FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Debian > Debian User

 
 
LinkBack Thread Tools
 
Old 06-04-2008, 02:02 PM
"Patrick Wiseman"
 
Default Running testing? -- read this.

On Wed, Jun 4, 2008 at 9:11 AM, Chris Bannister
<mockingbird@earthlight.co.nz> wrote:
> ----- Forwarded message from Manoj Srivastava <srivasta@debian.org> -----
>
> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.0.60 (gnu/linux) (x86_64-unknown-linux-gnu)
> From: Manoj Srivastava <srivasta@debian.org>
> Date: Tue, 03 Jun 2008 12:14:03 -0500
> To: debian-devel@lists.debian.org
> Subject: Re: Bug#484129: release.debian.org: packages in tasks should be fixed in priority and removed in last resort after
> discussion
>
> On Tue, 3 Jun 2008 12:05:45 -0400, Joey Hess <joeyh@debian.org> said:
>
>> Manoj Srivastava wrote:
>>> I thought I had answered that. The only version that th project
>>> releases for end users is stable.
>
>> Debian has been releasing versions of testing for end users for years.
>
> This is perhaps a matter of semantics.
>
> Where I come from, when you are testing something, it is not
> "released". Releases, in my view, have release numbers, so people can
> refer to them in bugs and recommend to other folks; our testing is
> volatile, and changes at least daily. I do agree that microsoft has
> blurred issue, but releasing things early to paying consumers to test,
> but hey.
>
> We call it testing for a reason: it is a a branch for testing
> what our next release will be. Do we try to keep it as close to a
> viable release as we can? Sure. Do we try to make it easy for people to
> test? Absolutely. But it would be diseembling to pretend that this
> version is anything but a test version of a product under development,
> and that stuff happens with development versions.
>
> People who run our test versions are part of our community, they
> are helping us test our development packages. We will not make things
> harder for them than they need to be, but they should also not be
> shielded from the fact (and it is a fact) that what they are ending up
> installing is a test version of a product under development.
>
> I also made an assumption: an end user is one who does not care
> to be a early adopter guinea pig; and for people who do not care for
> breakage and do not want to participate in product development and
> testing, the only variant we produce is called the stable release

So? I've been using testing for years, and have found it to be
remarkably stable - it's remarkable precisely because it IS 'testing'.
Sometimes (rarely) things break, but that's something I prefer to
live with so that I can have an up-to-date system. The current
'stable' is relatively up to date, as it's fairly recent, but there
tends to be a long period between stable releases.

Patrick


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-04-2008, 02:48 PM
Carl Fink
 
Default Running testing? -- read this.

On Wed, Jun 04, 2008 at 10:02:13AM -0400, Patrick Wiseman wrote:

> So? I've been using testing for years, and have found it to be
> remarkably stable - it's remarkable precisely because it IS 'testing'.
> Sometimes (rarely) things break, but that's something I prefer to
> live with so that I can have an up-to-date system. The current
> 'stable' is relatively up to date, as it's fairly recent, but there
> tends to be a long period between stable releases.

And in fact, for (using a number I just pulled out of thin air) 90% of the
time I've been using Debian, going back to Slink, Stable has been so
obsolete as to be hard to use for anything but some servers.
--
Carl Fink nitpicking@nitpicking.com

Read my blog at blog.nitpicking.com. Reviews! Observations!
Stupid mistakes you can correct!


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-04-2008, 03:49 PM
Michal Kapalka
 
Default Running testing? -- read this.

Carl Fink wrote:
> On Wed, Jun 04, 2008 at 10:02:13AM -0400, Patrick Wiseman wrote:
>
>> So? I've been using testing for years, and have found it to be
>> remarkably stable - it's remarkable precisely because it IS 'testing'.
>> Sometimes (rarely) things break, but that's something I prefer to
>> live with so that I can have an up-to-date system. The current
>> 'stable' is relatively up to date, as it's fairly recent, but there
>> tends to be a long period between stable releases.
>
> And in fact, for (using a number I just pulled out of thin air) 90% of the
> time I've been using Debian, going back to Slink, Stable has been so
> obsolete as to be hard to use for anything but some servers.

As usually, it all depends on what you need. I run Etch on my
(relatively old) laptop and so far there have been only few packages
that I needed to upgrade to their "testing" version. For me it seems
like a good deal to have a base system that is not breaking randomly on
updates, and just upgrade a few user-level applications. I used to use
Debian testing for a long time, but I realized that it just consumes too
much time to repair things that suddenly break, especially if the
problems are related to drivers/Xorg (e.g., using an external display
with my laptop needed a lot of tweaking after various Xorg updates of
"testing", and it works perfectly, and continuously, since I switched to
Etch, albeit without the new "xrandr" features).

On the other hand, the number of "release-critical" bugs is currently
higher in Etch than in Lenny -- which is quite surprising. Of course, it
is partially because, as far as I understand, many bugs get resolved
upstream and the patches are not backported to the old versions of
applications in Etch. But it is probably not the only explanation, is it?

Best,

Michal


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-04-2008, 04:05 PM
Johannes Wiedersich
 
Default Running testing? -- read this.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

This may be more interesting [1]:

Pierre Habouzit wrote:
[...]
> No it's not. The principal goal of testing is to evaluate what
> would be our next stable if we tried to release *RIGHT NOW*.
> Packages with RC bugs cannot be part of a release, so must be kept
> out. *I* don't really care about testing being fully usable all the
> time, I care about it being a good representation of what could be
> released. Testing was meant as a release management tool, not
> really as a usable distribution.

Pierre Habouzit is one of the release assistants responsible for the
shape of lenny. IMHO, he should at least care about testing being usable.

Johannes

[1] http://lists.debian.org/debian-devel/2008/06/msg00048.html
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIRr1dC1NzPRl9qEURAgZnAJ9QwudKRV+iKIrggaOTOV ecHQt+pQCcDCcm
sNltJIUpwUHn3Hqjac0OH6k=
=Fnkx
-----END PGP SIGNATURE-----


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-04-2008, 04:10 PM
Tzafrir Cohen
 
Default Running testing? -- read this.

On Wed, Jun 04, 2008 at 06:05:49PM +0200, Johannes Wiedersich wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> This may be more interesting [1]:
>
> Pierre Habouzit wrote:
> [...]
> > No it's not. The principal goal of testing is to evaluate what
> > would be our next stable if we tried to release *RIGHT NOW*.
> > Packages with RC bugs cannot be part of a release, so must be kept
> > out. *I* don't really care about testing being fully usable all the
> > time, I care about it being a good representation of what could be
> > released. Testing was meant as a release management tool, not
> > really as a usable distribution.
>
> Pierre Habouzit is one of the release assistants responsible for the
> shape of lenny. IMHO, he should at least care about testing being usable.

IMHO you should at least quote him with the proper context.
(E.g.: his clarifications in some messages that followed).

--
Tzafrir Cohen | tzafrir@jabber.org | VIM is
http://tzafrir.org.il | | a Mutt's
tzafrir@cohens.org.il | | best
ICQ# 16849754 | | friend


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-04-2008, 04:19 PM
"Douglas A. Tutty"
 
Default Running testing? -- read this.

On Wed, Jun 04, 2008 at 10:48:57AM -0400, Carl Fink wrote:
> On Wed, Jun 04, 2008 at 10:02:13AM -0400, Patrick Wiseman wrote:
>
> > So? I've been using testing for years, and have found it to be
> > remarkably stable - it's remarkable precisely because it IS 'testing'.
> > Sometimes (rarely) things break, but that's something I prefer to
> > live with so that I can have an up-to-date system. The current
> > 'stable' is relatively up to date, as it's fairly recent, but there
> > tends to be a long period between stable releases.
>
> And in fact, for (using a number I just pulled out of thin air) 90% of the
> time I've been using Debian, going back to Slink, Stable has been so
> obsolete as to be hard to use for anything but some servers.

Huh? The only time I ever used Testing was when I got a new box with
SATA drives and Sarge didn't handle it well. I've stuck with Etch
since. Actually, if Woody was still getting security updates, I'd have
stuck with that. OK, perhaps some web pages wouldn't now work with
Woody's Firefox.

Doug.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-04-2008, 05:16 PM
Carl Fink
 
Default Running testing? -- read this.

On Wed, Jun 04, 2008 at 12:19:33PM -0400, Douglas A. Tutty wrote:
I wrote:

> > And in fact, for (using a number I just pulled out of thin air) 90% of the
> > time I've been using Debian, going back to Slink, Stable has been so
> > obsolete as to be hard to use for anything but some servers.
>
> Huh? The only time I ever used Testing was when I got a new box with
> SATA drives and Sarge didn't handle it well. I've stuck with Etch
> since. Actually, if Woody was still getting security updates, I'd have
> stuck with that. OK, perhaps some web pages wouldn't now work with
> Woody's Firefox.

"... going back to Slink ..."

Etch is still relatively fresh. At the same time, I had to use Lenny to
support my laptop. My server runs Etch.

If Lenny isn't released within about a quarter, Stable will again begin to
be hard to use with new computers, people who want recent versions of
software, etc.
--
Carl Fink nitpicking@nitpicking.com

Read my blog at blog.nitpicking.com. Reviews! Observations!
Stupid mistakes you can correct!


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-04-2008, 08:41 PM
Michelle Konzack
 
Default Running testing? -- read this.

Am 2008-06-04 10:02:13, schrieb Patrick Wiseman:
> So? I've been using testing for years, and have found it to be
> remarkably stable - it's remarkable precisely because it IS 'testing'.

How can this be with the perl 5.10 transition?

The half of my system was down even by upgrading every day...

Note: I am Linux Developer and run Unstable/Testing/Stable all the time.

Thanks, Greetings and nice Day
Michelle Konzack
Systemadministrator
24V Electronic Engineer
Tamay Dogan Network
Debian GNU/Linux Consultant


--
Linux-User #280138 with the Linux Counter, http://counter.li.org/
##################### Debian GNU/Linux Consultant #####################
Michelle Konzack Apt. 917 ICQ #328449886
+49/177/9351947 50, rue de Soultz MSN LinuxMichi
+33/6/61925193 67100 Strasbourg/France IRC #Debian (irc.icq.com)
 
Old 06-05-2008, 10:42 AM
Johannes Wiedersich
 
Default Running testing? -- read this.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 2008-06-04 18:10, Tzafrir Cohen wrote:
> IMHO you should at least quote him with the proper context.
> (E.g.: his clarifications in some messages that followed).

IMHO his clarifications just confirm that *IMHO* he takes the decision
to remove packages too lightly.

I admit that I gave a short quote of a rather long thread, but I did
give a link to the whole discussion.

The whole discussion started with a bug report [1] of Daniel R.:

> This weekend (1-June-2008) several important gnome metapackages have been
> removed from Debian Lenny repositories:
>
> gnome-core
> gnome-desktop-environment
> gnome-cups-manager
> update-manager
> update-notifier
> libgnomecupsui1.0-1c2a
>
> Now, gnome desktop does not appear in Aptitude's tasks.

If a release assistant writes what I quote in this context, it alarms
me. There already have been discussions about whether there should be a
goal of keeping testing in a usable state vs. breaking it 'on purpose'
(see [1,2]). So, I don't think I put the quote out of proper context. He
did admit that gnome-core and gnome-desktop-environment were removed on
mistake (I don't blame anyone for a mistake), but I do have the
impression that the update-* packages as well as ntp (not mentioned
there) were removed *despite* the release managers wanting them back for
release. And I personally don't think that this is a good idea.

Packages *temporarily* removed from testing while making testing less
usable IMHO is not a good idea.

Packages removed from testing, because there is little or no chance to
get them in releasable shape, however, is a good idea.

That all is just the very humble opinion of a user of testing and who is
not a DD.

To make things clear: I believe that as a whole the developers and the
release managers do an *excellent* job and I thank them for their big
efforts. Testing is the very proof just by how much debian gets better
every day. Thank you soooooo much!!!!!

Debian rocks! Resistance is futile! ;-)

Cheers,
Johannes

[1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=484009
[2] http://lists.debian.org/debian-devel/2008/06/msg00060.html
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIR8MlC1NzPRl9qEURAp0IAJ487goZTqmP+k1dobIDms KlK0huXwCfaz9u
J/qiXxd+pFQ9jlJvHLZdbm8=
=4oxm
-----END PGP SIGNATURE-----


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 06-05-2008, 01:44 PM
"Damon L. Chesser"
 
Default Running testing? -- read this.

On Thu, 2008-06-05 at 01:11 +1200, Chris Bannister wrote:
> ----- Forwarded message from Manoj Srivastava <srivasta@debian.org> -----
>
> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.0.60 (gnu/linux) (x86_64-unknown-linux-gnu)
> From: Manoj Srivastava <srivasta@debian.org>
> Date: Tue, 03 Jun 2008 12:14:03 -0500
> To: debian-devel@lists.debian.org
> Subject: Re: Bug#484129: release.debian.org: packages in tasks should be fixed in priority and removed in last resort after
> discussion
SNIP

OK, my two cents worth: Testing is for testing. If you are not
testing, you don't run it. Testing is NOT meant to be a working
release, never has been. Nobody cares if Testing is broke. They will
fix it. But it will bet fixed when they have the fix for it: no one
will loose sleep over the non-functioning of Testing (except for the
user trying to use Testing as his/hers workstation).

Now the above is my observation, not the Debian policy.

see: http://www.us.debian.org/releases/ and
http://www.us.debian.org/doc/FAQ/ch-ftparchives#s-testing for the
official Debian policy. Joey Hess has said on this list that in his
experience Testing was seldom broke and seldom for any length of time
and should work as a desktop for most people (Joey Hess is one of the
Debian Developers).

Any time you leave Stable, you leave all promise behind that Debian will
work. I run Sid or Ubuntu. Sid has only become unusable for me only
once (PAM broke during an update). If you insists on running Testing OR
Unstable there is one program you MUST run: apt-get apt-listbugs you
have to read the output before proceeding to any updating. This will
save your bacon.

HTH
--
Damon L. Chesser
damon@damtek.com
http://www.linkedin.com/in/dchesser
 

Thread Tools




All times are GMT. The time now is 03:38 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org