libapache-mod-perl

[Justin A. McCright]

Is anyone going to try and get a version of this that works with newer
Apaches out before the freeze?


-- 
-jam

-BEGIN GEEK CODE BLOCK-
Version: 3.1
GCS d-(--) s: a--->? C++()>$ UL>$ P+ L+++>$ E W++(-)>$ N- o? K? w---
O-- M-- V? PS PE Y+ PGP t++ 5++ X+ R tv++ b+++(++) DI>+ D++ G>++ e-> h!>---
!r>+++ y
--END GEEK CODE BLOCK--

Re: libapache-mod-perl

[Dan Jacobowitz]

On Sun, Oct 11, 1998 at 08:01:42PM -0500, Justin A. McCright wrote:
> Is anyone going to try and get a version of this that works with newer
> Apaches out before the freeze?

I'm working on it.  It's very broken at the moment!

Dan

Re: The freeze and IMMINENT 2.2.0p1!!

[Manoj Srivastava]

Hi,
>>"Santiago" == Santiago Vila <[EMAIL PROTECTED]> writes:


 Santiago> Well, kernel-package is a single package but it would be surely
 Santiago> a lot of work, since there are a lot of new drivers.

What does not work for you using kernel-package on newer
 kernels? I have never had a problem, so far, all the way upto
 2.1.125. 

manoj
-- 
 They are ill discoverers that think there is no land, when they can
 see nothing but sea. Francis Bacon
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E

Re: KDE hurts Qt (was Re: LICENSES)

[jim]

> Chris Waters <[EMAIL PROTECTED]> wrote:
> > This is the really sad part about this whole mess. Qt is a nice
> > library. Non-free, but not everything has to be free. But because of
> > the refusal of the KDE developers to FIX THE KDE LICENSE PROBLEMS, a
> > lot of people are being turned off of Qt! Qt doesn't deserve this, and
> > I think the KDE team should: 1) fix their license problems, and 2)
> > apologize to Trolltech.
> 
> I think they should fix their license problems, but I do not think that an
> apology is warranted.  After all, the Qt license explicitly *encourages*
> application developers to license their code with the GPL.

And that's a good thing... but wil the GPL prevent distribution of binaries
made linked to Qt, or object files compiled against Qt?

If so, then is TrollTech implying falsehoods about the GPL? I.E., are they
implying that people can distribute such binaries? Or, is it the KDE folx that
are doing that implying? Or both?

If TrollTech is saying that people can release GPLed code compiled/linked
against Qt and this is not so and they are refusing to fix this, what is
the reason for this? Are they trying to get free software and/or ideas that
won't be permitted to be released in a normal way? How does this benefit
them, if in any way at all?

If it's KDE saying the same thing, then how do they benefit, if at all?

If both, then how does the system containing both Qt and KDE benefit?

If neither, why does the controversy exist?

> If anybody should apologize, it's Trolltech.  [If they choose.]

Maybe... I'm having trouble keeping up with who should apologise to whom...

I think KDE's use of Qt hurts KDE and also hurts the open source concept.
Whether it hurts Qt is not relevent, IMO. However, because Qt is non-free,
Qt also tends to harm free-software development. 

At one point, Bruce Perens suggested this harm is at the very core of linux.
I don't agree with this because the very core of linux is the kernel and only
the kernel. Outside of that, there are wayyy too many combinations of software
packages that people use regularly and each, to whatever degree, has its
following. So, I don't see this as one of the larger issues in the mass
usage of linux and supporting entities as a whole. However, I do see that
if linux were easier to install and configure, it would be in greater use.
I also see that if linux had a common face, this could also increase usage.
But I don't see linux getting a common face anytime soon, as each individual
will continue to use those software packages they prefer.

I think it boils down to this: everyone involved should read and understand
the GPL (and the ORIGINAL INTENT thereof). Everyone who distributes software
might need to consult a lawyer (especially KDE and Qt makers) to ensure that
their use of the GPL is actually in their interest, and that they are using
it correctly according to its letter and its original intent. Whoever is
using the GPL in a way that conflicts with its terms should stop immediately
and correct the situation.

Without further study into the situation, I can't tell who is or is not
doing so.

-Jim

Intent to package: oo2c, The Oberon-2 to ANSI-C Translator

[Anthony Fok]

Hello!  :-)  I hereby announce my intent to package (and upload) oo2c, The
(Optimizing) Oberon-2 to ANSI-C Translator.  I will try to upload it soon. 
:-)  Well, actually, already built it, but will rebuild it again and do some
more testing.  :-)

Package: oo2c
Priority: optional
Section: devel
Installed-Size: 3197
Maintainer: Anthony Fok <[EMAIL PROTECTED]>
Version: 1.3.10-1
Depends: libc6 (>= 2.0.7u), libgc4 (>= 4.12-4.13alpha1-1)
Description: The Oberon-2 to ANSI-C Translator
 oo2c is the first compiler of the Optimizing Oberon-2 Compiler project. 
 OOC's goal is to provide optimizing native code Oberon-2 compilers for a
 number of popular computer architectures.  The OOC home page is
   http://www.uni-kl.de/OOC/

Now, should I learn and use Modula-2, Modula-3 or Oberon-2?  Tough decision. 
(What about C/C++ and Ada?  Ahhh!  I don't know!  :-)

Cheers,

Anthony

-- 
Anthony Fok Tung-LingCivil and Environmental Engineering
[EMAIL PROTECTED], [EMAIL PROTECTED]University of Alberta, Canada
[EMAIL PROTECTED] Keep smiling!  *^_^*
Come visit Our Lady of Victory Camp -- http://olvc.home.ml.org/
or http://www.ualberta.ca/~foka/OLVC/

Re: Perl 5.005.02

[warp]

On Sun, Oct 11, 1998 at 11:28:06AM -0400, Roderick Schertler wrote:
> On 11 Oct 1998 03:08:22 -0700, "Darren/Torin/Who Ever..." <[EMAIL PROTECTED]> 
> said:
> > Andy Dougherty, in an immanent manifestation of deity, wrote:
> >>
> >> After some thought, I think I'd recommend that perl5.005_xx retain the
> >> same directory structure that perl5.00[34]_xx did. (with 5.005 in place of
> >> 5.00[34], of course).
> >
> > That's good enough for me.  I have boatloads of respect for Andy and his
> > understanding of Perl Install issues.  That's how it will be for
> > 5.005.02-3.
> 
> I don't think Andy is taking into account your plan of allowing both
> threaded and non-threaded Perls present on the system at the same time.
> 
> It seems to me that the only real problem we've got with the current
> layout is that the *.pm files for extensions which have XS portions are
> placed in /usr/lib/perl5 rather than in /usr/lib/perl5//.
> If that were changed so that such modules behaved like core modules
> (placing both *.pm and *.so in the arch/version hierarchy) then only
> modules with *.so files would need to be reinstalled when a new version
> is installed.  If this were in place previously then the recent 5.005
> install would only have broken modules with XS portions, as it should
> have.
> 
> Further, if the  part of that didn't necessarily track
> every new version, but only changed when a new version was binary
> incompatible, that would save even more recompilation.  That is, if
> 5.006 doesn't break binary compatibility, I'd recommend that you
> continue to use /usr/lib/perl5//5.005 rather than changing to
> /usr/lib/perl5//perl5.006.

I would suggest using symlinks here, have the version be the first one
which was binary compatible with all binarys there, and then for the
rest just symlink the new version dirs to the old until binary
compatibility is broken..

Also, when possible, could we try to lean towards having support for
more then one version of perl installed at the same time even if
currently its not possible?

(See my preavous posts on perl on a possible way to allow having more
then one installed at the same time)

Zephaniah E, Hull.
> 
> -- 
> Roderick Schertler
> [EMAIL PROTECTED]
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 


pgpnu47V19lyT.pgp
Description: PGP signature

Re: Perl policy for managing modules ?

[Manoj Srivastava]

Hi,
>>"Raphael" == Raphael Hertzog <[EMAIL PROTECTED]> writes:

 Raphael> Le Thu, Oct 08, 1998 at 09:57:36AM -0400, Dan Jacobowitz écrivait:
 >> My question is, why are we so intent on removing the versioned
 >> component even though we have lost binary compatibility?  I understand

 Raphael> Because most of the perl modules will work with all coming
 Raphael> version of perl.

I guess I am confused. How do we know this? How can we be
 certain that there shall not be any further binary incompatibilities
 introduced in Perl? The fact that the upstream releases are
 maintianing the versioned component seems to imply that no such
 (implied) promise has been made by the perl 5 porters group.

 Raphael> Because Debian doesn't allow different version of the same
 Raphael> package to be on the same machine.

Hmm. The statement is indeed true, I just seem to be too dense
 to see how it applies. Say the new Perl (released in jan 1999, say)
 is binary incompatible to 5.005. I have a module, foo. How does this
 scheme prevent a recompilation then?

Are we just putting all our eggs in the basket that despite
 the record, there shall never be any incompatible perl releases?

 Raphael> Because we don't want to have to update the packages that
 Raphael> contains modules each time a new perl version comes out.

This, I agree with. The solution proposed either is a kludge,
 or I am missing the point.

Me, I would create the current and all future versions of Perl
 which are binary compatible to have the link:
 /usr/lib/perl5/$arch/$version --> /usr/lib/perl5/$arch/

No modules need be recompiled when the new perl comes out,
 since all of them really reside in  dir.

When a new incompatible Perl comes out, we change the , and recompile.

What am I missing?

manoj
--
 When you meet a master swordsman, show him your sword. When you meet
 a man who is not a poet, do not show him your poem. Rinzai, ninth
 century Zen master
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E

Re: intent to remove libglide from non-free

[James A. Treacy]

On Sun, Oct 11, 1998 at 01:48:43PM -0700, Chris Waters wrote:
> James A. Treacy wrote:
> 
> > A number of people would like to see a 3dfx package of mesa. This can
> > not be done unless there is a legal package of glide (under the
> > current license I can't even get the libs since I don't own a 3dfx
> > card).
> 
> Any reason, aside from the lack of volunteers, why we can't do what we
> do with netscape/staroffice/etc.?  Even if we can't distribute it, can't
> we have a loader package?  (No, I'm not volunteering, I don't own a 3dfx
> card either.)
> 
Those are precompiled binaries that we don't distribute. Basically the
people who write the libglide licenses are idiots. The license makes some
sense for the library, but for header files it is absurd.

I am not working on mesa-glide packages until this license mess is worked
out. I don't have time to waste on it myself.

Jay Treacy

Re: LICENSES [was: Re: Have you seen this?]

[jim]

> > In my opinion, Qt is not a section of KDE, it is not derived from the
> > KDE and it must be considered independent and separate from the KDE.
> > In other words: The KDE's usage of the GPL does not cause the GPL, and
> > its terms, to apply to Qt.
> 
> Indeed Qt is not part of the problem

indeed

> > But when you distribute the same sections as part of a
> > whole which is a work based on the Program, the distribution of
> > the whole must be on the terms of this License, whose permissions
> > for other licensees extend to the entire whole, and thus to each
> > and every part regardless of who wrote it.
> > 
> > Qt is not distributed as part of KDE.  It is distributed as part of
> > various distributions that also include the KDE, but only by "mere
> > aggregation [...] on a volume of a storage or distribution medium"
> > which the GPL okays elsewhere in the text.
> 
> It is not a mere aggregation. If I remove Qt KDE is unusable. 

I note here, a point that most ppl probably know: It is certainly not the
fault of the Qt ppl that KDE would be unusable without Qt's presence. Qt is
not owned by the KDE ppl, so Qt folx should not be affected by KDE's 
decision to use KDE. If, on the other hand, Qt decides to say that they
"like" (whatever that means, and in whatever form it comes) KDE's use of Qt,
then at that point, they are possibly in it together. For example, if Qt does
anything to facilitate KDE's success in its present (with-Qt) state, then
I would take on the opinion that they are both in the situation together, and
responsible for the results thereof. (Like my opinion means anything :)

> KDE requires Qt currently. So KDE is non free. 

_No_. This does not necessarily follow, even if both statements may
both be true. KDE simply depends on something that is non-free. If KDE
itself can be (1) obtained in source, (2) altered and then (3)
redistributed in its altered form and (4) have all these with the only
restriction being that further restriction cannot be applied, it is
free. If it presently depends on something that is non-free, we know
the drill: the source is available. Therefore, anyone can fix it.

> Similarly Linus does not distribute KDE with the kernel so its not in
> the base distribution. 

I see your point here as "KDE does not come with the linux os (that being
the kernel) therefore the clause in the GPL, making an exception for a piece
of software included with an OS, does not apply." I would agree, however we
should watch this MicroSoft lawsuit carefully: it might redefine what can be
considered part of an OS. (Of course, it won't change what Linus distributes.)

I would make the same observation about Qt: Linus doesn't distribute Qt
with the kernel either. So KDE would have a reason to not assume the exception
in the GPL wrt Qt.

> On Solaris KDE is shipped even though no Sun
> product includes Qt. So the case there is even more blatant

Could you elaborate a bit here, if only addressing the elaboration to me?
why is this more blatent?

-Jim

Re: intent to remove libglide from non-free

[Joseph Carter]

On Sun, Oct 11, 1998 at 01:48:43PM -0700, Chris Waters wrote:
> Any reason, aside from the lack of volunteers, why we can't do what we
> do with netscape/staroffice/etc.?  Even if we can't distribute it, can't
> we have a loader package?  (No, I'm not volunteering, I don't own a 3dfx
> card either.)

Someone wanna send me a 3dfx?  I'll make an installer if I have one so I can
get the packages legally..  =>


pgp2jDiaV8b4B.pgp
Description: PGP signature

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Joseph Carter]

On Sun, Oct 11, 1998 at 10:52:19PM +0200, Gergely Madarasz wrote:
> > [...]
> > I agree that by using XForms in development, and XForms *is* needed to
> > compile and run LyX, we have implicitly allowd all users to link Lyx
> > with XForms.
> > [...]
> 
> I don't think so. It is not enough for KDE, why should it be enough for
> LyX ?

It's not enough for KDE because KDE includes things not written by the KDE
people.


pgpcyhQQtMrHq.pgp
Description: PGP signature

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Shaleh]

The main difference is that LyX is THEIR code.  The problem w/ KDE is not so
much its own code, rather it links other peoples GPL app w/QT and KDE to make
Kapp.  This is the brunt of the legal issue.  The authors of app where not asked
if it was ok to link w/QT, nor was the license modified to reflect this.

On 11-Oct-98 Gergely Madarasz wrote:
> On Sun, 11 Oct 1998, Michael Meskes wrote:
> 
>> Would this be enough for LyX? I think so.
>>
>> 
>> [...]
>> I agree that by using XForms in development, and XForms *is* needed to
>> compile and run LyX, we have implicitly allowd all users to link Lyx
>> with XForms.
>> [...]
> 
> I don't think so. It is not enough for KDE, why should it be enough for
> LyX ?
> 
> Greg
> 
> --
> Madarasz Gergely   [EMAIL PROTECTED] [EMAIL PROTECTED]
>   It's practically impossible to look at a penguin and feel angry.
>   Egy pingvinre gyakorlatilag lehetetlen haragosan nezni.
> HuLUG: http://mlf.linux.rulez.org/
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

--
E-Mail: Shaleh <[EMAIL PROTECTED]>
Date: 11-Oct-98
Time: 22:46:46

This message was sent by XFMail
--

Re: LICENSES [was: Re: Have you seen this?]

[jim]

> > However, the license for that derived work (I'll call it A) claims
> > that the whole of A must be GPL'd.  However, Qt is not part of A (the
> > GPL says "section of").  Qt provides services to A, and A depends on
> > those services: A very different thing.
> 
> Qt is part of the derived work. It is linked to it and the work A does not
> function without it. It is also not a public API and your message to Preston
> concerning possible legal action against harmony makes it clear you regard
> the item as actively protected IPR not an open API

I understood the meaning of "is derived from" to be "using the source code
to make a derivative of" as opposed to "using the services of".

If I use libc, I don't think I am creating a libc. Unless I am, I'm not
deriving, I think. If I use libc, I simply use the services. Hence, libc
is "a section of" the thing I am making, and does not derive from it.

How is this wrong? Is it "strategic" to look at sections as derivations?

-Jim

dbugging dselect

[David Stern]

Hi,

I'm having troubles with dselect in slink.  I tried addressing this 
issue in debian-user, but got absoutely no response, so I thought I'd 
try here.  I haven't been on this list for long, so if this is in any 
way inappropriate, I apologize.  I'm very grateful to and have great 
respect for all the developers.

A big part of the difficulty for me is figuring out exactly what 
problem I have. What happened was that when I tried to update my 
packages the other
night using dselect, a number of packages, mostly perl and pam, had 
errors.  Now dselect is stuck.  I can update the packages file, I can 
switch to another install method, but when it comes to running install, 
no go.

I've been stumbling around in bug tracking, and I successfully applied 
one "quick fix" (export PERL5LIB=/usr/lib/perl5), which bought me 
updating my available packages and changing access methods, but I still 
can't install.

If this is waiting for a fix, I don't mind waiting, I just want to know 
what I should be keeping an eye open for when a fix becomes available.  
 It's been a good number of days now.  OTOH, if I need to download some 
packages and manually install em, then I'd like to have at it.  You 
see..

  install attempt in dselect error message:
  -
  [..unmet dependencies, blah, blah.. ]
  dependencies:
perl-suid: Depends:perl
libpam0: Depends:libpam0g
perl: Depends:perl-base

Thanks,

David

Re: LICENSES [was: Re: Have you seen this?]

[Alan Cox]

> If I use libc, I don't think I am creating a libc. Unless I am, I'm not
> deriving, I think. If I use libc, I simply use the services. Hence, libc
> is "a section of" the thing I am making, and does not derive from it.

Your program derives from libc by being linked with it. This is precisely
why an LGPL has to exist. 

Alan

Re: Finding a source package

[Guy Maor]

Joey Hess <[EMAIL PROTECTED]> writes:

> Guy Maor wrote:
> > I'm suggesting that dpkg-scanpackages scan the dscs and put the
> > section and version in the Source field, or perhaps add a new field
> > Dsc which is simply the full path to the dsc, akin to the Filename
> > field.  Then downloading the source for a package is simple.  The web
> > pages would use this field too.
> 
> But .dsc's are pgp signed by thier creators. How would you edit them?

I'm speaking of the fields in the Packages line.  By fixing those, I
get the correct info into dpkg/apt's database.  It can then find the
dsc.  From that it can find the rest.


Guy

Re: Perl 5.005.02

[cfm]

My $.02 on this - and this is only personal feedback-
is that perl -MCPAN -e shell is even easier than 
apt-get.  So I maintain a perl5.tgz with our
various modules from CPAN already installed and whenever
debian blows away our perl on any particular machine we
just unpack it from our local distribution tree, maybe updating
with a new DebianNet.pm or some such.  What I would prefer is
perl -MCPAN -e shell  and  install Bundle::Debian.  I'm **almost**
tempted to bite on that too ;^)

> > > That's good enough for me.  I have boatloads of respect for Andy and his
> > > understanding of Perl Install issues.  That's how it will be for
> > > 5.005.02-3.

For better or worse, if the perl developers chose this new structure
for 5.005, then the decision to deviate incurs a huge cost, 
particularly when one considers the snowball effect and the 
costs of realignment in the future.

Best,

cfm

-- 

Christopher F. Miller, Publisher[EMAIL PROTECTED]
MaineStreet Communications, Inc208 Portland Road, Gray, ME  04039
1.207.657.5078  (MTRF 3-5pm)http://www.maine.com/

Re: LICENSES [was: Re: Have you seen this?]

[Craig Sanders]

On Mon, 12 Oct 1998, Alan Cox wrote:

> > If I use libc, I don't think I am creating a libc. Unless I
> > am, I'm not deriving, I think. If I use libc, I simply use the
> > services. Hence, libc is "a section of" the thing I am making, and
> > does not derive from it.
>
> Your program derives from libc by being linked with it. This is
> precisely why an LGPL has to exist.

true. more precisely: when you compile your program, the binary is a
combined work which is derived from both your source code and libc. that
derived work may only be distributed if ALL of it's parts (i.e. your
source AND the libc) may be distributed under the terms of the GPL.

note that there is also an exemption for libraries which normally come
with the operating system - and libc definitely qualifies there...but
that is a specific exemption which doesn't affect the general rule
above.

libc is a potentially confusing example, so s/libc/libFOO/ in my first
paragraph above.

craig

--
craig sanders

Re: LICENSES [was: Re: Have you seen this?]

[jim]

> > If I use libc, I don't think I am creating a libc. Unless I am, I'm not
> > deriving, I think. If I use libc, I simply use the services. Hence, libc
> > is "a section of" the thing I am making, and does not derive from it.
> 
> Your program derives from libc by being linked with it. This is precisely
> why an LGPL has to exist. 

So this isn't "derivative" in the sense of the OOP idiom "IS-A"...

and you're saying that all I have to do to "derive from" something, is to
include it unmodified or modified?

-Jim

Ropes in stl (was Re: lack of wstring in libstdc++2.8-dev)

[Chris]

On Sun, Oct 11, 1998 at 01:42:20PM +, Rob Browning wrote:
> Chris <[EMAIL PROTECTED]> writes:
> 
> > See http://www.sgi.com/Technology/STL for the current STL implementation.
> 
> And see 
> 
>   http://www.sgi.com/Technology/STL/string_discussion.html
> 
> for why you should probably be using ropes or vector instead.
> 
> We tend to use
> 
>   typedef rope string;
> 
> now.  The code changes to accomodate this were pretty minor.
> 

InterestingI cant seem to use ropes with libc++2.9-dev

I tried this:

#include // Also tried #include 

void main(void)
{
rope r;
}

But I get these errors:

In file included from /usr/include/g++-2/stl_rope.h:2107,
 from /usr/include/g++-2/rope.h:18,
 from blah.cc:1:
/usr/include/g++-2/ropeimpl.h:1085: warning: decimal integer constant is so
large that it is unsigned


The relevant section in ropeimpl.h contains numbers from the fibonacci
sequence up to 2^32(Anyone know why these are relevant to ropes?)


Anyone else have this problem?  Know a solution?


Thanks,


Chris



-- 

--
REALITY.SYS corrupted: Reboot universe? (Y/N/Q)   Debian GNU/Linux
--
Reply with subject 'request key' for PGP public key.  KeyID 0xA9E087D5


pgpVJtwLvC49O.pgp
Description: PGP signature

Re: Ropes in stl (was Re: lack of wstring in libstdc++2.8-dev)

[Chris]

Sorry to follow myself up so quickly, but I found the fault.

The ropeimpl.h file is in error complared to the version from SGI's STL.

Here's the diff:

--- ropeimpl.h.orig Mon Oct 12 14:18:53 1998
+++ ropeimpl.h  Mon Oct 12 14:17:25 1998
@@ -1082,7 +1082,7 @@
 /* 35 */24157817, /* 36 */39088169, /* 37 */63245986, /* 38 */102334155,
 /* 39 */165580141, /* 40 */267914296, /* 41 */433494437,
 /* 42 */701408733, /* 43 */1134903170, /* 44 */1836311903,
-/* 45 */2971215073 };
+/* 45 */2971215073u };
 // These are Fibonacci numbers < 2**32.
 
 template 



I'll post a bug against libc++2.9-dev


chris








On Mon, Oct 12, 1998 at 02:14:30PM +1000, Chris wrote:
> InterestingI cant seem to use ropes with libc++2.9-dev
> 
> I tried this:
> 
> #include // Also tried #include 
> 
> void main(void)
> {
>   rope r;
> }
> 
> But I get these errors:
> 
> In file included from /usr/include/g++-2/stl_rope.h:2107,
>  from /usr/include/g++-2/rope.h:18,
>  from blah.cc:1:
> /usr/include/g++-2/ropeimpl.h:1085: warning: decimal integer constant is so
> large that it is unsigned
> 
> 
> The relevant section in ropeimpl.h contains numbers from the fibonacci
> sequence up to 2^32(Anyone know why these are relevant to ropes?)
> 
> 
> Anyone else have this problem?  Know a solution?
> 
> 
> Thanks,
> 
> 
> Chris
> 
> 
> 
> -- 
> 
> --
> REALITY.SYS corrupted: Reboot universe? (Y/N/Q)   Debian GNU/Linux
> --
> Reply with subject 'request key' for PGP public key.  KeyID 0xA9E087D5



-- 

--
REALITY.SYS corrupted: Reboot universe? (Y/N/Q)   Debian GNU/Linux
--
Reply with subject 'request key' for PGP public key.  KeyID 0xA9E087D5


pgphvQm915KLy.pgp
Description: PGP signature

Re: Intent to package LEIM

[Milan Zamazal]

>Milan Zamazal <[EMAIL PROTECTED]> writes:

>> Is there any technical reason why LEIM (Emacs input methods) is not
>> available as a Debian package?  If not, I'll package it.

>FWIW It's already part of the emacs20 package.

I'd say including LEIM data *.elc files is worth.  I think it's
difficult for an unexperienced Emacs user to FTP and install LEIM
himself.

But maybe only an experienced Emacs user needs inputing through LEIM
instead of system keyboard (which doesn't work in X for Czech anyway:-)
for features like different keyboards in different buffers (which is
important for some languages) and occasional inputting characters of
foreign languages.  I'm not sure.

So do you really think leim package would be totally unnecessary?
Any other opinions?

Thanks.

Milan Zamazal

Re: dbugging dselect

[M.C. Vernon]

On Sun, 11 Oct 1998, David Stern wrote:

>   install attempt in dselect error message:
>   -
>   [..unmet dependencies, blah, blah.. ]
>   dependencies:
> perl-suid: Depends:perl
> libpam0: Depends:libpam0g
> perl: Depends:perl-base

I missed the original message, but what exactly are the dependancy probs?

If it's just those above, why not install them with dpkg and then run
dselect again?

Matthew

-- 
Elen sila lumenn' omentielvo

Steward of the Cambridge Tolkien Society
Selwyn College Computer Support
http://www.geocities.com/Area51/Chamber/8841/
http://www.cam.ac.uk/CambUniv/Societies/tolkien/
http://pick.sel.cam.ac.uk/

Re: Ropes in stl (was Re: lack of wstring in libstdc++2.8-dev)

[M.C. Vernon]

> void main(void)



BTW, main returns int, not void. See the comp.lang.c FAQ for the bit of
the C standard that defines this - main is incorrectly said to return void
in a number of texts though.

HTH,

Matthew

-- 
Elen sila lumenn' omentielvo

Steward of the Cambridge Tolkien Society
Selwyn College Computer Support
http://www.geocities.com/Area51/Chamber/8841/
http://www.cam.ac.uk/CambUniv/Societies/tolkien/
http://pick.sel.cam.ac.uk/

Re: Ropes in stl (was Re: lack of wstring in libstdc++2.8-dev)

[Chris Leishman]

On Mon, Oct 12, 1998 at 08:24:49AM +0100, M.C. Vernon wrote:
> 
> > void main(void)
> 
> 
> 
> BTW, main returns int, not void. See the comp.lang.c FAQ for the bit of
> the C standard that defines this - main is incorrectly said to return void
> in a number of texts though.
> 
> HTH,
> 
> Matthew
> 

Yeah, I know...but when I'm writing little 2second programs to check
something I tend to not fuss with returns, etc, etc...

Chris

-- 

--
REALITY.SYS corrupted: Reboot universe? (Y/N/Q)   Debian GNU/Linux
--
Reply with subject 'request key' for PGP public key.  KeyID 0xA9E087D5

Re: Ropes in stl (was Re: lack of wstring in libstdc++2.8-dev)

[Paul Slootman]

On Mon 12 Oct 1998, Chris Leishman wrote:
> On Mon, Oct 12, 1998 at 08:24:49AM +0100, M.C. Vernon wrote:
> > 
> > > void main(void)
> > 
> > 
> > 
> > BTW, main returns int, not void. See the comp.lang.c FAQ for the bit of
> > the C standard that defines this - main is incorrectly said to return void
> > in a number of texts though.
> 
> Yeah, I know...but when I'm writing little 2second programs to check
> something I tend to not fuss with returns, etc, etc...

Then you should just leave out the "void" in front of "main";
that's less typing and does the right thing :-)


Paul Slootman
-- 
home: [EMAIL PROTECTED] | work: [EMAIL PROTECTED] | debian: [EMAIL PROTECTED]
http://www.wurtel.demon.nl | Murphy Software,   Enschede,   the Netherlands

Re: [conrad@srl.caltech.edu: ANNOUNCE: Fulcrum scientific plotting tool update]

[John Lapeyre]

On Sat, 10 Oct 1998, Martin Schulze wrote:

joey>I wonder if somebody plans to package this one.
joey>

joey>   Fulcrum Scientific Analysis/Plotting Tool for Unix/GTK

I am swamped right now.  But , I'll try to do it if no one else
wants to.  I wonder if this is a new incarnation of yorick, which is
already packaged 
John


John Lapeyre <[EMAIL PROTECTED]>
Tucson,AZ http://www.physics.arizona.edu/~lapeyre

Re: KDE gone, Lyx next ?

[John Lapeyre]

On Sat, 10 Oct 1998, Shaya Potter wrote:

spotte>
spotte>-Original Message-
spotte>From: John Lapeyre <[EMAIL PROTECTED]>
spotte>
spotte>> Lyx is currently in contrib.
spotte>> Lyx is licensed under the GPL (version 2) .  It is dynamically
spotte>>linked against a non-free library (libforms) .
spotte>> According to the GPL and our interpretation of it in the KDE
spotte>>statement, this means we should not be distributing (binaries at least) 
of
spotte>>Lyx. For instance, these binaries use .h files from libforms.
spotte>> Unlike KDE, it may be all original code, so that a single change
spotte>>of license from the developers will do.
spotte>
spotte>
spotte>Boy, Mathias Ehtrich is going to think we have something against him. :)
spotte>
spotte>Shaya

I had no idea he worked on both projects when I wrote that.
Someone just mentioned something about lyx being under the GPL, and I
looked into it.
John


John Lapeyre <[EMAIL PROTECTED]>
Tucson,AZ http://www.physics.arizona.edu/~lapeyre

Re: Perl 5.005.02

[John Lapeyre]

On 11 Oct 1998, Darren/Torin/Who Ever... wrote:

torin>Andy Dougherty, in an immanent manifestation of deity, wrote:
torin>>After some thought, I think I'd recommend that perl5.005_xx retain the
torin>>same directory structure that perl5.00[34]_xx did. (with 5.005 in place 
of
torin>>5.00[34], of course).
torin>
torin>That's good enough for me.  I have boatloads of respect for Andy and his 
torin>understanding of Perl Install issues.  That's how it will be for
torin>5.005.02-3.

Good, I think that will cause the least problems for the rest of
Debian.
John


John Lapeyre <[EMAIL PROTECTED]>
Tucson,AZ http://www.physics.arizona.edu/~lapeyre

Electric VLSI Design System now open source!

[Dan Kegel]

Saw the following on http://slashdot.org.  Sounds like very good news.
- Dan

From: [EMAIL PROTECTED]
Date: 1998/10/10
Newsgroups:  alt.electronics.analog.vlsi, can.vlsi, comp.lsi,
comp.lsi.cad, cu.vlsi 


I am pleased to announce that the "Electric VLSI Design System" is now
available to anyone who wishes to use it.

Free source code and documentation can be found on the Free Software
Foundation web site.  For the latest version, download
ftp://ftp.gnu.org/pub/gnu/electric-5.4g3.tar.gz

Here are some of the amazing features that you will find in Electric:

> Platform independence.  This source release compiles under UNIX,
Windows, and Macintosh.

> Design both Schematic circuits and ASICs (has many IC design rule sets,
including the latest quad-metal MOSIS submicron rules).

> Many built-in tools (DRC, Simulation, Routing, Silicon Compilation, VHDL
interface, Network Consistency Checking, Compaction, Compensation, PLA
Generation, and much more).

> Built-in constraint system provides powerful design assistance.

> Flexible database and tool control make this an ideal workbench for CAD
tool development.

If you want further information, see the GNU web page at:
http://www.gnu.org/software/electric/electric.html
And if you want further information from the folks who wrote Electric,
see
their web page at:
http://www.electriceditor.com

Enjoy!
-Steven Rubin

Intent to package: ada-rm

[Samuel Tardieu]

-BEGIN PGP SIGNED MESSAGE-

ada-rm is a set of documentation containing the Ada 95 reference
manual, the Rationale and the changes with Ada 83.

  Sam
- -- 
Samuel Tardieu -- [EMAIL PROTECTED]

-BEGIN PGP SIGNATURE-
Version: 2.6.3ia
Charset: latin1

iQCVAwUBNiHEu4FdzKExeYBpAQFpOgP9HZQkCHUFN94mpZWtF/GaX7BklJXEZ+xm
wxU58h0UCSsjKd5P0hcQOsAue996UhTPlZc/qrgbeVIkYwmOZ5FADifiTrZXocFK
eGOdshl9I73L5MHe8RbBKf9btkx/NrJ+ZMALXEtaODNAfbyl7gyoBTCU6L2iejDF
b3gETR8cdEM=
=MMZY
-END PGP SIGNATURE-

Re: Intent to package LEIM

[Christophe Broult]

Milan Zamazal <[EMAIL PROTECTED]> writes:

> >Milan Zamazal <[EMAIL PROTECTED]> writes:
> 
> >> Is there any technical reason why LEIM (Emacs input methods) is not
> >> available as a Debian package?  If not, I'll package it.
> 
> >FWIW It's already part of the emacs20 package.

Something is missing because when I try the menu item `Mule/Toggle
Input Method' (or C-\ for short) I get the following error:

Signaling: (error "Can't use the Quail package `latin-1-prefix'")
  signal(error ("Can't use the Quail package `latin-1-prefix'"))
  error("Can't use the Quail package `%s'" "latin-1-prefix")
  quail-use-package("latin-1-prefix" "quail/latin-pre")
  apply(quail-use-package "latin-1-prefix" "quail/latin-pre")
  activate-input-method("latin-1-prefix")
  toggle-input-method(nil)
* call-interactively(toggle-input-method)

> 
> I'd say including LEIM data *.elc files is worth.  I think it's
> difficult for an unexperienced Emacs user to FTP and install LEIM
> himself.

I was used to compile Emacs 20.xx by myself and install the LEIM
package as well but I'd rather use a standard package. So I really
think that the missing files should be added.

> 
> But maybe only an experienced Emacs user needs inputing through LEIM
> instead of system keyboard (which doesn't work in X for Czech anyway:-)
> for features like different keyboards in different buffers (which is
> important for some languages) and occasional inputting characters of
> foreign languages.  I'm not sure.
> 
> So do you really think leim package would be totally unnecessary?

I don't think so.

> Any other opinions?

Could you please package LEIM? 

Anyway thank you for your packaging effort.

Chris

-- 
Looking for a cutting edge   | Christophe Broult
software validation technology?  | 
Check http://www.info.unicaen.fr/lpv | ``Smile, chuckle, giggle''

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Gergely Madarasz]

On Sun, 11 Oct 1998, Joseph Carter wrote:

> On Sun, Oct 11, 1998 at 10:52:19PM +0200, Gergely Madarasz wrote:
> > > [...]
> > > I agree that by using XForms in development, and XForms *is* needed to
> > > compile and run LyX, we have implicitly allowd all users to link Lyx
> > > with XForms.
> > > [...]
> > 
> > I don't think so. It is not enough for KDE, why should it be enough for
> > LyX ?
> 
> It's not enough for KDE because KDE includes things not written by the KDE
> people.

How can we be sure that LyX does not include things not written by them?
And anyway we're not given permission to distribute it.

--
Madarasz Gergely   [EMAIL PROTECTED] [EMAIL PROTECTED]
  It's practically impossible to look at a penguin and feel angry.
  Egy pingvinre gyakorlatilag lehetetlen haragosan nezni.
HuLUG: http://mlf.linux.rulez.org/

GnomeCar

[Michael Meskes]

Anyone working on this? I just saw the a screenshot that looks nice. But
since its source is only in cvs I cannot simply try it.

Michael
-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

[ettrich@troll.no: Re: copyright problem]

[Michael Meskes]

How about this one?

I told him I would remove the first sentence but other than that it looks
okay to me.

Michael

- Forwarded message from Matthias Ettrich <[EMAIL PROTECTED]> -
If we do something like this, I'd rather suggest a text like:

  The GPL is often a source of missunderstanding and confusion. As we
  understand the license, redistribution and use of LyX in source and
  binary forms, with or without modification, are permitted without any
  additional conditions. Even more, we would explicitely like to encourage
  people to distribute LyX in both source and binary forms. This permission
  certainly includes linking against GUI toolkits like XForms, Motif, GTK, Qt
  or Win32.


If that is still ok for Debian, I could live with it. Michael?

- End forwarded message -

-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Michael Meskes]

On Sun, Oct 11, 1998 at 04:07:31PM +, Raja R Harinath wrote:
> > I agree that by using XForms in development, and XForms *is* needed to
> > compile and run LyX, we have implicitly allowd all users to link Lyx
> > with XForms.
> 
> I don't see how it follows.  "we have implicitly allowed all users to
> link LyX with XForms" does not imply "we have implicitly allowed
> (re)distribution of the resulting LyX binaries", which I guess is the
> issue at hand.

I'm sorry, but for me this sounds like like nitpicking. But I try to solve
this. 

Boy, I wonder how many problemes with licenses we will find if we examine
all packages to that detail.

Michael
-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

Re: KDE gone, Lyx next ?

[Michael Meskes]

On Sun, Oct 11, 1998 at 12:09:15PM +1000, Hamish Moffatt wrote:
> There's probably plenty of other problematic packages in contrib too,
> as Raul has been telling us for a while. www-mysql, for example,
> although it'll move in to main once I reupload it (since mysql-base
> is in main now).

I idn't notice that mysql-base made it into main. I take it you need more
than base to set up a database, don't you?

Michael

-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Michael Meskes]

On Sun, Oct 11, 1998 at 11:18:53PM +0200, Bart Schuller wrote:
> Because *implicit* permission isn't good enough. By default *nothing* is
> allowed. So every right the authors grant you had better be written down
> in a license accompanying the software, otherwise one of the authors (or
> sometimes even their employers) can later sue you.

But the default does not hold. After all they said the GPL the code.

> In this particular case it is important to be explicit about the extra
> permissions granted, because people might get the mistaken belief that
> it is thus also ok to import other GPLed code into the project.

I try talking them into making it explicit.

Michael
-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

Re: KDE hurts Qt (LICENSES)

[Martin Konold]

On Sun, 11 Oct 1998, Alan Cox wrote:

Dear Alan,

> What you have to remember is this.
> 
>   SuSE are currently too small to be worth sueing in the US
>   on a licensing issue. Ditto most (all ?) other current
>   distributors
> 
> I'm not trying to belittle people like SuSE quite the opposite - Im
> confident SuSE will end up big enough to be worth sueing in the USA on
> such matters. What happens then ?

Alan: This is a perfect example of FUD!

>From ESR`s hackers dictionary:
":FUD: /fuhd/ n. Defined by Gene Amdahl after he left IBM to found
   his own company: "FUD is the fear, uncertainty, and doubt that IBM
   sales people instill in the minds of potential customers who might
   be considering [Amdahl] products."  The idea, of course, was to
   persuade them to go with safe IBM gear rather than with
   competitors' equipment.  This was traditionally done by promising
   that Good Things would happen to people who stuck with IBM, but
   Dark Shadows loomed over the future of competitors' equipment or
   software.  See {IBM}.

:FUD wars: /fuhd worz/ n. [from {FUD}] Political posturing engaged in
   by hardware and software vendors ostensibly committed to
   standardization but actually willing to fragment the market to
   protect their own shares.  The UNIX International vs. OSF conflict
   is but one outstanding example."


Facts: Unix does have most of its acceptance in Europe in contrast to the 
   US.
   SuSE is the most popular Linux distributor in Europe.
   SuSE has the biggest rate of growth of all Linux distributors in
   the US.
 


-- martin

// Martin Konold, Herrenbergerstr. 14, 72070 Tuebingen, Germany  //
// Email: [EMAIL PROTECTED] //
Anybody who's comfortable using KDE should use it. Anyone who wants to
tell other people what they should be using can go to work for Microsoft.
[EMAIL PROTECTED]: "BCPL gave birth to B, and the child of B was of 
   course C, since the ancestor of X is W, so the 
   sucessor to X must be K."

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Martin Schulze]

Michael Meskes wrote:
> On Sun, Oct 11, 1998 at 04:07:31PM +, Raja R Harinath wrote:
> > > I agree that by using XForms in development, and XForms *is* needed to
> > > compile and run LyX, we have implicitly allowd all users to link Lyx
> > > with XForms.
> > 
> > I don't see how it follows.  "we have implicitly allowed all users to
> > link LyX with XForms" does not imply "we have implicitly allowed
> > (re)distribution of the resulting LyX binaries", which I guess is the
> > issue at hand.
> 
> I'm sorry, but for me this sounds like like nitpicking. But I try to solve
> this. 
> 
> Boy, I wonder how many problemes with licenses we will find if we examine
> all packages to that detail.

A lot.

Regards,

Joey

-- 
Linux - the choice of a GNU generation

Re: [ettrich@troll.no: Re: copyright problem]

[Craig Sanders]

On Mon, 12 Oct 1998, Michael Meskes wrote:

> How about this one?
> 
> I told him I would remove the first sentence but other than that it looks
> okay to me.

looks good to me, with or without the first sentence.  

it's true, anyway.  the GPL is often a source of misunderstanding and
confusion.  witness KDE, for example.

if ettrich is willing to write this for LyX, then maybe he'll do the same
for KDE?  i hope so.

> Michael
> 
> - Forwarded message from Matthias Ettrich <[EMAIL PROTECTED]> -
> If we do something like this, I'd rather suggest a text like:
> 
>   The GPL is often a source of missunderstanding and confusion. As we
>   understand the license, redistribution and use of LyX in source and
>   binary forms, with or without modification, are permitted without any
>   additional conditions. Even more, we would explicitely like to encourage
>   people to distribute LyX in both source and binary forms. This permission
>   certainly includes linking against GUI toolkits like XForms, Motif, GTK, Qt
>   or Win32.
> 
> 
> If that is still ok for Debian, I could live with it. Michael?
> 
> - End forwarded message -

craig

--
craig sanders

Intend to package, create OSS/Free

[Guenter Geiger]

Hi there !

The current way to add sound support on Debian is either installing the 
kernel-source and compiling OSS
or installing the ALSA packages.

What about providing a modularized, precompiled OSS package and a sound 
installation tool ?

This sound installation tool should use isapnp to detect soundcards, or let the 
user specifiy the hardware settings and
the soundcard he wishes to use. The tool should handle both, ALSA and OSS, 
setup the loading of modules and 
the isapnp.conf file.

URLS:
 


Please CC replies to me

Guenter

RE: lesstif

[Guenter Geiger]

Same for me, snd doesn´t work with current lesstif  version

Guenter

Re: LICENSES [was: Re: Have you seen this?]

[Raul Miller]

Craig Sanders <[EMAIL PROTECTED]> wrote:
> note that there is also an exemption for libraries which normally come
> with the operating system - and libc definitely qualifies there...

Nope.

Some of the time, libc would qualify for that special excemption.
But it doesn't qualify for anything shipped with the OS (which includes,
at a minimum, the kernel and libc).

-- 
Raul

Re: PROPOSAL: one debian list for all porting efforts

[Hartmut Koptein]

> > to increase communication betweenm the ports and between porters and
> > non-porters, I'd propose a new list:
> > 
> > debian-porting
> > or sim.
> 
> I fully support this proposal (The name debian-porting seems fine to me)

No, we haven't enough topics for this new list.

> IMHO, it makes sence to create a new list, since it seems 90% of the
> Debian developers use i386 only...

:-)   debian/i386 is also a port!

MfG,

Hartmut



-- 
 Hartmut Koptein   EMail:
 Friedrich-van-Senden-Str. 7   [EMAIL PROTECTED]
 26603 Aurich   
 Tel.: +49-4941-10390  [EMAIL PROTECTED]

Re: dpkg config files in /etc ?

[Ilya L Ovchinnikov]

On Thu, Oct 08, 1998 at 12:47:09PM -0400, Stephen J. Carpenter wrote:
> On Thu, Oct 08, 1998 at 11:48:08AM -0400, Peter S Galbraith wrote:
> > 
> > "Thomas Gebhardt" <[EMAIL PROTECTED]> writes:
> > 
> > > the configuration files of all debian packages are located in /etc.
> > > That's really fine.
> > >
> > > But the package manager stores its configuration (access method,
> > > list of selected packages, ...) somewhere in /var/lib/dpkg. Why?
> > 
> > Steve Dunham wrote:
> >  
> > > Configuration goes in /etc, state goes in /var.
> > 
> > But the access method is not a state, it's a configuration.
> 
> I would disagree...
> In my mind (read; I dn't have a formal definition in front of me)
> "State" and "Configuration" have somewhat overlapping definitions.
> 
> My general rule of thinking about it is:
> state is an opion within the program which can be changed and
> should be remembered next time.
> esp something which reasonably could change every time the program
> is used (it is concievable I have a CD today...in a month I am FTP
> upgrading)
> 
> this is not something which is meant to be changed "by hand"
> besides...
> the main rational for /var is to allow other partitions to be mounted
> read only...
> if this were stored in /etc/etc would HAVE to be mounted read-write

/etc CAN NOT be mounted, in should be part of root filesystem.
And root filesystem is mounted read-write.  
And /etc/mtab  definitly is a state but it should be in root partition.

But /var/lib/dpkg/info changes only during upgrading the system.
/usr must be remounted read-write for that.  So maybe it should go
to /usr ?  And in any case  dpkg should check if /usr is mounted read-write
before unpacking a package.

Re: LICENSES [was: Re: Have you seen this?]

[Raul Miller]

Alan Cox <[EMAIL PROTECTED]> wrote:
> > KDE requires Qt currently. So KDE is non free. 

[EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> _No_. This does not necessarily follow, even if both statements may
> both be true. KDE simply depends on something that is non-free.

Except that KDE programs have been written (or modified) to require
Qt, to not work without Qt, and everybody that uses it lives with this
design aspect.

-- 
Raul

Re: Electric VLSI Design System now open source!

[Hamish Moffatt]

On Sun, Oct 11, 1998 at 09:27:07AM -0700, Dan Kegel wrote:
> I am pleased to announce that the "Electric VLSI Design System" is now
> available to anyone who wishes to use it.

Unless someone has beaten me to it I will attempt to package it asap
(which could be a couple of weeks). Sounds very good although I haven't
yet had a chance to try it out.

Hamish
-- 
Hamish Moffatt VK3TYD  [EMAIL PROTECTED], [EMAIL PROTECTED]
Latest Debian packages at ftp://ftp.rising.com.au/pub/hamish. PGP#EFA6B9D5
CCs of replies from mailing lists are welcome.   http://hamish.home.ml.org

Re: KDE gone, Lyx next ?

[Hamish Moffatt]

On Mon, Oct 12, 1998 at 09:13:44AM +0200, Michael Meskes wrote:
> On Sun, Oct 11, 1998 at 12:09:15PM +1000, Hamish Moffatt wrote:
> > There's probably plenty of other problematic packages in contrib too,
> > as Raul has been telling us for a while. www-mysql, for example,
> > although it'll move in to main once I reupload it (since mysql-base
> > is in main now).
> 
> I idn't notice that mysql-base made it into main. I take it you need more
> than base to set up a database, don't you?

The server remains non-DFSG-free, although for most users it is gratis.
The client code is all in main.


Hamish
-- 
Hamish Moffatt VK3TYD  [EMAIL PROTECTED], [EMAIL PROTECTED]
Latest Debian packages at ftp://ftp.rising.com.au/pub/hamish. PGP#EFA6B9D5
CCs of replies from mailing lists are welcome.   http://hamish.home.ml.org

Re: Intend to package, create OSS/Free

[Hamish Moffatt]

On Mon, Oct 12, 1998 at 01:51:24PM -0700, Guenter Geiger wrote:
> The current way to add sound support on Debian is either installing the 
> kernel-source and compiling OSS
> or installing the ALSA packages.
> 
> What about providing a modularized, precompiled OSS package and a sound 
> installation tool ?

Sounds great. Hopefully we can leave out OSS/"Free" though.
I haven't looked at ALSA though.

Hamish
-- 
Hamish Moffatt VK3TYD  [EMAIL PROTECTED], [EMAIL PROTECTED]
Latest Debian packages at ftp://ftp.rising.com.au/pub/hamish. PGP#EFA6B9D5
CCs of replies from mailing lists are welcome.   http://hamish.home.ml.org

GPL source policy.

[Masato Taruishi]

Hi,

There is a section from GPL:

b) Accompany it with a written offer, valid for at least three
years, to give any third party, for a charge no more than your
cost of physically performing source distribution, a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange; or,


In this, the licence sait that `valid for at least three years', but I can't
understand what date the beginnings of `at least three years' starts
actually from. From the date when the vendor began to sell it, or when
buyers bought it, or else?
 

Masato Taruishi <[EMAIL PROTECTED]> | University of Electro Comunications
<[EMAIL PROTECTED]>  |   Department of Computer Science
<[EMAIL PROTECTED]> |  Junior
http://www.sunicom.co.jp/~taruisma/  |   Chofu City Tokyo, JAPAN  
   Key fingerprint = 49 46 74 E1 8D D1 EB 56  8D CA 2A 20 14 9E A9 25

Re: KDE gone, Lyx next ?

[Raul Miller]

Craig Sanders <[EMAIL PROTECTED]> wrote:
> there is no combined work until the source is compiled, linked to the
> non-free library, and a binary produced.

Please show me where the GPL says this.

I'm tired of pointing out this is false, quoting from the GPL to show
you were it says different, and having you ignore that.

-- 
Raul

Re: I2O specs mailed to webmaster

[Gregory S. Stark]

Craig Sanders <[EMAIL PROTECTED]> writes:

> it was mailed from a dummy hotmail account ([EMAIL PROTECTED]), and the
> originating IP was from an ISP in Norway.

On the off chance that the original sender is reading this, or looking at the
e-mail archive: Hotmail is not an anonymous mailing system, and makes no
pretense of such. They will happily hand over records if needed.

If you want to send something anonymously I suggest using the cypherpunk PGP
remailers. Of course there's no reason not to combine that with sending from
hotmail or something like that, but don't count on a system like hotmail for
your anonymity.

greg

Re: sendmail & libc6 [was: Squid2, how to handle incompatible upgrade]

[peloy]

Miquel van Smoorenburg <[EMAIL PROTECTED]> wrote:
> In article <[EMAIL PROTECTED]>,
>  <[EMAIL PROTECTED]> wrote:
>>The only pain I had to face was that I had to upgrade my libc6 and
>>that upgrade broke sendmail, so I had to upgrade sendmail as well.
> 
> Uh - oh .. please check out this bug:
> 
> #27334: libc6: breaks sendmail, probably problem in resolver
> 
> details at http://www.debian.org/Bugs/db/27/27334.html
> 
> This is release-critical, IMO.

I agree this is release critical. However, the problem went away when
I upgraded to the sendmail in Slink (8.9.1, I think). Before that, I
was using the sendmail in Hamm.

I did not downgraded my libc6.

peloy.-

Re: copyright problem

Okay, Matthias also agrees with my version. Let´s see if the other LyX guys
say. Could we get this into the LyX package ASAP? Who´s in charge now?
Mark?

Michael
-- Weitergeleitet von Mummert&Partner
MeskesM/D/ExternalStaff/WLB on 12.10.98 16:02 ---


Matthias Ettrich <[EMAIL PROTECTED]> on 12.10.98 12:54:20

Bitte antworten an lyx@via.ecp.fr

An:   lyx@via.ecp.fr
Kopie:Michael Meskes <[EMAIL PROTECTED]> (Blindkopie: Mummert&Partner
  MeskesM/D/ExternalStaff/WLB)
Thema:Re: copyright problem




>>
>> If that is still ok for Debian, I could live with it. Michael?
>
>I could also. However, I would use a slightly different wording. I would
do
>it this way:
>
>   As we understand the license of LyX, redistribution and use of LyX in
>   source and binary forms, with or without modification, are permitted
>   without any additional conditions. Even more, we would explicitely like
>   to encourage people to distribute LyX in both source and binary forms.
>   This permission certainly includes linking against GUI toolkits like
>   XForms, Motif, GTK, Qt or Win32.
>
>Is that okay, too?
It's still fine with me.
Matthias






  Mit freundlichen Grüßen


  Dr. Michael Meskes, Senior-Consultant
  Mummert+Partner Unternehmensberatung AG
  Tel.: +49211 826  4616

Antwort: Re: [ettrich@troll.no: Re: copyright problem]

>looks good to me, with or without the first sentence.

For me too.
>it's true, anyway.  the GPL is often a source of misunderstanding and
>confusion.  witness KDE, for example.

Yes, your right. But I think this sentence doens´t fit well into a license
file.

>if ettrich is willing to write this for LyX, then maybe he'll do the same
>for KDE?  i hope so.
I´m in touch with him on that. But he doesn´t like to have his packages on
Debian while his friends´ packages are not. He´s afraid of a Debian KDE
package being not even close to the real thing.

Michael



Dr. Michael Meskes, Senior-Consultant
Mummert+Partner Unternehmensberatung AG
Tel.: +49211 826  4616

September LJ

[Federico Di Gregorio]

Hi all and sorry for the "SPAM"... but I just read my Spetember issue
of the Linux Journal and I noted that on page 6 the Editor says:

This is one reason we choose the Debian distribution to
use in our office.   ^^
 that's US!
 
Whoa! RH take that! eh eh eh
(Sorry for the out-of-topic-spammish-style again...)
Federico

Re: I2O specs mailed to webmaster

[Thomas Lakofski]

On 12 Oct 1998, Gregory S. Stark wrote:

> On the off chance that the original sender is reading this, or looking at the
> e-mail archive: Hotmail is not an anonymous mailing system, and makes no
> pretense of such. They will happily hand over records if needed.

Equally the information you supply to hotmail can be complete garbage and
you can access their servers via an anonymizing proxy.

-thomas

Local IP address / Java Incompatibility

[Rainer Dorsch]

We found an incompatibility between Java on Solaris 2.5.1/2.6 and on Debian 
2.0.r2 Could anybody suggest how to find out, if it is a bug in Debian Linux 
or JDK?

Additions to the included bug report:

- $ hostname -i
  129.69.183.3

  gives the correct answer in a shell.

- Debian 2.0 comes with jdk-1.1.5v5-1


Thanks.



--- Begin Message ---
Hello!

I have tried to get the IP address and host name of the host running my 
Java application using the following Java-Code:

  InetAddress inetadr = InetAddress.getLocalHost();
  System.out.println("LocalIP:   " + inetadr.getHostAddress());
  System.out.println("LocalName: " + inetadr.getHostName());

This works well for a SUN system, but I always get 127.0.0.1 and
localhost on my Debian Linux system. Is this a bug in Linux-Java or
is there any other way to determine the name and IP address of my
localhost???

Can anyone help me?

bye Thomas!


Below are some of my System.Properties:

-- listing properties --
user.language=en
java.home=/usr/lib/jdk1.1/bin/..
java.vendor.url.bug=http://java.sun.com/cgi-bin/bugreport...
file.encoding.pkg=sun.io
java.version=cls:03/11/11-08:49
file.separator=/
line.separator=
 
file.encoding=8859_1
java.vendor=Sun Microsystems Inc., ported by Rand...
user.timezone=CET
user.name=schwarts
os.arch=x86
os.name=Linux
java.vendor.url=http://java.blackdown.org/java-linux
user.dir=/home/hiwi/schwarts/RAJava
java.class.path=.:/usr/lib/jdk1.1/bin/../classes:/usr...
java.class.version=45.3
os.version=2.0.33
path.separator=:
user.home=/home/hiwi/schwarts

--- End Message ---
-- 
Rainer Dorsch
Abt. Rechnerarchitektur  e-mail:[EMAIL PROTECTED]
Uni StuttgartTel.: 0711-7816-215

Re: KDE hurts Qt (LICENSES)

[Alan Cox]

> Alan: This is a perfect example of FUD!

No

>SuSE has the biggest rate of growth of all Linux distributors in
>the US.

And SuSE and Red Hat and all of them put together are not worth a US lawsuit
yet. Price yourself a US lawsuit then judge again.

Make them 5 times bigger and yes then its worth it.

Alan

LyX & KDE

I might be able to get a similar license agreement for KDE as the one I
send for LyX. Would that be enough to get at least major parts of KDE back
on the site? I have no idea how much we would have to keep out. I know
kghostview and kdvi, but other than that? Since I use Gnome I cannot simply
check. No KDE on my machine. :-)

Michael
--
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein
Fire!
Mummert+Partner | private: [EMAIL PROTECTED] | Use Debian
Unternehmensberatung AG |  [EMAIL PROTECTED] | GNU/Linux!

Re: Local IP address / Java Incompatibility

[Rainer Dorsch]

Yes, it works now! Thanks.

Could anybody think of negative implications of doing this reverse ordering of 
the localhost and ip address entries?

> Rainer Dorsch wrote:
> > 
> > We found an incompatibility between Java on Solaris 2.5.1/2.6 and on Debian
> > 2.0.r2 Could anybody suggest how to find out, if it is a bug in Debian Linux
> > or JDK?
> > 
> Put the hostname above the localhost entry
> in the /etc/hosts.
> 
> I believe it's a libc/glibc fault.
> 
> Joe
> 
> -- 
> Joe Carter  Software Engineer
> Brite Voice Systems Ltd, Gatley, Cheshire. UK.
> mailto:[EMAIL PROTECTED]
> 

-- 
Rainer Dorsch
Abt. Rechnerarchitektur  e-mail:[EMAIL PROTECTED]
Uni StuttgartTel.: 0711-7816-215

Re: Intend to package, create OSS/Free

[Manoj Srivastava]

Hi,
>>"Guenter" == Guenter Geiger <[EMAIL PROTECTED]> writes:

 Guenter> What about providing a modularized, precompiled OSS package
 Guenter> and a sound installation tool ? 

Please also consider a src.deb package (look at pcmcia_cs
 packages for example) that puts the sources in /usr/src/modules// 
 so that the sound module can be built when the kernel packages are
 created by the user using kernel-package.

This would then enable people using the sound package to
 recompile if they happen to get a new kernel.

manoj
-- 
 It was a saying of the ancients, "Truth lies in a well;" and to carry
 on this metaphor, we may justly say that logic does supply us with
 steps, whereby we may go down to reach the water.  -- Dr. I. Watts
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E

Re: KDE hurts Qt (LICENSES)

[Martin Konold]

On Mon, 12 Oct 1998, Alan Cox wrote:

> > Alan: This is a perfect example of FUD!
> >SuSE has the biggest rate of growth of all Linux distributors in
> >the US.
> 
> And SuSE and Red Hat and all of them put together are not worth a US lawsuit
> yet. Price yourself a US lawsuit then judge again.
> 
> Make them 5 times bigger and yes then its worth it.

Wrong! If you are not protecting your rights. You are running into the
danger that you loose a case because you did accept the wrong behaviour
too long in case you had detailed information years before.

Just threadening with sueing is simply an action of FUD.

I therefore make the following _personal_ offer to you and RMS:

I will link GNU Emacs to kde and distribute the resulting source and
binary to you and RMS. 

You then have to sue ME or shut up!

Sorry for my harsh words. But it looks to me like some people are trying
to keep kde people from making even better free software because they do
have trouble to succeed with their own competing project.

Regards,
-- martin

// Martin Konold, Herrenbergerstr. 14, 72070 Tuebingen, Germany  //
// Email: [EMAIL PROTECTED] //
Anybody who's comfortable using KDE should use it. Anyone who wants to
tell other people what they should be using can go to work for Microsoft.
[EMAIL PROTECTED]: "BCPL gave birth to B, and the child of B was of 
   course C, since the ancestor of X is W, so the 
   sucessor to X must be K."

Re: sendmail & libc6 [was: Squid2, how to handle incompatible upgrade]

[Miquel van Smoorenburg]

In article <[EMAIL PROTECTED]>,
 <[EMAIL PROTECTED]> wrote:
>Miquel van Smoorenburg <[EMAIL PROTECTED]> wrote:
>> #27334: libc6: breaks sendmail, probably problem in resolver
>> details at http://www.debian.org/Bugs/db/27/27334.html
>> This is release-critical, IMO.
>
>I agree this is release critical. However, the problem went away when
>I upgraded to the sendmail in Slink (8.9.1, I think). Before that, I
>was using the sendmail in Hamm.
>I did not downgraded my libc6.

I had this problem with sendmail-8.9.1-x and libc6_2.0.7u-2. Only
downgrading to libc6_2.0.7t-1 helped.

Mike.
-- 
  "Did I ever tell you about the illusion of free will?"
-- Sheriff Lucas Buck, ultimate BOFH.

Re: KDE hurts Qt (LICENSES)

[Alan Cox]

> Just threadening with sueing is simply an action of FUD.

I haven't threatened to sue anyone. You must have been listening to 
Matthais foaming at the mouth too much.

> Sorry for my harsh words. But it looks to me like some people are trying
> to keep kde people from making even better free software because they do
> have trouble to succeed with their own competing project.

I hope not. Had people like Miguel been involved that might be a reasonable
suspicion. I want to see a KDE that works well, is on a solid legal ground
and preferably is totally free. KDE works pretty well, its quite usable
even on my 486SLC palmtop.[1] Most of number 2 is easy to resolve by
folks putting in the explicit clarifications that they think linking with
Qt is fine by them. The only hard bit is asking people outside the KDE
project and dealing with anyone who doesn't like the idea.

I suspect the number of objections will be few.

Alan

[1] Yes I've got a box with KDE on it, and one with Gnome and one with
Windowmaker ;)

updates to Debian pages

[James A. Treacy]

This is just to help translators keep up to date with changes to the
Debian pages.

 - distrib/distrib.wml was renamed to distrib/index.wml
   This caused changes to template/debian/menubar.wml
  and template/debian/navbar.wml

   I believe those are the only links affected, but I'm about to run a 
urlchecker
   over the pages to see if anything got missed.

 - news.wml and news1997.wml have been removed. This was actually changed a 
while
   ago. All the news is now in directories by year under News/. The dependencies
   in the News/Makefile may not be totally correct so please report any 
problems.

   It is intended that this setup enable translators to translate only those
   announcements that they feel are important enough or they have time for.
   It will also enable language specific announcements to be made (for example
   a meeting of Spanish speaking Debian people in Madrid).
   If an item is translated, then its title will replace the english one in the
   list of news items in that languages news page. Since this hasn't been 
thoroughly
   tested I wouldn't mind a guinea pig to translate something to help test it 
out.

 - Pics/ has been added to CVS. This enables wml to find the images used in the 
web
   pages so it can add width and height tags. Of course this should have been 
done
   ages ago. So fire me. :)

   A link has been added from all the /Pics -> ../english/Pics on master.
   They weren't added to CVS as it doesn't handle special files very well.
   Its only important that master have them anyway, so the pages the public sees
   have the tags.

Jay Treacy

Re: KDE hurts Qt (LICENSES)

[Raul Miller]

On Mon, 12 Oct 1998, Alan Cox wrote:
> > And SuSE and Red Hat and all of them put together are not worth a US lawsuit
> > yet. Price yourself a US lawsuit then judge again.
> > 
> > Make them 5 times bigger and yes then its worth it.

Martin Konold <[EMAIL PROTECTED]> wrote:
> Wrong! If you are not protecting your rights. You are running into
> the danger that you loose a case because you did accept the wrong
> behaviour too long in case you had detailed information years before.

You're confusing copyright law and trademark law.  This statement
would be true for trademark law but is not true of copyright law.

-- 
Raul

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Russell Coker]

>> > > I agree that by using XForms in development, and XForms *is* needed to
>> > > compile and run LyX, we have implicitly allowd all users to link Lyx
>> > > with XForms.
>> > > [...]
>> > 
>> > I don't think so. It is not enough for KDE, why should it be enough for
>> > LyX ?
>> 
>> It's not enough for KDE because KDE includes things not written by the KDE
>> people.
>
>How can we be sure that LyX does not include things not written by them?
>And anyway we're not given permission to distribute it.

Surely if a piece of software is released under a particular license
agreement and if you contribute some source code to the maintainer of the
software for inclusion in that software then the same license conditions must
implicitely apply.  If a piece of software is GPL then you can't give some
source code to the maintainer and then say "lines 10-20 of file foo.c are
commercial and al users must pay me".  If things were otherwise then all
current GPL projects would be void and all new ones would require written
statements explicitely agreeing to the license conditions.
Changing a license from one that implies something (may be linked to Xforms)
to one which states it directly and clearly is not altering the license
conditions merely clarifying them.  So my opinion is that the main developers
in the LyX project can get together and change the license in this fashion
after a quick vote without any problems.
Of course I'm not a lawyer and even a lawyers opinion won't mean that much
unless a magistrate agrees...

--
Got no future, got no past.
Here today, built to last.

Re: A Detailed Analysis of the GPL For KDE/QT

[Martin Konold]

On Mon, 12 Oct 1998, Raul Miller wrote:

> Finally, if a library wasn't a part of the program as a whole, what's
> the point of the LGPL?

You seem to mix things up again.

Programs linked to GPL'd library must be GPL, because by using the GPL'd
library you have to comply to the license terms of this library. The main
point is that USING a GPL'd library for a program is only allowed if the
resulting program becomes GPL'd. The library never becomes part of the
program which is very obvious if you consider the fact the the GPL does
not mention linking at all.

Programs linked to LGPL'd library _can_ be non GPL. LGPL does protect the
code of the library very similiar to the GPL but allows link (static and
dynamic) to non GPL and non LGPL'd code.

Programs linked to non GPL'd library can be GPL.

The main point is that a library can force some restrictions on a program 
which uses the library (Like the Qt Free Software License forces to
release the written program under a free license) but NOT vice versa.

So to answer your question:  LGPL was created in order to NOT force the
GPL on programs which use it.  (The license does cover the use of the
library and modifications/redistributions of the library itself.) It was
not created because it is part of the reulting binary.

Regards,
-- martin

// Martin Konold, Herrenbergerstr. 14, 72070 Tuebingen, Germany  //
// Email: [EMAIL PROTECTED] //
Anybody who's comfortable using KDE should use it. Anyone who wants to
tell other people what they should be using can go to work for Microsoft.
[EMAIL PROTECTED]: "BCPL gave birth to B, and the child of B was of 
   course C, since the ancestor of X is W, so the 
   sucessor to X must be K."

Re: exim really does need to be the standard MTA in slink

[Tony Finch]

Paul Slootman <[EMAIL PROTECTED]> wrote:
>
>I personally have confidence in Exim's quality in this regard.
>Demon (a large ISP in the UK and the Netherlands, www.demon.net)
>uses Exim as its customer-facing smtp interface, so I guess that they're
>convinced as well.

To be precise, we use Exim for the smarthosts. Incoming customer mail
and delivery to customers is handled by MMDF; company staff email is
handled by Sendmail. This heterogeneity is mostly for historical
reasons -- Demon's earliest systems were based on SCO Unix which used
MMDF as the standard MTA. The Internal mail server was hacked together
overnight just to win a bet that it would be "easy" to set up a
replacement for the pre-existing system. Exim was introduced recently
as aprt of improvements to the architecture of the customer mail
system.

Tony.
-- 
   7yuc zhd2**f.a.n.finch
[EMAIL PROTECTED]
   [EMAIL PROTECTED]

GNotepad

[Ole J. Tetlie]

Is anyone packing gnotepad?

[ Sorry if anyone tried to a post of mine and bounced. I played with
exim.conf and forgot to "unplay" the rewrite. ]

-- 
Eschew obfuscation(go on; look them both up)
   (Brian White)
[EMAIL PROTECTED]   [-: .elOle. :-]   [EMAIL PROTECTED]

Re: GNotepad

[Martin Schulze]

Ole J. Tetlie wrote:
> Is anyone packing gnotepad?
> 
> [ Sorry if anyone tried to a post of mine and bounced. I played with
> exim.conf and forgot to "unplay" the rewrite. ]

wget + ./configure + vi src/main.c + make just finished.  It looks
nice, it seems to work.  We should include it.  However if I push
the exit button I get a "Gdk segfault" message.

Regards,

Joey

-- 
Linux - the choice of a GNU generation

Re: A Detailed Analysis of the GPL For KDE/QT

[Raul Miller]

Martin Konold <[EMAIL PROTECTED]> wrote:
> Programs linked to GPL'd library must be GPL, because by using the
> GPL'd library you have to comply to the license terms of this library.
> The main point is that USING a GPL'd library for a program is only
> allowed if the resulting program becomes GPL'd. The library never
> becomes part of the program which is very obvious if you consider the
> fact the the GPL does not mention linking at all.

If the library isn't part of the program, why would the GPL apply
to the program?

> The main point is that a library can force some restrictions on a program 
> which uses the library (Like the Qt Free Software License forces to
> release the written program under a free license) but NOT vice versa.

Er... where did you get this idea from?  It's certainly not a part
of copyright law, nor is it a part of the GPL.

-- 
Raul

RE: updates to Debian pages

[Darren Benham]

On 12-Oct-98 James A. Treacy wrote:
> This is just to help translators keep up to date with changes to the
> Debian pages.
> 
>  - distrib/distrib.wml was renamed to distrib/index.wml
>This caused changes to template/debian/menubar.wml
>   and template/debian/navbar.wml
> 
>I believe those are the only links affected, but I'm about to run a
> urlchecker
>over the pages to see if anything got missed.
Also, developers_corner.wml has been moved/renamed to devel/index.wml (also
affecting templage/debian/menubar.wml and template/debian/navbar.wml as well as
one of the news files in News/1997  (19971125.wml)

=
* http://benham.net/index.html <><  *
*  * -BEGIN GEEK CODE BLOCK- ---*
*Darren Benham * Version: 3.1   *
*  <[EMAIL PROTECTED]>  * GCS d+(-) s:+ a29 C++$ UL++> P+++$ L++>*
*  * E? W+++$ N+(-) o? K- w+++$(--) O M-- V- PS--   *
*   Debian Developer   * PE++ Y++ PGP++ t+ 5 X R+ !tv b DI+++ D++   *
*  <[EMAIL PROTECTED]>  * G++>G+++ e h+ r* y+*
*  * --END GEEK CODE BLOCK-- ---*
=

Re: [ettrich@troll.no: Re: copyright problem]

[Nicolás Lichtmaier]

>   people to distribute LyX in both source and binary forms. This permission
>   certainly includes linking against GUI toolkits like XForms, Motif, GTK, Qt
>   or Win32.

 `... and distributing the resulting binary.' should be added.

 You can always link in the privacy of your home. What GPL forbids is to
distribute the `derived work'.

Re: LICENSES [was: Re: Have you seen this?]

[john]

Jim writes:
> So this isn't "derivative" in the sense of the OOP idiom "IS-A"...

> and you're saying that all I have to do to "derive from" something, is to
> include it unmodified or modified?

In copyright law "is a derivative of" means "contains a copy of all or part
of".  Copyright is about making copies.
-- 
John HaslerThis posting is in the public domain.
[EMAIL PROTECTED]  Do with it what you will.
Dancing Horse Hill Make money from it if you can; I don't mind.
Elmwood, Wisconsin Do not send email advertisements to this address.

Re: GPL source policy.

[john]

Masato Taruishi writes:
> In this, the licence sait that `valid for at least three years', but I
> can't understand what date the beginnings of `at least three years'
> starts actually from. From the date when the vendor began to sell it, or
> when buyers bought it, or else?

>From the moment the vendor transferred ownership of the specific copy in
question.
-- 
John Hasler
[EMAIL PROTECTED] (John Hasler)
Dancing Horse Hill
Elmwood, WI

Re: updates to Debian pages

[James A. Treacy]

On Mon, Oct 12, 1998 at 11:09:34AM -0700, Darren Benham wrote:
> Also, developers_corner.wml has been moved/renamed to devel/index.wml (also
> affecting templage/debian/menubar.wml and template/debian/navbar.wml as well 
> as
> one of the news files in News/1997  (19971125.wml)
> 
One thing I forgot to mention to you. Could you please add the final / on
directories? It is proper to add them, but not mandatory. Additionally
it confuses the urlchecker I use (yeah I'll fix it eventually).
I have already fixed the occurrences of devel/ and got a few others in the
process.

Jay Treacy

problem with new icewm

[Michael Meskes]

I installed both packages. After starting X I found that the old name is no
longer the one compiled for gnome which removed all my applets from the
panel. I changed my setup to call icewm-gnome instead and re-created my
panels but had to notice that the panel applet no longer works. Not only
does it not appear, but no applet added after it will appear. I've set up my
panel without the pager again and it works fine but I'd like to get the
pager back.

Also I would prefer an alternative setup for these two packages with
icewm-gnome getting the higher one. Or even divided into three:
icewm-common, icewm-gnome, icewm-nognome.

Michael
-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

Packages that disappeared

[Michael Meskes]

I noticed that some packages disappeared from the site. Could anyone
enlighten me whether they are superceeded, not needed or whatelse happend?

libstdc++2.8
libg++2.8

I know these are special cases since there is no source. But as long
as we need the libs we need the packages, don't we? Or do we have
recompiled everything? On my system there are packages depending on
them (well I checked only libstdc++2.8 to be honest).

gnome-mico
gnome-mico-dev
libgtktty-dev
libgtktty0
pentium-builder
movemail
xadmin
x11amp-static
mp3.8hz

Michael

-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Michael Meskes]

On Mon, Oct 12, 1998 at 12:25:12PM +0200, Gergely Madarasz wrote:
> How can we be sure that LyX does not include things not written by them?

Wait a moment. Don't let this become ridiculous. How can we be sure that
Ulrich Depper didn't include non-GPL stuff in his glibc? You can ask this
ofr every single package. Hey, it's not our job to check that.

> And anyway we're not given permission to distribute it.

Why?

Michael
-- 
Dr. Michael Meskes  | Th.-Heuss-Str. 61, D-41812 Erkelenz | Go SF49ers!
Senior-Consultant   | business: [EMAIL PROTECTED] | Go Rhein Fire!
Mummert+Partner |  private: [EMAIL PROTECTED]| Use Debian
Unternehmensberatung AG |   [EMAIL PROTECTED]| GNU/Linux!

Re: Intent to package: smtpfeed and sendmail-wide

[Fumitoshi UKAI]

At Thu, 8 Oct 1998 10:02:06 -0400 (EDT),
Richard A Nelson <[EMAIL PROTECTED]> wrote:

> hrm... 
> The requested URL /~monotori/sendmail.html was not found on this server.

Oops, sorry, it's typo.
The URL should be http://www.wide.ad.jp/~motonori/sendmail.html

> Is there somewhere else I can look?  am I correct in assuming that
> WIDE is support for DBCS?

No, WIDE is the research projects in Japan, stands for 
`Widely Integrated Distributed Environments.'
This WIDE patch does not mean support for DBCS.

-- 
Fumitoshi UKAI

Re: Packages that disappeared

[Martin Schulze]

Michael Meskes wrote:
> xadmin

Request by maintainer=author, iirc.

> x11amp-static
> mp3.8hz

You didn't watch the 100 messages thread on debian-private?

Regards,

Joey

-- 
Linux - the choice of a GNU generation

Re: problem with new icewm

[Martin Schulze]

Michael Meskes wrote:
> I installed both packages. After starting X I found that the old name is no
> longer the one compiled for gnome which removed all my applets from the
> panel. I changed my setup to call icewm-gnome instead and re-created my
> panels but had to notice that the panel applet no longer works. Not only
> does it not appear, but no applet added after it will appear. I've set up my
> panel without the pager again and it works fine but I'd like to get the
> pager back.
> 
> Also I would prefer an alternative setup for these two packages with
> icewm-gnome getting the higher one. Or even divided into three:
> icewm-common, icewm-gnome, icewm-nognome.
 ^
This should be kept as `icewm' imho.

Regards,

Joey
-- 
Linux - the choice of a GNU generation

Re: Packages that disappeared

[J.H.M. Dassen \(Ray\)]

On Mon, Oct 12, 1998 at 06:13:33PM +0200, Michael Meskes wrote:
> gnome-mico
> gnome-mico-dev

Replaced by an ORB written by the GNOME folks themselves, "orbit".

Ray
-- 
UNFAIR  Term applied to advantages enjoyed by other people which we tried 
to cheat them out of and didn't manage. See also DISHONESTY, SNEAKY, 
UNDERHAND and JUST LUCKY I GUESS. 
- The Hipcrime Vocab by Chad C. Mulligan  

Re: [larsbj@ifi.uio.no: Re: copyright problem]

[Gergely Madarasz]

On Mon, 12 Oct 1998, Michael Meskes wrote:

> On Mon, Oct 12, 1998 at 12:25:12PM +0200, Gergely Madarasz wrote:
> > How can we be sure that LyX does not include things not written by them?
> 
> Wait a moment. Don't let this become ridiculous. How can we be sure that
> Ulrich Depper didn't include non-GPL stuff in his glibc? You can ask this
> ofr every single package. Hey, it's not our job to check that.

The problem here is that we can't distribute it under the terms of the GPL
(read the kde announcement) while they say it is plain GPL, so they say
they can include other people's GPL-ed stuff. glibc2 doesnt have a
contradictory licence like this.

> > And anyway we're not given permission to distribute it.
> 
> Why?

Did you see it written down? :) GPL forbits it, unless explicitly
stated in the licence. And when we're given permission to distribute it in
written form (it should be part of the licence or something) then it will
be clear it is not exactly GPL, though it is open source.

-- 
Madarasz Gergely   [EMAIL PROTECTED] [EMAIL PROTECTED]
  It's practically impossible to look at a penguin and feel angry.
  Egy pingvinre gyakorlatilag lehetetlen haragosan nezni.
HuLUG: http://mlf.linux.rulez.org/

Re: problem with new icewm

[Gergely Madarasz]

On Mon, 12 Oct 1998, Michael Meskes wrote:

> I installed both packages. After starting X I found that the old name is no
> longer the one compiled for gnome which removed all my applets from the
> panel. 
> I changed my setup to call icewm-gnome instead and re-created my
> panels but had to notice that the panel applet no longer works. Not only
> does it not appear, but no applet added after it will appear. I've set up my
> panel without the pager again and it works fine but I'd like to get the
> pager back.

Hmpf... icewm versions up to 0.9.13 weren't compiled for gnome either.
Well, they were gnome-compliant in some cases, like the pager, etc, but
didnt need to be linked against libgnome, etc. I thought this linking gave
some additional features (like the gnome menu), but didn't remove those
from the plain version which were available before, and since I dont
actively use gnome, I didn't test it thoroughly. I'll look into it.

> Also I would prefer an alternative setup for these two packages with
> icewm-gnome getting the higher one. Or even divided into three:
> icewm-common, icewm-gnome, icewm-nognome.

I'll probably do this.

Thanks,

Greg

-- 
Madarasz Gergely   [EMAIL PROTECTED] [EMAIL PROTECTED]
  It's practically impossible to look at a penguin and feel angry.
  Egy pingvinre gyakorlatilag lehetetlen haragosan nezni.
HuLUG: http://mlf.linux.rulez.org/

Re: [ettrich@troll.no: Re: copyright problem]

[Raul Miller]

Michael Meskes <[EMAIL PROTECTED]> wrote:
> I told him I would remove the first sentence but other than that it looks
> okay to me.

Yeah.

With that first sentence in, I think he'd argue that he doesn't need
anyone's permission to apply it to third-party GPLed software: he's
declaring what the GPL says.  [If nothing else, we should fly 
such a statement past RMS's lawyer.]

Without it, it looks like a simple granting of permission.

-- 
Raul

Re: Intent to package LEIM

[Milan Zamazal]

Grr, after I uploaded it I've found emacs20 *source* package already
contains all the LEIM data files, so I removed the uploaded leim
package.

I think the best solution would be to produce leim binary package from
the emacs20 source package.  Rob, could you do so in the next version of
the emacs20 package please?  Thanks.

Milan Zamazal

Re: gnome and gtk--

[Ben Gertzfield]

> "Chris" == Chris Waters <[EMAIL PROTECTED]> writes:

Chris> Perhaps I'm missing something, but I don't see any way of
Chris> using gtk-- with gnome at the moment.  I was going to try
Chris> packaging gnome-hack (for my own use -- I'd want to check
Chris> with the nethack maintainer before doing anything more with
Chris> it), but it seems to require gtk-- and gtk1.1, and the two
Chris> don't seem to work together at this point.

I'm the nethack maintainer. If you wish to package up and maintain
gnomehack, I'd be perfectly happy, so long as you use the same sort of
debian/* files that I use in the nethack package, and if you find bugs
in them you let me know :)

-- 
Brought to you by the letters W and T and the number 3.
"* denotes Hot and Spicy!" -- *Ben Gertzfield
Debian GNU/Linux -- where do you want to go tomorrow? http://www.debian.org/
I'm on FurryMUCK as Che, and EFNet and YiffNet IRC as Che_Fox.

Thoughts on installation

[Dave Swegen]

First of all a quick apology if it turns out the following message isn't
being posted to the correct forum.

A few weeks ago I switched from RH to debian, and was slightly dismayed at
the installation process, which I found to be less than flexible. After
wrestling with it for some six-seven hours I finally got it up and running
properly. There were two steps which caused most of the problems:
Installing lilo, and getting X installed.

I will briefly give an outline of my HD structure:
/dev/hda1   DOS
/dev/hda2   linux part. used for backup
/dev/hdb1   swap
/dev/hdb2   /
/dev/hdb3   /usr

Now, RH had always placed the boot loader on hda. Debian OTOH insisted on
placing it on hdb2. Now, if I had had floppies available (which I didn't)
things would have been easier (I ended up having to take a one hour bike
ride to get some floppies - who said computing doesn't keep you fit :)
What would have been even easier if there had been an option to edit
lilo.conf there and then in some manner. I would also like to say I think
the default lilo.conf is a mess, to put it mildly. It is badly
structured, makes the process of adding other images less than obvious and
I was thankful I had my old lilo.conf hanging about.

The second problem was installing X (an issue which I get the impression
is already being addressed). Basically the X configuration process died on
me all 4 times I tried it during install, leaving .dkpg-new files
everywhere, so the only option was to install only the base system, and
then manually install and configure X later on.

In case this comes across as a wholly negative impression I maybe should
mention the things I did like: The preselected setups, allowing for a
quick start (well, OK, so they wouldn't have worked in my case :); the
wealth of packages on the main CD (but no xv on any of the CDs, even the
non-free); and the hints about what the next step is, or what alternative
steps exist was also very useful.

While I'm at it, I might as well comment on another aspects of Debian that
IMHO could be improved: The practice of naming packages which are only
installer scripts (netscape, star-office) is confusing (OK, so the 23k
size gave it away) and I have seen quite a few people query about them.
Just adding the postfix '-installer' or somesuch would save a lot of
confusion.

Also, how is one supposed to say 'dpkg' without tying ones tongue up in
a knot? ;)

Anyway, apologies for the long rambling nature of this mail, and keep up
the good work.

Cheers
Dave

Re: Ropes in stl (was Re: lack of wstring in libstdc++2.8-dev)

[Rob Browning]

Chris <[EMAIL PROTECTED]> writes:

> But I get these errors:
> 
> In file included from /usr/include/g++-2/stl_rope.h:2107,
>  from /usr/include/g++-2/rope.h:18,
>  from blah.cc:1:
> /usr/include/g++-2/ropeimpl.h:1085: warning: decimal integer constant is so
> large that it is unsigned

As I recall, this isn't an error, just a warning.  There are also a
number of other annoying warnings that spew like crazy in some of the
other stl files.  I filed a bug with patches a while back.  Hopefully
they'll be forwarded.  I don't think any of these bugs hurt anything,
the compiler does the right thing, it just complains.

-- 
Rob Browning <[EMAIL PROTECTED]> PGP=E80E0D04F521A094 532B97F5D64E3930

Re: Intent to package LEIM

[Rob Browning]

Milan Zamazal <[EMAIL PROTECTED]> writes:

> Grr, after I uploaded it I've found emacs20 *source* package already
> contains all the LEIM data files, so I removed the uploaded leim
> package.
> 
> I think the best solution would be to produce leim binary package from
> the emacs20 source package.  Rob, could you do so in the next version of
> the emacs20 package please?  Thanks.

I'm not really an expert on LEIM, so I'm not sure what you're asking
for.  What exactly is the problem, and how can I fix it?  I'd be happy
to as soon as I understand it.

Thanks

-- 
Rob Browning <[EMAIL PROTECTED]> PGP=E80E0D04F521A094 532B97F5D64E3930

Re: updates to Debian pages

[Philipp Frauenfelder]

Hi

James A. Treacy wrote:
> One thing I forgot to mention to you. Could you please add the
> final / on directories? It is proper to add them, but not
> mandatory. Additionally it confuses the urlchecker I use (yeah
> I'll fix it eventually). I have already fixed the occurrences
> of devel/ and got a few others in the process.

Adding the / is one of the easiest ways to save bandwidth. A
browser asking for a directory without a / gets a 301
redirection:

--
$ telnet www.niederglatt.lugs.ch 80
Trying 10.0.24.1...
Connected to www.niederglatt.lugs.ch.
Escape character is '^]'.
GET /Buchhaltung HTTP/1.0

HTTP/1.1 301 Moved Permanently
Date: Mon, 12 Oct 1998 20:24:44 GMT
Server: mod_perl/1.15 Apache/1.3.1 (Unix) Debian/GNU
mod_perl/1.15
Location: http://www.niederglatt.lugs.ch/Buchhaltung/
Connection: close
Content-Type: text/html



301 Moved Permanently

Moved Permanently
The document has moved http://www.niederglatt.lugs.ch/Buchhaltung/";>here.

Connection closed by foreign host.
---

Btw, http://www.niederglatt.lugs.ch/Buchhaltung/index.html
exists.

Regards,
Philipp

Re: Local IP address / Java Incompatibility

[Bernd Eckenfels]

On Mon, Oct 12, 1998 at 04:18:54PM +0200, Rainer Dorsch wrote:
> We found an incompatibility between Java on Solaris 2.5.1/2.6 and on Debian 
> 2.0.r2 Could anybody suggest how to find out, if it is a bug in Debian Linux 
> or JDK?

You have to use getLocalHost() on a connected Socket, not on InetAddress to
get the local address of the socket.

Greetings
Bernd

> 
>   InetAddress inetadr = InetAddress.getLocalHost();
>   System.out.println("LocalIP:   " + inetadr.getHostAddress());
>   System.out.println("LocalName: " + inetadr.getHostName());
> 
-- 
  (OO)  -- [EMAIL PROTECTED] --
 ( .. )  [EMAIL PROTECTED],linux.de,debian.org} http://home.pages.de/~eckes/
  o--o *plush*  2048/93600EFD  [EMAIL PROTECTED]  +497257930613  BE5-RIPE
(OO)   If privacy is outlawed only Outlaws have privacy

Re: debian/rules and find

[Rob Browning]

Martin Schulze <[EMAIL PROTECTED]> writes:

> So please check your debian/rules files for constructs like the
> following:
> 
>chmod g+w `find debian/tmp -name foo`
> 
>find debian/tmp -name foo|xargs chmod g+w
> 
> the correct way to implement this would be
> 
>find debian/tmp -name foo|xargs -r chmod g+w

I'd say that this is *only* correct if debian/tmp being empty is *not*
an error condition.

You shouldn't make this change in cases where you know that there's
supposed to be something in debian/tmp (or wherever); you'll just mask
the problem.  In that case, the code *should* fail.  This is analogous
to whether or not you should say:

  for(int i = 0; i < foo; i++) { bar(); }

or 

  for(int i = 0; i != foo; i++) { bar(); }

Technically (though I know no one ever does it) the latter is
preferable.  In general you want the weakest test rather than the
strongest so that failures happen sooner rather than later --
i.e. once they've been compounded.

MHO

-- 
Rob Browning <[EMAIL PROTECTED]> PGP=E80E0D04F521A094 532B97F5D64E3930

Re: GNotepad

[Martin Schulze]

Ole J. Tetlie wrote:
> Is anyone packing gnotepad?
> 
> [ Sorry if anyone tried to a post of mine and bounced. I played with
> exim.conf and forgot to "unplay" the rewrite. ]

Since nobody stepped forward, I take it for now.  It in the process
of being built right now.

Regards,

Joey

-- 
Linux - the choice of a GNU generation

Re: Bug#27753: libpgjava: depends on jdk1.1-runtime, which is now included in jdk1.1

[Stephen Zander]

> "Hamish" == Hamish Moffatt <[EMAIL PROTECTED]> writes:
Hamish> Why doesn't jdk1.1 provide java-virtual-machine? That's an
Hamish> officially listed virtual package and would have avoided
Hamish> this problem, and would also solve two bug reports filed
Hamish> against guavac.

Because I wasn't sure if that was still a valid virtual package name? :)

We've gone round the issue of virtual packages for jvm's a couple of
times & I'm still not certain anything got decided.  Next release will
comply.

-- 
Stephen
---
Perl is really designed more for the guys that will hack Perl at least
20 minutes a day for the rest of their career.  TCL/Python is more a
"20 minutes a week", and VB is probably in that "20 minutes a month"
group. :) -- Randal Schwartz

Re: gnome and gtk--

[Marcus Brinkmann]

Hello,

On Sun, Oct 11, 1998 at 03:28:27PM -0700, Chris Waters wrote:
> Havoc Pennington wrote:
> 
> > On Sun, 11 Oct 1998, Chris Waters wrote:
> 
> > > I think it would really be nice to get a gnome-supporting version
> > > of gtk-- in before the slink freeze.  Is anyone working on this?
> 
> > Not really possible without hacking Gtk-- (which can be done, but
> > it's work). Gtk-- can be built with either Gtk 1.0 or Gtk 1.1, if you
> > install both things would get, uh, confused.
> 
> 'Bout what I figured, but wouldn't it be possible to produce two
> versions which conflict?  Not a perfect solution, but it would make it
> possible for people like me who want to work on gnome-related gtk--
> stuff to do so.  The conflicts could be cleaned up when someone had the
> time to hack on it (presumably post-slink).

I sthere any reason why I can't build gtk-- libraries with gnome support? Do
we need two versions? Why would we need a version without gnome support?

About the gtk+ version to use:
 
>From the release notes page:
"NOTE! Its now a policy that gtk-- works with 1.0.x gtk+'s and only gtk+1.1
from the CVS. Thus to use gtk1.1, you will want to have both gtk+ and gtk-- from
CVS!"

So Gtk-- 0.9.x will be build with the stable release of gtk+.

> Just a thought -- I'm about to try building my own personal gnome-gtkmm
> package (which will conflict with gtkmm), but I don't yet have any
> experience at packaging libraries, so I'm a little scared.  I doubt if
> I'll be able to finish in time for the freeze.

Could you please talk to me about this? I assume both will be build from the
same sources, so it does make sense to just include the rules for them in my
packaging scripts.

Marcus,
gtk-- debian maintainer

-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09