RE: Proposal to delay release of Precise Pangolin

[Sebastien Bacher]

Le mercredi 19 octobre 2011 à 16:43 +0200, Rodrigo Moya a écrit :
> Usually, the x.x.1 release of GNOME is much better, since it includes
> lots of fixes for lots of issues as people start using the final
> stable release in their distros. 

To be fair we always had our schedule adapted to include .1 until 10.10,
then the schedule got shifted to accomodate 10.10.10 and stayed shifted,
we should probably revisit and go back to what we were doing...

--
Sebastien Bacher


-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Proposal to delay release of Precise Pangolin

[Jeremy Bicha]

On 19 October 2011 10:58, Andrew Starr-Bochicchio  wrote:
> In this case, the LTS release will be after the  x.x.1 release of
> GNOME, but not by much. GNOME 3.4.1 is scheduled for April 18th [0].
> 12.04 is scheduled for April 26th, but the Release Candidate is
> scheduled for April 19th [1].
>
> Of course, this is mostly a moot point for Precise. It's still under
> discussion, but it seems like we're not going to have 3.4 in Precise.
> According to the desktop team list the plan seem to be to stick with
> 3.2.x.

My unscientific prediction is that we'll have at least half of 3.4 in
Precise. The question is more about how many and what specific apps we
want to hold back to 3.2 to focus packaging and bugfixing efforts for
the best quality by release time. Even in Oneiric's case, we held back
Totem (3.2 seems to require working 3D graphics), Epiphany (confusion
about whether new WebKit would be stable in time), and GDM (lack of
manpower/attention).

Jeremy Bicha

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Proposal to delay release of Precise Pangolin

[Allison Randal]

On 10/19/2011 06:02 AM, nick rundy wrote:
> Put the "rapid release" schedule
> on hold temporarily and make an LTS that fixes this stuff.

You'll be pleased to know that bug fixing and quality improvements are
precisely what people are focusing on for the LTS. You won't see much in
the way of new features proposed at UDS-P. This is common for LTS
cycles, but especially strong this cycle. "Quality" is a big part of
being "Precise". :)


It's a common fallacy in software development, thinking "If I only had a
few more weeks, I'd make it so much better". In fact, time-based
releases don't decrease quality, they improve it. Martin Michlmayr gave
a great talk on this at Google a few years ago, based on his research at
Cambridge:

http://video.google.com/videoplay?docid=-5503858974016723264

Allison

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Proposal to delay release of Precise Pangolin

[Andrew Starr-Bochicchio]

On Wed, Oct 19, 2011 at 10:43 AM, Rodrigo Moya
 wrote:
> I really think a bit more of time between the GNOME releases and the
> Ubuntu final release would help a lot in cleaning lots of these bugs.
> Usually, the x.x.1 release of GNOME is much better, since it includes
> lots of fixes for lots of issues as people start using the final stable
> release in their distros. So, getting x.x.1 in the final Ubuntu release
> would help a lot.

In this case, the LTS release will be after the  x.x.1 release of
GNOME, but not by much. GNOME 3.4.1 is scheduled for April 18th [0].
12.04 is scheduled for April 26th, but the Release Candidate is
scheduled for April 19th [1].

Of course, this is mostly a moot point for Precise. It's still under
discussion, but it seems like we're not going to have 3.4 in Precise.
According to the desktop team list the plan seem to be to stick with
3.2.x.

[0] https://live.gnome.org/action/show/ThreePointThree
[1] https://wiki.ubuntu.com/PrecisePangolin/ReleaseSchedule

-- Andrew Starr-Bochicchio

   Ubuntu Developer 
   Debian Maintainer

   PGP/GPG Key ID: D53FDCB1

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

RE: Proposal to delay release of Precise Pangolin

[Rodrigo Moya]

On mié, 2011-10-19 at 09:02 -0400, nick rundy wrote:
> I appreciate your e-mail, Martin :) 
> 
> > * These types of bugs are too big/complex for quick patches and too
> > small or unimportant for critical attention.
> 
> This is what I'm getting at. I don't doubt this is true. However,
> fixing this stuff is what's going to make a critical difference in
> users coming to, enjoying, and staying with Ubuntu. Put the "rapid
> release" schedule on hold temporarily and make an LTS that fixes this
> stuff. Then the LTS can last a long time as the face of Ubuntu. The
> status quo has become new releases perpetuating old bugs that are
> years old. The LTS releases up to this point are better than the
> 6-months but they still contain these bugs. I'm proposing that if bug
> #1 is going to be "Fix Released," the current full speed ahead
> rapid-release approach has to at least take a break for a cycle and
> address this stuff. NOW is an appropriate time because of Unity and
> GNOME 3.2. There's a lot of stuff that needs fixing in Unity and GNOME
> 3.2.
> 
I really think a bit more of time between the GNOME releases and the
Ubuntu final release would help a lot in cleaning lots of these bugs.
Usually, the x.x.1 release of GNOME is much better, since it includes
lots of fixes for lots of issues as people start using the final stable
release in their distros. So, getting x.x.1 in the final Ubuntu release
would help a lot.

Of course, not saying all bugs would get fixed, but it would help in
having more of this kind of bugs fixed.

Also, still talking about GNOME, the desktop team doesn't have enough
man power to fix Ubuntu and GNOME upstream bugs, so yes, we also need
25/50 more desktop developers :)



-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Proposal to delay release of Precise Pangolin

[Dustin Kirkland]

On Tue, Oct 18, 2011 at 9:15 PM, nick rundy  wrote:
> Canonical/Ubuntu, please don't feel obligated to release Precise Pangolin in
> April 2012. A delayed release would strengthen stability and allow more bugs
> to be fixed in both Unity and GNOME 3.2.
>
> Considering the "long-lived" nature of an LTS release, it would be
> preferable if Precise Pangolin was delayed a month or two (or more) than for
> it to be released on time with visible bugs. There are so many bugs that
> plague Oneiric. Many exist in GNOME 3.2. Perhaps Precise could be delayed a
> month or two and Ubuntu developers could fix some of the "minor" bugs
> plaguing GNOME 3.2?
>
> Although ranked as "minor," some of these bugs have existed for years and
> really hurt the usability of Ubuntu. For example, please see bug
> https://bugzilla.gnome.org/show_bug.cgi?id=552093 and take a look at the
> screenshots posted by the bug's commentators. John Strandberg recently
> posted  a screenshot of Oneiric that highlights how much this bug hurts
> productivity. Yet the bug has existed for more than 3 years. Sadly, the same
> can be said for many other bugs.
>
> I love Oneiric, but it has too many bugs. Please consider delaying release
> and having Ubuntu developers fix as many bugs as possible for Precise, even
> if it means fixing bugs that GNOME themselves should be fixing.
>
> I feel confident that the community will have no problem with a delay, even
> if it means skipping a 6 month release for once. The integrity of the LTS is
> worth it.

https://wiki.ubuntu.com/PrecisePangolin/ReleaseSchedule

So that sort of happens, with the LTS "dot" releases :-)

The 12.04.1 (the first of the "dot" releases") is scheduled for 23
August 2012, which is about 4 months after the 12.04 release (26 April
2012).  The "dot" release is, in fact, a bug-fix and
hardware-enablement only release cycle.  Realistically, some
enterprise server and corporate desktop users won't upgrade until that
first "dot" release.

For those interested in stabilization and quality assurance of Ubuntu
12.04, we'd very much invite you to get involved with your friendly
Ubuntu QA, SRU, and LTS dot release teams!

Cheers,
-- 
:-Dustin

Dustin Kirkland
Ubuntu Core Developer

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Proposal to delay release of Precise Pangolin

[Evan Huus]

On Wed, Oct 19, 2011 at 9:27 AM, Jonathan Carter (highvoltage)
 wrote:
> Hi Joseph
>
> On 11-10-19 12:21 AM, Joseph Toppi wrote:
>> Because no one else seemed willing to check, compact view does remove the
>> needless amount of margin, but also switches to a more list-like look and
>> changes the scrolling to horizontal. I checked in Nautilus 2.32.2.1 the
>> version that ships with 11.04 with all updates applied.
>>
>> What kind of QA process is there before a release, how can I help with that?
>
> The QA team can always use some more hands, there's even a dedicated QA
> site where you can get all the information you need: http://qa.ubuntu.com/
>
> It seems that the Desktop testing program is of particular interest to you:
> http://qa.ubuntu.com/testing/desktop-testing-program/

The QA team is already doing a fantastic job from my perspective :)

More to the point, they're not in desperate need of extra help
(although I'm sure they wouldn't say no to it).

It seems to me that we need more developers fixing bugs, not more
people trying to find bugs. We're already finding more bugs than we
fix.

Just my two cents,
Evan

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Proposal to delay release of Precise Pangolin

[Jonathan Carter (highvoltage)]

Hi Joseph

On 11-10-19 12:21 AM, Joseph Toppi wrote:
> Because no one else seemed willing to check, compact view does remove the
> needless amount of margin, but also switches to a more list-like look and
> changes the scrolling to horizontal. I checked in Nautilus 2.32.2.1 the
> version that ships with 11.04 with all updates applied.
> 
> What kind of QA process is there before a release, how can I help with that?

The QA team can always use some more hands, there's even a dedicated QA
site where you can get all the information you need: http://qa.ubuntu.com/

It seems that the Desktop testing program is of particular interest to you:
http://qa.ubuntu.com/testing/desktop-testing-program/

-Jonathan

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

RE: Proposal to delay release of Precise Pangolin

[nick rundy]

I appreciate your e-mail, Martin :) 

>  * These types of bugs are too big/complex for quick patches and too
> small or unimportant for critical attention.

This is what I'm getting at. I don't doubt this is true. However, fixing this 
stuff is what's going to make a critical difference in users coming to, 
enjoying, and staying with Ubuntu. Put the "rapid release" schedule on hold 
temporarily and make an LTS that fixes this stuff. Then the LTS can last a long 
time as the face of Ubuntu. The status quo has become new releases perpetuating 
old bugs that are years old. The LTS releases up to this point are better than 
the 6-months but they still contain these bugs. I'm proposing that if bug #1 is 
going to be "Fix Released," the current full speed ahead rapid-release approach 
has to at least take a break for a cycle and address this stuff. NOW is an 
appropriate time because of Unity and GNOME 3.2. There's a lot of stuff that 
needs fixing in Unity and GNOME 3.2.

>  *The gnome programmer deals
> with bugs as he feels like it and expects patches.

Is there anything to prevent Ubuntu developers from saying, "Hey, is it okay if 
I take this and fix this bug?"


>  * No one answered the question 'did you try compact layout'

Compact Layout works to some extent. But what about use scenarios where the 
user needs the increased zoom only available in Icon View?

> I understand your point Nick, I'd really like a cycle that focuses
> _only_ on bug fixing and nothing else. But I'd also like a cycle that
> took everyone off coding to train a 100 new kernel hackers and 50 new
> xorg slaves.

Perhaps this should be done? Delay the release of Precise Pangolin and really 
refine it. Then miss a 6-month release or two and spend the time training 100 
new kernel hackers and 50 new xorg slaves. With a strong, refined LTS, Ubuntu 
will be fine for 2-3 years. To use an analogy from athletics: when you've been 
training hard for a long time, taking some time off will help you make gains 
because it gives the body time to grow from the training. Let Ubuntu grow into 
Unity and GNOME 3.2 bug-free. Unity was an awesome accomplishment IMHO. I'm 
proposing that NOW is the time to pull it all together by making an 
ultra-refined bug-free LTS product that can truly begin to tackle Bug #1. It's 
going to be an LTS that takes on Windows, not one of the 6-month releases. 



> Subject: Re: Proposal to delay release of Precise Pangolin
> From: docto...@gmail.com
> To: nru...@hotmail.com
> CC: ubuntu-devel-discuss@lists.ubuntu.com
> Date: Tue, 18 Oct 2011 23:27:01 -0400
> 
> On Tue, 2011-10-18 at 22:15 -0400, nick rundy wrote:
> > Yet the bug has existed for more than 3 years. Sadly, the same can be
> > said for many other bugs.
> 
> To be fair to the bug:
> 
>  * No one answered the question 'did you try compact layout'
>  * Nautilus is a 'special' codebase which I wouldn't want to touch again
> this side of the 21st century, ugly and duplicative spaghetti.
>  * Anything to do with how something looks, workflow or speed is not
> going to get fixed by the fire fighters or cathedral builders. 
>  * These types of bugs are too big/complex for quick patches and too
> small or unimportant for critical attention.
>  * No user continues to pay for bug fixes, no economics and no other
> relationship between programmer and user. The gnome programmer deals
> with bugs as he feels like it and expects patches.
> 
> I understand your point Nick, I'd really like a cycle that focuses
> _only_ on bug fixing and nothing else. But I'd also like a cycle that
> took everyone off coding to train a 100 new kernel hackers and 50 new
> xorg slaves.
> 
> If wishes could be put in dishes the world would be delicious.
> 
> Best Regards, Martin Owens
> 
  -- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss