Hi all,
I am a ubuntu user. It has been an quite an interesting
discussion about reportbug as well as reportbug-ng. I too use
reportbug (more than reportbug-ng) when I need to report some bug
which might be upstream as I believe there are quite few packages
which ubuntu syncs from debian. Now reportbug is cool in the point
that I do get the dependency package info. with version & a single
line description of the dependancy package . Something akin to what
dpkg -l <package-name> does but automated. Now I also have issues
with it needing some kind of mail client, if it used some way to
transport the messages without going through the mail client/server it
would be so much easier. At this point, I use webmail & have to make
a point to remember that the subject line should have the packagename
otherwise it doesn't go to the right place.
[OT] I am somewhat surprised that nobody thought it worthy or
useful to mention apport. This is the ubuntu bug-reporting tool. A
wiki telling all about it.
https://wiki.ubuntu.com/AutomatedProblemReports as well as
http://packages.ubuntu.com/gutsy/gnome/apport-gtk . Ok now what apport
does which reportbug or report-bug ng doesn't is it sends the
bug-report without resorting to mail client/server so that much less
of a steep curve. The issue/hassle here though lies in creating a
launchpad account although its similar process as signing up for
webmail or having an account to any site. Also launchpad is not just a
BTS but also handles translations, repository functions etc.
https://wiki.ubuntu.com/LaunchPad as well as
https://launchpad.net/ubuntu [/OT]
The reason for bringing the apport/launchpad reporting
infrastructure is not to have a replica but if you guys could find a
middle way. Atleast for some of the users it would be one less hassle
if it could be used to work with the browser itself. I dunno how you
guys feel but for me its always better to look at other tools also &
if possible if some of the feature-set can be replicated/used for your
bug-reporting tool/package.
[OT] Lastly, I do know that you guys have stats regarding
package installations as to how many people installed a package but
obviously it doesn't give the whole idea. For e.g. I have evolution
installed (by default) but haven't felt the need to configure it or
use it. So at some point in future could you guys think of not just
using the installation as a benchmark for popularity of a tool but
have some kind of usage pattern associated to benchmark real-life
circumstances. Do understand there are privacy issues here (apart from
basic ones) which I might not be aware of hence maybe something to
have a look at. [/OT]
Sorry for the longish mail.
Cheers!
--
Shirish Agarwal
This email is licensed under http://creativecommons.org/licenses/by-nc/3.0/
065C 6D79 A68C E7EA 52B3 8D70 950D 53FB 729A 8B17
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
