Re: The number of popcon.debian.org-submissions is falling

[Tim Clewlow]

On Tuesday 20 July 2010 14:41:49 Petter Reinholdtsen wrote:
> The number of submissions to the Debian popularity-contest collector
> is falling, and has done so for some time now.  This can be easily
> seen on http://popcon.debian.org/stat/sub-i386.png >.
>
> This is mostly caused by a fall in the number of Lenny installations,
> as can be seen from
> http://popcon.debian.org/stat/release-1year.png >.
>
> Anyone got any idea how to can get more machines to report to
> popcon.debian.org?  Or can there be some other problem causing the
> fall in the number of submissions?
.

I suspect it has more to do with the general acceptance of *nix
based systems. In the past, *nix users were often want to proclaim
the use of *nix widely and loudly as a way of promoting *nix to the
general public (admit it, we all did it at some time :-). Popcon was
just one more (perceived) way of helping with the promotion. But
now, *nix is well known to the general public, they may not use it,
but they know it exists, and so the need to promote is reduced.

I know for myself I dont bother to install popcon because I really
dont care about popularity any more. I am happy that there is
already a massive *nix base in the community. When I choose to
install *nix, its because of its functionality - all the
tools/applications I need are either already installed, or easy to
install.

Winning a popularity contest just isn't one of my motivations.

Cheers, Tim.

PS - If you want to record how many people are using different
things, then perhaps if the update servers record IP addresses and
requested updates then this could give you more data as it does not
require a system has specifically installed a popcon type
application for this method to work.

-- 
I got the bends modelling, must have been all them lines, and points.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/221afa7816052de55d45087c07ae494b.squir...@192.168.1.100

Re: teaching users how to submit good bug reports

[Ben Hutchings]

Julien Cristau wrote:
> On Sat, Jul 24, 2010 at 23:50:36 +0100, Ben Hutchings wrote:
> 
> > On Sat, 2010-07-24 at 15:06 -0400, Holger Levsen wrote:
> > [...]
> > > and b.) IMHO reportbug-ng should be installed by the default 
> > > desktop task.
> > 
> > Yes, if it supports bug scripts properly now.
> > 
> It doesn't.  I'm still tempted to add Conflicts on it in my packages
> because it's a source of additional work.  Thankfully I seem to be
> getting fewer reports from rng than I did in the past (that said I'm not
> reading half of them, so I'm not sure).

reportbug does have a GUI (Gtk-based) and a desktop file, but the GUI
requires python-gtk2 and python-vte which the reportbug package only
suggests.  Maybe the GUI code and desktop file should be moved to a new
package, reportbug-gtk, which depends on them.  Either way, this GUI
does support bug scripts and is therefore a better candidate for the
desktop task.

Ben.

-- 
Ben Hutchings
Once a job is fouled up, anything done to improve it makes it worse.


signature.asc
Description: This is a digitally signed message part

Bug#590214: support for submitting bug reports via http

[Stefano Zacchiroli]

Package: reportbug
Version: 4.12.4
Severity: normal

On Sat, Jul 24, 2010 at 11:50:36PM +0100, Ben Hutchings wrote:
> On Sat, 2010-07-24 at 15:06 -0400, Holger Levsen wrote:
> > So, to summarize: a.) I still think reportbug should be able to submit bugs 
> > using port 80
> 
> Agreed.

Ditto.
As I haven't been able to find a feature request for that, here is one!

Dear reportbug maintainers, the contest of this request is the -devel
(sub)thread started at
. In
subsequent messages, it has been argued that permitting to report bugs
via http (actually, via port 80 ...) would remove some more barriers to
bug reporting, due to firewalling and the like.

I understand that for such a feature mere support in reportbug isn't
enough, so you might want to block this feature request by some other
feature request on debbugs (I haven't yet checked whether the latter
request already exists or not).

Cheers


PS that thread just made me try again the GTK2 UI of reportbug: it's
   just *great* nowadays, kudos!

-- 
Stefano Zacchiroli -o- PhD in Computer Science \ PostDoc @ Univ. Paris 7
z...@{upsilon.cc,pps.jussieu.fr,debian.org} -<>- http://upsilon.cc/zack/
Dietro un grande uomo c'è ..|  .  |. Et ne m'en veux pas si je te tutoie
sempre uno zaino ...| ..: | Je dis tu à tous ceux que j'aime


signature.asc
Description: Digital signature

Re: teaching users how to submit good bug reports

[Ron Johnson]

On 07/24/2010 02:06 PM, Holger Levsen wrote:
[snip]


And then there is the (nowadays perceived) problem that reportbug needs a
working MTA setup or at least outgoing traffic on port 25/587. Both ports are
blocked on almost all my machines, so I still have not much bothered with
reportbug. (I'd use it for when a maintainer tells me to use it as it will
collect some information automatically, but thats it.)



For at least a couple of years, reportbug has been able to send mail 
via the user's ISP's smtp server, just like he sends "regular" email.


I'd bet that, given the correct ~/.reportbugrc options, you can also 
use smtp.google.com.


--
Seek truth from facts.


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4c4b7128.1040...@cox.net

Re: teaching users how to submit good bug reports

[Julien Cristau]

On Sat, Jul 24, 2010 at 23:50:36 +0100, Ben Hutchings wrote:

> On Sat, 2010-07-24 at 15:06 -0400, Holger Levsen wrote:
> [...]
> > and b.) IMHO reportbug-ng should be installed by the default 
> > desktop task.
> 
> Yes, if it supports bug scripts properly now.
> 
It doesn't.  I'm still tempted to add Conflicts on it in my packages
because it's a source of additional work.  Thankfully I seem to be
getting fewer reports from rng than I did in the past (that said I'm not
reading half of them, so I'm not sure).

Cheers,
Julien


signature.asc
Description: Digital signature

Re: teaching users how to submit good bug reports

[Ben Hutchings]

On Sat, 2010-07-24 at 15:06 -0400, Holger Levsen wrote:
[...]
> And then there is the (nowadays perceived) problem that reportbug needs a 
> working MTA setup or at least outgoing traffic on port 25/587. Both ports are 
> blocked on almost all my machines, so I still have not much bothered with 
> reportbug. (I'd use it for when a maintainer tells me to use it as it will 
> collect some information automatically, but thats it.)

Given that you know the benefit of reportbug to maintainers receiving
bug reports, this really is quite selfish behaviour.

> reportbug-ng works well on my laptop (but it's not in lenny, so no cookie for 
> me (as someday I stopped backporting it)), but most of my systems dont have a 
> GUI neither.
> 
> So, to summarize: a.) I still think reportbug should be able to submit bugs 
> using port 80

Agreed.

> and b.) IMHO reportbug-ng should be installed by the default 
> desktop task.

Yes, if it supports bug scripts properly now.

Ben.

-- 
Ben Hutchings
Once a job is fouled up, anything done to improve it makes it worse.


signature.asc
Description: This is a digitally signed message part

Re: The number of popcon.debian.org-submissions is falling

[Steve Langasek]

On Thu, Jul 22, 2010 at 06:14:23AM -0700, Russ Allbery wrote:
> Goswin von Brederlow  writes:
> > Russ Allbery  writes:

> >> Maybe with some tagging as to the origin derivative, so that we can
> >> handle cases where that data isn't useful?  There's no guarantee that
> >> the Ubuntu package has much of anything to do with the Debian package
> >> of the same name; it might not even be the same software.  They're
> >> *normally* the same, but we could get some weird artifacts.

> > Well, ubuntu is quite good with tagging the version. So anything
> > reported from popularity-contest 1.48ubuntu1 is from ubuntu. :)

> This is true for packages that Ubuntu has imported from Debian and then
> updated, but I don't believe this is true for packages that are introduced
> directly into Ubuntu and then separately introduced directly into Debian
> later.

It is not /universally/ true for such packages, yes.  Third-party software
that's uploaded to Ubuntu before it's uploaded to Ubuntu will generally
still have ubuntu-annotated version numbers, but there are exceptions.

Cheers,
-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Re: How to make Debian more attractive for users, was: Re: The number of popcon.debian.org-submissions is falling

[Steve Langasek]

On Fri, Jul 23, 2010 at 01:53:22AM +0200, Jesús M. Navarro wrote:
> Hi again, Russ:

> On Thursday 22 July 2010 14:21:09 Russ Allbery wrote:
> > "Jesús M. Navarro"  writes:
> [...]

> > I don't agree; I think it's very hard to say the same thing about testing.

> I already told you that's about perceptions and that each one has his own so 
> I'll try this once more, after that I'll leave.

> > Yes, sid sometimes breaks hard,

> It's more than that: Sid is *intended* to break hard; it's not a undesired 
> side effect.

No.  sid is *intended* to continuously integrate new versions of software
into Debian, for testing and use.  Breaking hard *is* an undesired (but
often unavoidable) side effect.

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Bug#590195: ITP: libnet-plesk-perl -- Perl extension for the Plesk XML Remote API

[Ivan Kohler]

Package: wnpp
Severity: wishlist
Owner: Ivan Kohler 

* Package name: libnet-plesk-perl
  Version : 0.03
  Upstream Author : Jeff Finucane 
* URL : http://search.cpan.org/dist/Net-Plesk/
* License : Perl
  Programming Lang: Perl
  Description : Perl extension for the Plesk XML Remote API

 Net::Plesk implements a client interface to SWSOFT's Plesk Remote API,
 enabling a perl application to talk to a Plesk managed server.



-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20100724184801.27451.6213.report...@skydancer.freeside.biz

Re: teaching users how to submit good bug reports

[Holger Levsen]

Hi,

On Donnerstag, 22. Juli 2010, Don Armstrong wrote:
> On Thu, 22 Jul 2010, Stefano Zacchiroli wrote:
> > So, point 2: are we *advertising* reportbug enough to our users?
>
> Perhaps not, but it's literally the very first thing listed on
>
> http://www.debian.org/Bugs/Reporting
>
> which is linked from http://www.debian.org and the first result for
> reporting bugs in debian, and the second for debian bug.
>
> What would probably be useful is if people like your friend who didn't
> know about reportbug first were asked what steps they'd take (or
> actually did take) to try to report bugs in the first place.

I have reported bugs for several years until I first heard about^w^wnoticed 
reportbug. I guess the reason is/was: that page first tells you very briefly 
to install a new tool (which then forces me to learn something new) and then 
it explains in easy steps how to report a bug using a technique I already 
know: by sending an email. And thats what I have been doing ever since then.

(When having the chance to do something using known or unknown techniques, 
most people will choose the known way.)

And then there is the (nowadays perceived) problem that reportbug needs a 
working MTA setup or at least outgoing traffic on port 25/587. Both ports are 
blocked on almost all my machines, so I still have not much bothered with 
reportbug. (I'd use it for when a maintainer tells me to use it as it will 
collect some information automatically, but thats it.)

reportbug-ng works well on my laptop (but it's not in lenny, so no cookie for 
me (as someday I stopped backporting it)), but most of my systems dont have a 
GUI neither.

So, to summarize: a.) I still think reportbug should be able to submit bugs 
using port 80 and b.) IMHO reportbug-ng should be installed by the default 
desktop task.


cheers,
Holger


signature.asc
Description: This is a digitally signed message part.

Re: Flag images

[Magnus Holmgren]

On onsdagen den 17 februari 2010, Dmitry E. Oboukhov wrote:
> MH> On the other hand, one application will want 16x10 icons, another one
> MH> 24x15, another one may have some effects applied on the flags to better
> MH> fit the UI design, etc.
> 
> May be the size must be included into path?
> 
> like
> flags/countires//16x10/
> flags/countires//24x15/
> 
> etc?

Not all flags have the same aspect ratio.

-- 
Magnus Holmgrenholmg...@debian.org
Debian Developer 


signature.asc
Description: This is a digitally signed message part.

Bug#590182: ITP: fgrun -- FGRun is a graphical frontend for running FlightGear

[Christopher Baines]

Package: wnpp
Severity: wishlist
Owner: Christopher Baines 


* Package name: fgrun
  Version : 1.5.2
  Upstream Author : Frederic Bouvier (Project admin) 

* URL : http://fgrun.sourceforge.net/
* License : GPL
  Programming Lang: C++
  Description : FGRun is a graphical frontend for running FlightGear

FlightGear Launch Control (FGRun) is a graphical frontend for running the
FlightGear Flight Simulator. It allows you to select a airport and aircraft,
and includes advanced options like launching from an aircraft carrier or, 
changing the screen resoluton. 
 



-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20100724160026.3261.7295.report...@chris-debian-desktop

Re: How to make Debian more attractive for users, was: Re: The number of popcon.debian.org-submissions is falling

[vishnu vardhan]

Hello DD's and Debian Users,

I have installed Debian around an year ago. Recently, I have subscribed to
various mailing lists and reading them on daily basis, whether they are
relevant to me or not. I am still learning about debian.

Prior to the installation of Debian, I don't have any experience with Linux
or any kind of Linux Distributions. I did some research and concluded that
Debian suits me well due to it's rock solid stability.

When I was first installing Debian, my room mate was also with me. My room
mate has some kind of exposure to linux distros. When the option has come
for installation of popularity-contest, he suggested me not to install it.
He told me that "it will take you to places". Eventhough I never understood
what that phrase was meant. I have contacted some other people and they too
has same kind of impression.

The issue lies with a person's perception and privacy fears. The issue has
to be dealt in a careful manner and with a holistic approach.

This is my two step approach :

Step 1 : We should know a person's idea about the package.

Q1 : Did you install the popularity-contest package at the time of
installation ?
A  : Yes or No.

Q2 : Do you know the purpose and use of popularity-contest package ?
A  : Yes or No.

Q3 : What is a popularity-contest package ?
A  : O1 : It sends the installed packages list to the Debian server
anonymously.
 O2 : It collects the data about the system architecture anonmyously.
 O3 : It collects the information about your system hardware and
installed packages and
  distributes to various third party organisations for commercial
and non-commercial
  purposes.

Step 2 : The above step will give us a clear impression about the package
perception. If most of them think that package is

[1] invading the privacy : then we should dispel their fear. The popcon
website should provide more information and it should be more interactive
too. Like providing stats on weekly, monthly, quarterly and yearly basis for
each architecture.

[2] if they have right impression but did not installed the package : There
are two reasons for the answer : [a] they do not have an active internet
connection [b] they are not simply interested, since they find that it has
no value.

[a] We can have a package similar to bugreport and ask them to send an
e-mail. The user will know what they are disclosing exactly.

[b] Can be solved with more information and interactive popcon website.

The Step 1 can achieved by a detailed survey. The survey might be carried on
the website or wiki page of debian. We can also encourage debian users with
websites or blogs to carry out the surveys independently.

To me there is nothing wrong collecting information about system
architecture and installed packages, as long it is anonymous. I hope this
might be helpful for the community. Excuse for broken english.

vishnuvardhan.

Re: How to make Debian more attractive for users, was: Re: The number of popcon.debian.org-submissions is falling

[Bernd Zeimetz]

On 07/22/2010 02:46 PM, Hans-J. Ullrich wrote:
> Hi all, 
> 
> and there is another point, I would like to mention. 
> 
> The naming of the repository is not well chosen, as it let new and 
> unexperienced people to debian feel a wrong way. The names "stable" "testing" 
> and "unstable" let the poeople think, debian is using crippled software, 
> which 
> is unstable, not well tested. In fact, even software from "unstable" is often 
> running better, than other (including closed source) software.
> 
> I suggest for this, to thinlk about other names, for example 
> 
> - stable = server
> 
> - testing = desktop
> 
> - unstable = super_modern

So all the OMGTHISISCOOL kids install unstable and have a messed up system
because there is something broken in unstable? Thats probably the worst example
for a renaming ever.

> 
> Just an example. 
> 
> Have fun!
> 
> Hans-J. Ullrich
> 
> Just an example
> 
> 


-- 
 Bernd ZeimetzDebian GNU/Linux Developer
 http://bzed.dehttp://www.debian.org
 GPG Fingerprints: 06C8 C9A2 EAAD E37E 5B2C BE93 067A AD04 C93B FF79
   ECA1 E3F2 8E11 2432 D485 DD95 EB36 171A 6FF9 435F


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4c4ad066.4020...@bzed.de

Re: [pkg-nvidia-devel] Moving diversions between packages

[Goswin von Brederlow]

Russ Allbery  writes:

> Ian Jackson  writes:
>
>> Why not have the new package ship libGL.so.1 to a more specific filename
>> and create a symlink named libGL.so.1 by hand in its postinst ?  That
>> way you can defer doing the diversion until that part of the postinst,
>> by which time the old package and its diversion are gone.
>
> Oh, huh, I hadn't thought of that.  But wouldn't this cause the library to
> temporarily disappear, which would be contrary to the last paragraph of
> Policy 8.1?  Am I being too conservative about that?

For the split second between the dpkg-divert and the ln call on install
or between the rm and dpkg-divert on removal. I think you are too
conservative there.

You are refering to "between the time that dpkg installs it and the time
that ldconfig is run in the postinst script", right? But the disapearing
would be only for a split second while your postinst runs.

MfG
Goswin


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87mxthp8t2@frosties.localdomain