradigm were able to enter the mainstream 
DEs, it would really be a nice thing for DE users.



"berenger.mo...@neutralite.org, debian-user@lists.debian.org"

You should really fix your mail configuration :) (I do not really mind, 
but that could annoy other people)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/57072640062d64840f96ca90a5f39...@neutralite.org



Re: pcmanfm no window/gui, one user only

2013-10-14 Thread berenger . morel



Le 13.10.2013 13:44, Tazman Deville a écrit :

On Fri, Sep 27, 2013 at 08:11:10PM +0200, Tony Baldwin wrote:

Friends,

For some reason, when I try to use pcmanfm as my user, I can not.
I can start it as other users on the system, or with gksu or sudo,
but not for my user.


So it must be a configuration problem, or other users would have the 
same problem.
You could try a diff between your configuration and another user ( do 
not only compare pcmanfm's files, it could be related to other stuff ).


Or a longer but reliable technique:
1) move all your configuration files into another directory, so that 
the system won't be able to find them.

2) Run pcmanfm.
3) if it works, move 1 configuration file from the temp directory you 
made to it's original place. If it does not, move it to another 
directory (say, conf_problem, for example).

4) go to 2 until you still have files into the temporary directory.
5) Take a look at all files in conf_problem, and read them to 
understand what could be problematic. If you do not understand, then 
integrate their content bloc per bloc, as you did for files.


This is the desperate solution I use, when everything failed. It's long 
and boring, but always worked for me. Using diff (or meld, which is 
probably smarter since it is an interactive application, not a simple 
program. For comparing a lot of data, diff is not the right tool for me. 
Simply my opinion.) can give you a faster understanding


Or, really, what happens is, it seems to start (can find processes 
and kill them),

but no window appears. It seems to hang.

I tried to use gdb, but get no debug information.
What happens there is the whole thing (gdb and pcmanfm) hangs,
doing nothing, until I kill pcmanfm again, and gdb tells me nothing.


You need pcmanfm-dbg to have symbols, it might help. However, I have 
never tried it myself.
Also, you will need to put a breakpoint before running, or will need to 
manually interrupt application's workflow ( I have no idea about how to 
do that, but I know it is possible since some frontends seems to be able 
to do that)
But I honestly doubt that you will find anything interesting this way. 
Maybe if you compile it yourself, at least you could have the source 
code when debugging, which will give you more precise hints about where 
it have a problem.



When I run it from terminal, likewise, I find no errors, nothing.
Just hangs (even if I run the command with pcmanfm &, it just hangs 
the

terminal).


Sadly, it does not seem to have a verbose mode.

Yes, also, of course, I have killalled any such processes several 
times

before trying to start it again.


Several times is not useful, if it does not work at start, it won't 
work until the problem is here. At least for such "simple" programs.
I mean by simple, that it does not have to use lot of external info, 
only filesystem and user's action, when it runs. Speaking about 
filesystem, do you have a different partition than other users? Do you 
use cryptography? Or other file-related voodoo magic? That would not be 
very usual, but who knows...


I've also tried with other WMs (I use openbox as a standalone, but 
have
also now tried with LXDE and with wmii, and still no joy. Have no 
other

WMs on the system at this time).
This is on wheezy.


Again, if it works with a WM for other users, it should work with the 
same for you. So the problem is obviously not the WM.



I've even tried replacing my conf files in
~/.config/pcmanfm/{default,LXDE}/pcmanfm.conf
with the files from another user (and chowning them to me, of 
course),

to determine if there were something amiss in my config files,
but, alas, this too proved unproductive.
I don't know what else to do.


It may be a problem made by other configuration files. Like a gtk 
theme. No problem with other application? It could give you some hints.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/811d225888aa4b5040e0196fa912d...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-14 Thread berenger . morel



Le 13.10.2013 15:40, Joel Rees a écrit :
On Sat, Oct 12, 2013 at 7:11 AM,   
wrote:

[...] if you think that people are free to
give themselves the label they want, so you must accept that other 
are also

free to give the labels they want.
Long time ago, I studied the "dark side of computer sciences", and 
the first
things I have learn are that you can not claim to be a hacker, or 
elite,
or... If you do so, then people will name you lamer. You are a 
hacker if

other people recognize you as such.


There is a difference between the three words.


Of course.



Elite is something that truly elite people do not try to be. Nor do
they care if they are called such. That's the irony of "l33+".

Hacker is, again, not something you try to be. Either you hack or you
hesitate. People (like me) who tend to talk tend not to hack so much.

System administrator is actually a role that needs to be filled.

The truth here is simple: you are not what you want, only other 
people can

define who you really are.


Which I can acknowledge is relevant about "elite hackers", but I 
think

it's missing the point about system administrators.

If you (the general "you") own hardware that doesn't have a system
administrator, you need to fix that situation. Maybe it means you 
need

to step into the role.


Being able to fit the role for minor and simple tasks does not makes 
you able to do the whole role. Installing OSes and applications in a 
end-user way is quite trivial.
sysadmin is, for me and other (I hope), someone which is of course able 
to do those minor tasks, but also to manage really more difficult ones. 
Like installing OSes and applications automatically on more than one 
computer, depending on the use of that computer (programming, writing 
for the boss, taking care of money, etc).
For example, a windows sysadmin will need to understand how active 
directory works. In other things.
This is really more complex than the basic usage. Anyway, I think the 
problem we really have in this discussion is that we do not use same 
words for same things.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/9658a9f886e9b9ccd3e039d49ab02...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-14 Thread berenger . morel

Le 13.10.2013 14:41, Joel Rees a écrit :
On Sat, Oct 12, 2013 at 6:21 AM,   
wrote:

Le 11.10.2013 23:06, Brian a écrit :


"are you root?"



It does only means you own the system. Not that you can claim to be 
a

sysadmin. I own my car. I am not a mechanic, but I anyway have the
*authorizations* to tinker it. It's what root, or to be more 
precise, uid=0

means in linux OSes.


In some countries, owning a car does not authorize you to tinker with 
it.


I did not known that. Not even changing a wheel or repairing motor, 
direction? Sounds strange to me.
Here we have rights to tinker with our stuff, but to be authorized to 
use it on public space you need the vehicle to fit some conditions. 
Changing a motor to have a more powerful one or adding passenger places 
which were not originally thought are good examples. If you do some of 
those "tinkering", then you have to make a check (not yourself, of 
course, but by an organization. I do not know a lot more about that.)



Many who are the defacto admin for their system(s) do not claim to be
a sysadmin. But they are still the only admin the system has.


I have no idea about how it works in other countries, but in France, 
when the enterprise is big enough, sysadmins does not take care of 
single systems. That job is left to people with less qualifications.


Sysadmin has multiple meanings, and possession of a piece of paper 
is,

frankly, one of the less meaningful meanings I can think of.


Could not agree more. Sadly French guys seems to love those damned 
pieces of paper. It is quite problematic for self-learners (as I).
Sometimes I think that if I had better english skills I could try to 
work in other countries.



(I still
plan to take the LPIC level 2 when I have some extra money.)

But being able to install and update a debian box is part of what 
gets

tested in the LPIC exams.



If you can get a debian box up and a Fedora
box up, if you can read a shell script and have some idea what's 
going

on, if you can set apache up, if you can fiddle with your X server,
that's most of a passing grade on the LPIC level 1, and then you can
be a Jr. Sysadmin on paper.


You are right. But only (so, not being able to understand scripts) 
being able to install your debian box, and then to add it some softwares 
does not mean you could be a sysadmin.



(Well, there are a few more things you want to get down, too.
Permissions basics, basics of TCP-IP, SSH and such, but you generally
pick those up while you're learning how to install the system and
packages.)


I wonder if I could pass that test. 1st level does not seems so hard 
when I read you. How many does it costs?



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/b14a12e57f47953f6204e62ce51df...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-14 Thread berenger . morel



Le 13.10.2013 20:41, Chris Bannister a écrit :

On Sat, Oct 12, 2013 at 12:11:01AM +0200,
berenger.mo...@neutralite.org wrote:

>If you own a system you control it and can do whatever you like
>with it.
>You can give yourself whatever label you want (sysadmin,
>superuser, top
>dog etc) - it matters not. How about "Debian Despot"?

Oh, of course, if you speak about giving yourself a label, then,
fine. Take the one you want. But, it does not mean that you can
claim to be a professional, or that you can say someone is a
professional.


Nowhere has anyone said that administering your own system makes you 
a

professional system administrator.


In my mind, sysadmin is a term to qualify professionals, or people able 
to take that role on real organizations, not simply on the family level. 
(1)



Take the label you want. But if you take the label of "programmer"
because you can only write a hello world, and will own the source
code. But then do not be surprised if other people gives you the
label of "liar".


WTF! What is this "label" business? Do qualifications exist in your
country?


Of course they do. And they are regarded like if they were everything.
The problem is, that they mean almost nothing in reality.
Someone with a high qualification is not necessarily better than people 
without any.



When someone applies for a job, whatever "label" they give
themselves doesn't matter a hoot, if they don't have the necessary
experience/qualifications then they won't get the job.


A quite simplistic vision.


Is the level of corruption in your country an issue in this regard?


I can not judge corruption levels, but, for example, if you have a 
friend which works somewhere, it can become suddenly easier to work 
there, if he have a good reputation and talk for you.
Another example if is your friend (or family member, of course) 
represent a huge client of an enterprise, it can help too. You will be 
paid according to your qualifications, of course, and for some works you 
will need those qualifications and/or experience anyway, but it makes 
things easier.


Easier does not means automatic, of course.


It is the same with sysadmin. You can own your computer, be only
able to install softwares and use those excuses to label yourself a
sysadmin. But then, other people are also free to give you the label
of liar.


Don't confuse "role" and  "profession/career". A person can be a 
weekend

mechanic (role) but not be a mechanic (profession/career).


(1)
In the sysadmin role, there is much more than simply install things. 
Most people I know do not even upgrade their systems, if the system does 
not do it by default.

Luckily, windows have automatic updates.
Still, most people I know installs lot of crap, without understanding 
anything about that. Sometimes, they do not even understand that they 
are installing something.


How could I consider those people as sysadmin? It would be an insult to 
the real ones.
It is exactly like considering the 1st guy able to "build" an excel 
file with 3 formulas make by the software (you know, when you ask it to 
interpret your actions to make a formula, which can result in more 
"if-then-else" that the software can show... I have seen that, it really 
exists. I had to use gnumeric to be able to see the entire formula, and 
to paste it in a real text editor to understand it...), and call him 
programmer. Or to say that someone only able to change his car's oil 
with the oil someone gave him/her (so, not being able to check if it is 
the right oil considering the motor and the temperature of where he/she 
lives) is a mechanic.


Yes, that person would be able to take very minor, trivial parts of the 
role, but not the whole role. Really, maintaining your tools is part of 
their usage. So I would name people able to install his system a user. 
Maybe advanced user, depending on the mood and the guy. Not an admin. 
I'm not a mechanic. I'm a car user. I'm not a sysadmin, I am a computer 
user (well, not strictly true since I do some programming)



The truth here is simple: you are not what you want, only other
people can define who you really are.


That is very sad that you think this way.


I think it's how things work. People can want to be something a lot, 
being sysadmin or electrician or whatever, they won't become one by 
simple will. They need to work to become something. By work, I do not 
mean have a job, of course.
And if you effectively *are* something but nobody recognize you, then I 
do not think you will think to be one yourself.

At least, it is my opinion, and I can not see why it is sad.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/bd7db744fa325a368ad08d7bb6f2b...@neutralite.org



Re: Installing Cinnamon 2.0

2013-10-14 Thread berenger . morel



Le 13.10.2013 23:23, Bob Proulx a écrit :

Mark Allums wrote:

Ralf Mardorf wrote:
> I don't nag about endless lines,
> ...
> It's not the end of the world, but unfavourable when you argument 
about
> not running into issues with your computer, while causing issues 
for the

> list, when you're using Windows,
> ...
> IOW you might not experience issues with your computer, with Mate 
and/or

> Windows, but you at least cause issues when using Windows.


I believe it is _possible_ to use Windows without causing problems.
In theory anyway.  But it is definitely very much harder to do so.


In practice too. When I was a windows user, I did not gave problems to 
other people. At least, in the last years when I had enough knowledge :)



This has nothing to do with Windows.  You need to set your MUA to
use word wrap.  Stop bitching about Windows and fix your MUA.


Negative.  This is controlled by the sender.  If you want your text 
to

be word wrapped by the reader then the sender is obliged to set
format=flowed.  Outlook has long been wrongly sending long lined
messages without using that setting.  But it is wrong.

Bob


Or simply do not use HTML. I do not format my mails, I simply write 
them in plain text, and let something doing it. It seems to be ok for 
readers? ( this mail is a 3 line one, seen on my screen. And because 
this one is automatically wrapped). I believe it can be achieved with 
outlook or whatever crappy client you use.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/14546d05d3850e1b2b458d82306cb...@neutralite.org



Re: apt-get vs. aptitude

2013-10-14 Thread berenger . morel

Le 12.10.2013 18:10, msl09 a écrit :

Oh I have fond memories of aptitude breaking my system. Once it
suggested me to remove most of my system, including apt, I thought it
was going to upgrade it so I confirmed it. I had to reinstall apt 
from

the debian packages website.



That's why I do not use simple CLI tools to manage my system. They just 
fail to show important stuff, by flooding a lot of informations.
Using aptitude in it's ncurse mode would have prevented such a PEBCAK 
error, because updates and removal are shown with different colors.


But, it definitely is a user's... no, sorry, a root's error. A root 
should not act or think like a user. (damn, outside of linux context, 
that phrase really means nothing...)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1d4dcb83ee2bee862ae59cd20e3c5...@neutralite.org



Re: apt-get vs. aptitude

2013-10-14 Thread berenger . morel



Le 13.10.2013 04:18, green a écrit :

Tom H wrote at 2013-10-12 18:40 -0500:
I suspect that the problem's in the examples above are simply 
PEBKAC.


I use aptitude, and find it to be *more* useful than apt because of
its *interactive* dependency resolver.  Probably if people have
trouble with aptitude it is because the package selection scoring
(which is configurable) is not quite perfect for all cases and they 
do

not know about the interactive resolver (normally accessed with `e`)
when a package is marked as "broken".

The resolver allows the user to mark a particular suggestion of the
resolver (eg. "remove gnome") as *rejected* (`r`).  The resolver will
never suggest that action again and will usually soon find an
appropriate action, or that there is no possible action to satisfy
both the user and dependencies.

This feature of aptitude is somewhat more useful to me because I have
some extra repositories enabled beside Debian stable (I also use
pinning).


I love aptitude, but I really would like to be able to disable that 
solver. It makes aptitude slow when you are solving issues by yourself.


What I like with aptitude is the ncurse interface, with it's colors to 
show package's states, the preview, and to be able to browse in repo to 
find the package which fit your needs more than the others (debtags is 
nice for that).



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/70e96382252dfcdce184cec1ed269...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-14 Thread berenger . morel



Le 14.10.2013 10:37, Joe a écrit :

On Mon, 14 Oct 2013 07:53:50 +1300
Chris Bannister  wrote:


On Sat, Oct 12, 2013 at 09:19:18PM +0100, Joe wrote:
> > though most include routers and other
> > useless stuff.
>
>  ..when it is normally customary to refer to them as routers.
> Pedants might call them modem-routers, but nobody else does.

Um, you can get routers without a modem, so the difference is
important and not just pedantry!



Yes. I know that as well, having two myself. But I am unusual in that
respect, as exceedingly few homes or small-to-medium size businesses
have any need for even one.

In the large majority of networks, 'router' is taken to mean the 
(A)DSL

modem-router-DNS-server-DHCP-server-firewall-etc that plugs into the
telephone line.


And which does the interface between 2 networks. This one being the 
most important role from the end-user point of view, I think that the 
word router is good enough for them.


I know I wont teach that to anyone here, but modems are not computing 
stuff, at all. They are simply here to transform numeric signals to 
analogical ones, and vice versa. I wonder why someone would explicitly 
call the boxes "router-modem"... It would be like calling PC: 
computer-screen-mouse-keyboard-windows ( or linux, this one was for the 
trolling).
Nobody does, because it is obvious that the computer will need some 
interfaces with humans. For routers, it is quite obvious that it will 
need to be able to communicate with networks it is connected to, so 
really no use for the modem word here.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8b1b1851c8319dc2e96f97976940b...@neutralite.org



Re: apt-get vs. aptitude

2013-10-14 Thread berenger . morel



Le 13.10.2013 19:44, Frank McCormick a écrit :

On 13/10/13 01:02 PM, Chris Bannister wrote:

On Sat, Oct 12, 2013 at 01:10:14PM -0300, msl09 wrote:

Oh I have fond memories of aptitude breaking my system. Once it
suggested me to remove most of my system, including apt, I thought 
it
was going to upgrade it so I confirmed it. I had to reinstall apt 
from

the debian packages website.

In this new installation I gave it another try but when it started
suggesting very weird plans(like remove all gnome packages) ...


It seems like aptitude has gotten a lot smarter lately. :)




  Not this morning it wasn't.

  Aptitude has been refusing to do a full upgrade on my Jessie system
for the past two weeks because it said it needed xorg-video-abi-12 
but

it said it is not installable. Well, not so. I tried running Synaptic
this morning and it had no problem finding what it needed and
installing it. I still don't understand what the difference was but
Synaptic did what aptitude said it couldn't do. What could be the
difference ? Does Synaptic not use the same repo source files 
aptitude

uses?

Cheers


Did you tried synaptic just after aptitude, without updating the 
package list? If not, then maybe the package which gave you problems 
with aptitude was added by that update.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/e2b253c0735db78aa50ac87bf1d8e...@neutralite.org



Re: Installing Cinnamon 2.0

2013-10-14 Thread berenger . morel



Le 14.10.2013 15:29, Ralf Mardorf a écrit :
On Mon, 2013-10-14 at 15:21 +0200, berenger.mo...@neutralite.org 
wrote:

Or simply do not use HTML.


Endless line, for Evolution available by selecting "Preformatted":
This has nothing to do with HTML, I can format my emails using plain
text too, HTML is completely unneeded to do this.

Automatic line break after 72 signs, Evolution calls this "Normal":
This has nothing to do with HTML, I can format my emails using plain
text too, HTML is completely unneeded to do this.


Now that you say it, it's obvious.
I was simply thinking that HTML was the source of the problem, because 
I only noticed such problems with HTML mails. Was a stupid reasoning.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/c1cc8f7885e3f09cb6e809ca0e26a...@neutralite.org



Re: Linux Professional Institute Certification (Re: sysadmin qualifications (Re: apt-get vs. aptitude))

2013-10-14 Thread berenger . morel

Le 14.10.2013 17:01, Joel Rees a écrit :

I'm pretty sure the LPI site is translated into French, too. Yep.


I have absolutely no problem with English... at least when it is 
written :) my speaking is probably ugly, since I can rarely practice it.



Check it out:

http://www.lpi-francophonie.org/


< snip >

The English site is here:

http://www.lpi.org/



Thanks. I will take a look at that.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/560e4d38d7301f48a2a654899...@neutralite.org



Re: apt-get vs. aptitude

2013-10-14 Thread berenger . morel



Le 14.10.2013 16:38, Frank McCormick a écrit :

On 14/10/13 09:52 AM, berenger.mo...@neutralite.org wrote:



Le 13.10.2013 19:44, Frank McCormick a écrit :

On 13/10/13 01:02 PM, Chris Bannister wrote:

On Sat, Oct 12, 2013 at 01:10:14PM -0300, msl09 wrote:

Oh I have fond memories of aptitude breaking my system. Once it
suggested me to remove most of my system, including apt, I 
thought it
was going to upgrade it so I confirmed it. I had to reinstall apt 
from

the debian packages website.

In this new installation I gave it another try but when it 
started

suggesting very weird plans(like remove all gnome packages) ...


It seems like aptitude has gotten a lot smarter lately. :)




  Not this morning it wasn't.

  Aptitude has been refusing to do a full upgrade on my Jessie 
system
for the past two weeks because it said it needed xorg-video-abi-12 
but
it said it is not installable. Well, not so. I tried running 
Synaptic

this morning and it had no problem finding what it needed and
installing it. I still don't understand what the difference was but
Synaptic did what aptitude said it couldn't do. What could be the
difference ? Does Synaptic not use the same repo source files 
aptitude

uses?

Cheers


Did you tried synaptic just after aptitude, without updating the 
package

list? If not, then maybe the package which gave you problems with
aptitude was added by that update.




   Immediately afterward. I still don't understand it...then again
after using Linux for 10+ years...there are a lot of things I still
don't understand :)

Cheers


Hum... are recommended packages automatically installed by both 
softwares in your configuration? It may be the reason, if aptitude tries 
to upgrade something which recommend a 2nd package, which depends on the 
non-available package, so the recommended package is broken and you need 
to fix the breakage.
If synaptic does not, then the recommended package is not installed, 
and so not broken.


I do not really know, since I usually fix errors myself, by freezing or 
purging packages (I see no real interest into simple removal) which 
cause the problem... but I am really curious :)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/387fc00cab4b4aaa905ad90179d3e...@neutralite.org



Re: who could take the time to figure out how to contact linux ? so complicated !

2013-10-15 Thread berenger . morel



Le 14.10.2013 21:37, Konrad Neitzel a écrit :

Hi!

On Mon, 2013-10-14 at 13:57 -0500, Catherine Gramze wrote:


We need to know first if Larry has a successfully running Debian
installation, and if so, is it Wheezy, Jessie, Squeeze or Sid?



I think Larry needs some help to modify his sources.list file to
include the non-free repository. I know any Debian installation I 
have
done does not include the non-free repository by default. Once he 
has
that done he can use synaptic to update and then retry to install 
Flash

from there.


Note that it can be asked at installation time, if you do not use the 
default "question level". In expert mode, it will ask you if you want to 
enable contrib and/or non-free. I think that behavior is strange: people 
who use the expert mode are able to modify the sources.list file, but 
obviously people who are doing their 1st debian install are not, and 
they will not select the expert install.



So, Larry: there should be a file /etc/apt/sources.list on your
Debian computer. This file will have the ftp and http addresses your
computer uses to find the software that is prepackaged to run 
correctly

on your system. Can you find this file?


The non-free part is not required because the flashplugin-nonfree
package is inside the contrib part. (As it is written inside the 
wiki!)


With kind regards,

Konrad


contrib and non-free repos are not enabled by default installation, so 
the problem is the same :)




--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/18c85300ec8c03cfa3c80e1f1ac5d...@neutralite.org



Re: apt-get upgrade (security packages)

2013-10-15 Thread berenger . morel

Le 14.10.2013 22:11, Pol Hallen a écrit :
I can't everytime do updates from main repository because many 
packages

of this server are patched.


Using pinning for all of your packages is a solution, but I would not 
call it the easiest one.

Why not simply freezing them in aptitude/apt-*/dpkg?

For aptitude, I use it with the ncurse interface, so I do not know how 
to do that in command line. You will use the '=' key to freeze the 
package you currently have selected.

For apt-*, use apt-mark hold.
For dpkg... well, man dpkg :p (man apt-mark says that itself is a 
wrapper on dpkg, so you can find how to do what you need with few 
searches)



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/babcac55ba2571f5dc32f4c9a5f72...@neutralite.org



Re: pcmanfm no window/gui, one user only

2013-10-15 Thread berenger . morel



Le 15.10.2013 12:42, Tazman Deville a écrit :

On Mon, Oct 14, 2013 at 02:22:06PM +0200,
berenger.mo...@neutralite.org wrote:



Le 13.10.2013 13:44, Tazman Deville a écrit :
>On Fri, Sep 27, 2013 at 08:11:10PM +0200, Tony Baldwin wrote:
>>Friends,
>>
>>For some reason, when I try to use pcmanfm as my user, I can not.
>>I can start it as other users on the system, or with gksu or sudo,
>>but not for my user.

So it must be a configuration problem, or other users would have the
same problem.


This is one of the first things I thought,
but, as mentioned in my original message,
I replaced MY configs with those of another user
(and chowned them to me), and this made no difference.



Yes, but you only did this on pcmanfm's configuration files, right? I 
think the problem is not directly a pcmanfm's one, but maybe related to 
something it uses. This is why I suggested you to move all config files. 
Do something like that:


$ mkdir OLD_CONFIG
$ mv .* OLD_CONFIG
$ pcmanfm

Here, pcmanfm should work. If so, you can move back files one after one 
until pcmanfm stop to work again, so that you could identify which file 
is making problem.



So far, I'd only tried replacing the configs, not just simply making
then unavailable.


Making them unavailable will force the softwares to create the default 
ones, so you will have the default config.




...

okay, moving the .config/pcmanfm contents to somewhere, thus leaving
the directory empty (no configs) seems to make no difference.


Obviously. The problem does not come from pcmanfm, from what I can 
guess. pcmanfm not working is the consequence of the problem, not the 
problem itself.



You need pcmanfm-dbg to have symbols, it might help. However, I have
never tried it myself.


I do have that installed.


Honestly, I tried to use the *dgb files. I had no results, I was never 
able to understand their use.
It seems that it is better to compile the software with the -g option, 
and debug the compiled version. At least, you will have the source code, 
with comments, which makes things really easier, and you also might use 
some IDE or a debugger with a real human usable interface. I do not say 
that gdb is bad, but it needs learning. Lot of RTFM for even simple 
debugging...


The debugger I use (which are, in fact, simple frontends to gdb) are 
the plug-in of codeblocks, and cgdb (I know prefer this last one, except 
that I still do not know how to interrupt a program, unlike in C::B.). 
You can also use ddd, or the debugging feature in qtcreator, for 
example. All of them uses gdb in background, but they are far easier to 
use than gdb.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8e0001a4614d8d56a29793b0d0144...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-15 Thread berenger . morel



Le 15.10.2013 03:28, Catherine Gramze a écrit :

On Mon, 14 Oct 2013 21:12:32 -0400
Jerry Stuckle  wrote:



Maybe where you are, but not in the world scheme of things.

A router is a specific box.  A (A)DSL modem may also contain a
firewall, etc.  But most (A)DSL modems, cable modems, etc., only 
have

one Ethernet port.  So people install routers in addition to the one
which may or may not be in the (A)DSL modem.

Modems and routers are two entirely different things, with 
completely

different uses.  One box may contain both - but that does not mean
all modems are routers (or vice versa).

Hear, hear, Jerry! This is how I have always heard them referred to 
when
I worked as a network admin. A router is a router, and a "cable 
modem"
may or may not (usually not!) have any routing capability. It is 
really
a bridge connecting two networks, as I mentioned previously. It 
doesn't

do any modulating or demodulating. It simply allows the packets to go
from one network to the other.


I think that most people here knows about that: 
http://en.wikipedia.org/wiki/OSI_model


Modem are stuff of layer 1 (or 2, not sure), while routers works on 
layer 3 (or 4, I am not sure again... I'm not an expert on those stuff, 
I apologize). Modem and routers are different things, working on 
different levels.
So, routers may include modems, if they need to route packages between 
networks using different level 1 protocols. But modems have no reason to 
have routing features: they are on a very lower level.


Calling a router which include a modem a modem, would be the same as 
calling modem a computer. Because computers includes modems (it is not 
mandatory, but I think that nowadays, every computer includes a sound 
card able to have both input and output, right? So, to convert analog 
signals to numeric ones, and vice versa, which is the work of modems). 
Modems are not networking stuff, they are electrical stuff, which are 
sometimes used by some networking stuff.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/b613ff330a5fdbb51f2e748eddbc3...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-15 Thread berenger . morel

Le 15.10.2013 15:36, Jerry Stuckle a écrit :

If you want to confuse things by getting into the OSI model, modems
are Layer 1 (Physical).  Routers are Layer 3 (Network).


Thanks for precision.


I stayed away
from this because it's mainly of interest to engineers and network
programmers; the only thing most programmers have to worry about is
Layer 7 (Application).


Honestly, I think that it is often a good idea to know what is around 
your domain.
Programmers nowadays do not have to manage computer's memory too, but 
it seem that when they know how low level works they write more robust 
programs.
Sure, I am a programmer, and do not care at all about all those layers. 
But knowing that they exist allows me to have a better understanding 
about why something could have problems, and to speak with other people 
with different interests.


By example, in my last job, I was able to speak with the sysadmins, 
because I have knowledge about stuff that I do not need to know: linux, 
networking, firewalls (hardware and software), active directory, and 
that kind of stuff that programmers do not need. But it always helps to 
have knowledge you do not need, and if I did not liked that idea when I 
was younger, I now love it.
It is named general culture I think (if that translation from the 
French expression is correct).


In my opinion, programmers which only cares about their stupid 
languages and paradigms are doing a big error. You can learn as much 
high level stuff as you want. It you do not have a basic understanding 
of what allows you to stay so high, you take the risk of falling, and 
the higher you are, the more dangerous the fall is.
I would really like that teachers teach asm x86 at school. It is 
useless nowadays, right. Except that thanks to that, I had no problem to 
understand C pointers. Still thanks to that, I was very good in my 
electronic lessons, which in turns have helps me a lot in my 
programming, because I know why some operations are faster than others. 
Yes, it is useless, in the end. But it makes so many things damn easier 
to learn and tinker.


Plus, here, we are not discussing about programming, but about 
networking, so I think minimal knowledge of network stuff and some 
electric basis can be useful ;)



Calling a router which include a modem a modem, would be the same as
calling modem a computer. Because computers includes modems (it is 
not
mandatory, but I think that nowadays, every computer includes a 
sound
card able to have both input and output, right? So, to convert 
analog
signals to numeric ones, and vice versa, which is the work of 
modems).
Modems are not networking stuff, they are electrical stuff, which 
are

sometimes used by some networking stuff.




True.  Except it's normally the other way around.  Companies don't
design routers then add a modem to them; they design the modem then
decide if they need to add a router.  A small, but significant
difference.

Jerry


I think that this way is strange, but conception can be made in so many 
ways... The only important point in conception is, imho, that in the 
end, you should have good results.
Companies can build my netbook by creating cells and putting stuff 
around (random words, it is for the example. Plus, that would make sense 
since the cell determine how many power you can use), but I won't name 
my computer a cell.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/fbab2fde1a5e3741aa91b052109bc...@neutralite.org



Re: pcmanfm no window/gui, one user only

2013-10-15 Thread berenger . morel



Le 15.10.2013 23:38, Tazman Deville a écrit :

You're talking about moving every .configfile in my $HOME?


Yes.


That sounds a bit drastic, and possibly a bad idea.


Why that? It is only moving. Doing a backup, in fact. If you are feared 
by that, you can also create a new user, and copy configuration files 
one after the other until you have the same problematic behavior.



Can we narrow it down to what possible files might actually affect
it without implementing such drastic measures?


I have no other idea... but theoretically, the problems should be 
related to graphical stuff. I would guess that it could be linked to 
gtk, fonts, and stuff like that. Take a look at it's dependency chain, 
and see what in it have some configuration files, that should give you a 
good hint.



Honestly, I tried to use the *dgb files. I had no results, I was
never able to understand their use.
It seems that it is better to compile the software with the -g
option, and debug the compiled version. At least, you will have the
source code, with comments, which makes things really easier, and


Only if it's written in a language I can make sense of, of course.


No. Even without knowing a language, you should be able to read 
comments, variable names and functions names. I personally consider 
that, in fact, comments are not as useful as it is taught. Good names 
for classes, functions and variables are the best documentation and 
makes comments almost useless. ( note the almost here... some comments 
are useful, sometimes, when you need to do hacking tricks )



VIM IS THE ONLY IDE!!!


I like vim a lot, but it is not an IDE. It is only an editor. My IDE is 
my whole system: i3, gdb and cgdb, vim, git, lxterminal, 
bash-completion, meld, man, opera... and no classic all-in-one IDE can 
compete with it for my needs, this I can assure you ;)


I don't thnk we're getting anywhere with gdb or other debuggers, 
because

it's simply stopping and hanging at some point.
Or maybe some other debugger will be able to tell me where, and 
possibly

why?


As I said, gdb can. Because IDE are able to do so, and they are simply 
front-ends to gdb.

In fact, they are able to interrupt program at user's will.

Here is a trick I have learn something like 10 years ago, when I was 
studying reverse engineering ( well, ok, cracking, but what has always 
interested me was to add features to closed source softwares. Or 
removes, so not only the cracking stuff ) : run the targeted program, 
interrupt it when it shows the (un)desired behavior, and ask the 
debugger to "run until RET ( asm instruction )" several times so that 
you finally are in the program itself ( at interrupt time, you will 
probably be in system functions or in a library ). This way, you can 
find the section which interest you, quickly.


Sadly, I was used to do that with softice ( a ring0 debugger, a very 
nice and efficient tool, able to debug the OS itself. I wonder why rr0d 
- it means Rasta Ring 0 Debugger - , the only one I knows able to run 
other OSes than windows, is not so known...), on windows. w32disasm or 
ollydbg too.
Windows' debuggers are really very good tools, and I know no gdb 
frontend which can compete with them, sadly. But they are better for 
their interface, not because GDB lacks the feature. I simply do not know 
how to do that with GDB... If you or anyone else find how to, I am 
really interested.


But, I know that Code::Blocks is also able to interrupt a program you 
are debugging. I am certain of that, so GDB should be able to do so. I 
think I could find how it does it if I take the time to read the 
debugger plugin... I should do that now I think about it.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/9e59cd137715735ab9cad85bdae75...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-15 Thread berenger . morel



Le 15.10.2013 19:32, Chris Bannister a écrit :

On Mon, Oct 14, 2013 at 03:43:21PM +0200,
berenger.mo...@neutralite.org wrote:

I know I wont teach that to anyone here, but modems are not
computing stuff, at all. They are simply here to transform numeric
signals to analogical ones, and vice versa. I wonder why someone
would explicitly call the boxes "router-modem"...


Ummm, Under the router which is plugged into the phone line:
"ADSL2+ WiFi Modem Router" (without the hyphen!)

Under my other router "Wireless router" and yes, it has no modem
capability (as the name implies!)


Do you think the radio waves are binary signal?

So IOW, if you are purchasing a router which you also wish to use as 
a

modem then it better be written on the box!, cause the store owners
won't take too kindly in ripping off the shrink wrap to examine the
contents.


If I buy a computer to listen music, I should ensure myself that there 
is a sound card in it. Obviously. And I should take some more time to 
determine if that sound card is good enough for my needs, or change 
computer (or only adding the card, depending on my knowledge).



For routers, it is quite obvious that it
will need to be able to communicate with networks it is connected
to, so really no use for the modem word here.


Unless you need one with a modem!


Of course. If one of the networks is based on a numeric signal, and the 
other based on analog signal, you will need an adequate modem.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8d73b04435b91f2b53ddd57e10d0c...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-15 Thread berenger . morel



Le 15.10.2013 18:28, Jerry Stuckle a écrit :

Why?  Do you know how a TV signal is encoded at the station?  How it
is modulated onto the carrier?  The operation of the RF and IF strips
in your TV?  The frequencies of the local oscillator(s) being used?
How the RF signal is demodulated?  How the video and audio are
decoded?  How the video is displayed on the screen?  How the audio
ends up at the speakers?

Or do you just turn it on and watch your favorite show?


If I am a simple user, I can live without knowing nothing, yes. But I 
can hardly consider programmers as simple users... And, no, I do not 
know everything about the TV, but I know some stuff related to it's 
general working, like, for example, that to have an image I need to capt 
a radio signal from a satellite (ok, it's not the only way, I know), 
which is turned into an electrical analog signal, itself converted to a 
numerical one, which is finally interpreted to generate and image and 
sound.


And this small knowledge allows me to understand why and how I need to 
move the parabolic antenna, so that I am able to simply use my TV then, 
without the need of giving a lot of money to someone who will need lot 
of time to fix it for me (lot of time, because I'll have to wait him to 
go at my home). So, it makes me able to use my TV  quicker, and to fix 
minor problems if, by example, a strong wind moved the antenna.



By example, in my last job, I was able to speak with the sysadmins,
because I have knowledge about stuff that I do not need to know: 
linux,

networking, firewalls (hardware and software), active directory, and
that kind of stuff that programmers do not need. But it always helps 
to
have knowledge you do not need, and if I did not liked that idea 
when I

was younger, I now love it.
It is named general culture I think (if that translation from the 
French

expression is correct).



None of which has anything to do with the OSI layers or programming.
They are all sysadmin functions.


Yes, I was speaking about the interest of being able to understand 
something that you do not really need to know.



It doesn't hurt to have knowledge you don't need now.  But this world
is way too complex to know everything about everything.


I agree. But to be able to communicate with someone which knows what 
you ignore, you will need common bases to explain what you need. People 
with very very good knowledge of their domain, but no knowledge in 
others' domains will need someone able to act as an translator. I have 
heard lot of stories of very very good computer scientist who mades very 
very good programs. But the programs were not used, because they were 
not adequate to user's needs.


That's why people have to know basics of other's sciences, and why 
general culture is important, for me. But maybe I think that because it 
is the way teaching is made in France, from centuries. I know that the 
biological knowledge (very basic one, of course, stuff about cellular 
atomic kernels, DNA, and alike.) I have acquired at a time is not very 
useful. But it made me able to discuss with my neighbor when I was in 
higher studies, and also inspired me some ideas for algorithms.



Just like I
don't program in assembler (for Intel or Motorola MPUs or IBM
mainframes), although I could do any of them still.


I do not do it either. But by being able to do so, I can understood why 
some instructions will slow down programs more than others. Of course, 
early optimization is root of evil, but I know that if I have to 
divide/multiply integers by a power of 2, I can use the << and >>  
operators. It also helps me when I need to debug programs, even if I do 
not have the source code.



Pointers have nothing to do with assembler.


Pointers are memory addresses, which are very important in asm. So, 
yes, knowing asm helped me a lot to understand C pointers. I understood 
them without any problem, unlike my classmates. And those guys were, as 
me, coming from electronic studies, so they were supposed to know basics 
about processors.



C is not the only
language with pointers


Of course. They all need to use them if they offer dynamic stuff, but 
they try to hide them. Is it the good solution or not? I do not know, 
but if yes, I wonder why most games are written in C or C++? I think 
that guys who write them knows what is memory, and how it works. I hope 
for them at least.



But it doesn't require knowledge of underlying memory
access to use pointers.
I've taught many C and C++ classes over the
years (I used to do corporate programmer training), and never once 
did

we get into how memory works.


I wonder how you taught how to use them, so. Just saying that they are 
variables which indicates where in memory are located other variables? 
And people just accept that?



Those who never programmed in languages
with pointers, i.e. COBOL, had a little trouble getting used to them,


Yes. It is what I said. General culture helps to learn new things 
fast

Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-15 Thread berenger . morel

Le 16.10.2013 00:42, Miles Fidelman a écrit :

I'm trying to figure out what kinds of things you see "programmers"
working on that don't need serious knowledge of the underlying
operating system, computer hardware, and i/o environment.


I do not think every programmer needs serious knowledge outside the 
scope of his application, but I really think we need at least bases, as 
a minimal requirement.


I had a boss which said me, one time, that he preferred people able to 
understand C/asm/CPUs to build SQL requests. The link? People without 
that "low level" knowledge do not understand why it takes time to 
compare strings, and their requests were too slow. Just hell too slow.


Saying that programmers do not need to care about hardware, is 
something my teachers tried to convince me about. Luckily enough, I had 
knowledge at that time, being self-learner.
Now, take a look at real world, and explain me why we need so powerful 
computers to do basic text formatting? Explains me, why we need to 
destroy the planet to build more powerful hardware, which requires 
living and dead memory, and so more energy to run?
My explanation is, because people of the past taught to other guys 
things like "Take a int to store that value which is comprised between 0 
and 10. If client needs to make the software faster, he can just buy new 
hardware.", when the correct phrase should have be: "Since we know that 
this variable have a limited range of values, take an unsigned char. 
That will cost nothing to you.".


We even have languages that are not aware that there are unsigned 
types... like java. Encouraging people to be expert of their domain and 
being ignorant of other things will lead to inefficient softwares, even 
if they might work. Ah, and, of course, not knowing the hardware can 
also cause "strange" bugs, like a value which become a negative number 
after an increment. Or the classic problems with real numbers.
Basic hardware comprehension is useless to build programs? That's 
exactly the reason why I take so much care when I select the softwares I 
install on my computer, which makes my computers with cheap hardware 
faster than some with expensive hardware.


I consider that building efficient softwares is a duty, not a choice.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/e2fe3f037375089bcd0170ac8055c...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-16 Thread berenger . morel

Le 16.10.2013 03:25, Jerry Stuckle a écrit :

Ah, but you are more than a "simple user".


I guess so. I am not even a TV user anymore in fact, but that's not the 
question.
The point is that I can hardly consider a programmer to be a simple 
user of a computer, because when you write a program, you will probably 
have to know how to install it, and so know the system you target. 
Because it's the programmer who knows what the program needs. Root 
access maybe? Or will it listen on a port? Which configuration files 
will it needs? Which installed lib?
Those are not the work of the admin, even if the admins can be able to 
understand about what the programmer is talking, so that they can then 
know when there is a problem how to fix it.



Just like I
don't program in assembler (for Intel or Motorola MPUs or IBM
mainframes), although I could do any of them still.


I do not do it either. But by being able to do so, I can understood 
why
some instructions will slow down programs more than others. Of 
course,

early optimization is root of evil, but I know that if I have to
divide/multiply integers by a power of 2, I can use the << and >>
operators. It also helps me when I need to debug programs, even if I 
do

not have the source code.


Pointers have nothing to do with assembler.


Pointers are memory addresses, which are very important in asm. So, 
yes,
knowing asm helped me a lot to understand C pointers. I understood 
them
without any problem, unlike my classmates. And those guys were, as 
me,

coming from electronic studies, so they were supposed to know basics
about processors.



Yes, my C/C++ students sometimes had initial problems with pointers,
but some real-world (non-programming) examples got the point across
quickly and they grew to at least accept them, if they didn't like
them. :)


I only know few people who actually likes them :)
I liked them too, at a time, but since I can now use standard smart 
pointers in C++, I tend to avoid them. I had so much troubles with them, 
so now I only use them for polymorphism and sometimes RTTI.
I hope that someday references will become usable in standard 
containers... (I think they are not because of technical problems, but I 
do not know a lot about that. C++ is easy to learn, but hard to master.)



C is not the only
language with pointers


Of course. They all need to use them if they offer dynamic stuff, 
but
they try to hide them. Is it the good solution or not? I do not 
know,

but if yes, I wonder why most games are written in C or C++? I think
that guys who write them knows what is memory, and how it works. I 
hope

for them at least.



C was never meant to be an applications language - K&R designed it
for creating OS's (specifically Unix).  But because of that design, a
good programmer can write code that is smaller and faster than with
other languages (except assembler, of course).


Yep. It is designed to be an efficient language, allowing to give 
people full control on their tool, in a portable way. This is risky, 
because you can shoot your feet, but taking that risk is needed to have 
efficient softwares.


Plus, in an OS, there are applications. Kernels, drivers, and 
applications.
Take windows, and say honestly that it does not contains applications? 
explorer, mspaint, calc, msconfig, notepad, etc. Those are applications, 
nothing more, nothing less, and they are part of the OS. They simply 
have to manage with the OS's API, as you will with any other 
applications. Of course, you can use more and more layers between your 
application the the OS's API, to stay in a pure windows environment, 
there are (or were) for example MFC and .NET. To be more general, Qt, 
wxWidgets, gtk are other tools.


For Debian, in it's standard installation (I insist on the standard 
installation, the one I never do), it will come with the gnome DE. I do 
not know the tools it provides, but they are probably applications, too. 
And it is part of the Debian OS.
I know, OSes have evolved since the first UNIX. But languages and the 
libs available in them too. C was invented 40 years ago. I have seen 
some of the codes which were valid at that time, and it really had great 
enhancements (imo).
But all of this have nothing related to the need of understanding 
basics of what you use when doing a program. Not understanding how a 
resources you acquired works in its big lines, imply that you will not 
be able to manage it correctly by yourself. It is valid for RAM memory, 
but also for CPU, network sockets, etc.



A bigger advantage is
the code is machine-independent.


Which is why C and his little brother C++ are probably the reason of my 
switch to linux. See, if those languages were never used to write 
applications, there would not be so many portable and efficient one, and 
so I would have probably stayed to windows (and not annoying people on 
that list :p), instead of changing my tools one after one until I was 
able to change the OS itself without chang

Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-16 Thread berenger . morel



Le 16.10.2013 08:24, Erwan David a écrit :
On Wed, Oct 16, 2013 at 01:10:42AM CEST, 
berenger.mo...@neutralite.org said:



Le 15.10.2013 19:32, Chris Bannister a écrit :
>On Mon, Oct 14, 2013 at 03:43:21PM +0200,
>berenger.mo...@neutralite.org wrote:
>>I know I wont teach that to anyone here, but modems are not
>>computing stuff, at all. They are simply here to transform numeric
>>signals to analogical ones, and vice versa. I wonder why someone
>>would explicitly call the boxes "router-modem"...
>
>Ummm, Under the router which is plugged into the phone line:
>"ADSL2+ WiFi Modem Router" (without the hyphen!)
>
>Under my other router "Wireless router" and yes, it has no modem
>capability (as the name implies!)

Do you think the radio waves are binary signal?


Just as much as the electric signal in an ethernet cable...


You got me :)


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/3c878f1e4893496527843544e6c90...@neutralite.org



Re: sysadmin qualifications (Re: apt-get vs. aptitude)

2013-10-16 Thread berenger . morel

Le 16.10.2013 13:04, Jerry Stuckle a écrit :
Anybody who thinks that being able to write code (be it Java, C, or 
.NET
crap), without knowing a lot about the environment their code is 
going

to run in, much less general analytic and design skills, is going to
have a very short-lived career.



Anyone who can't write good cross-platform code which doesn't depend
on specific hardware and software has already limited his career.
Anyone who can write good cross-platform code has a much greater
career ahead of him. It is much harder than writing platform-specific
code.



If writing portable code is harder than platform-specific code (which 
is arguable nowadays), then, could it be because you have to take about 
type's min/max values? To take care to be able to use /home/foo/.bar, 
/home/foo/.config/bar, c:\users\foo\I\do\not\know\what depending on the 
platform and what the system provides? Those are, of course, only 
examples.


However, I disagree that it is harder, because checking those kind of 
problems should be made in every software. For types limitations, never 
taking care of them is a good way to have over/under-flow problems, and 
for the file considerations, what if the user have a different 
configuration than what you thought, on the same system (Of course, I 
know that you can access system variables to solve those problems)?
Does not it means that you have to know basics of your targets to be 
able to take care of difference between them?


Of course, you can simply rely on portable libs. But then, when you 
have a bug which does not comes from what you did, how can you determine 
that it comes from a lib you used?


I remember having a portability problem, once. A code worked perfectly 
on a compiler, and not at all on another one. It was not a problem of 
hardware, but of software: both had to do a choice on a standard's lack 
of specification (which is something I did not known at that point. I 
have never read the standard sadly.). I had to take a look at asm 
generated code for both compilers to understand the error, and find a 
workaround.
What allowed me to understand the problem, was that I had that asm 
knowledge, which was not a requirement to do what I did.


Of course, I have far less experience and grades than it seem you both 
have, and if I gave a minimalistic sample of the problem you could think 
that it was stupid, but it does not change that I only was able to fix 
the problem because of my knowledge of stuff that I have no real need to 
know.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/3fdf3b9523eb59e72677cafa445e5...@neutralite.org



  1   2   3   4   5   6   7   >