Re: 2588 - Journal unusable

2008-12-12 Thread Greg Smith
Hi All,

This is a great thread! Very respectful but on point and addressing a 
core concern which needs to become a core competency.

Mikus, James, Gary and the other lead developers who pull down joyride 
regularly are critical to the success of the next release. They proved 
it in the last release.

I agree with James suggestion to get people to test new code in a 
private stream before they put it in joyride. Whether that can be done 
or not, we need to be more clear about when Mikus and the cutting edge 
team should try out the latest version.

There will be bumps along the way, miscommunication, lost time and 
wasted bandwidth. You can't make an omelette without breaking eggs.

As long as we continuously improve and we respect each others time and 
input, we'll get there.

This is open source at its best and we have to become great at it for 
the success of the project.

We're off to a good start but we need to see continual improvement on 
communicating status and quality of Joyride from now until release.

85 days until we send XO Software Release 9.1.0 to manufacturing!

Thanks,

Greg S

**
Date: Fri, 12 Dec 2008 15:27:21 +1100
From: James Cameron qu...@laptop.org
Subject: Re: 2588 - Journal unusable
To: Chris Ball c...@laptop.org
Cc: devel@lists.laptop.org
Message-ID: 20081212042721.gg6...@us.netrek.org
Content-Type: text/plain; charset=us-ascii

G'day Chris,

I'll give a partial line of reasoning response ... this is not complete,
I'm short of time.

Where I said public, I meant developer builds that can be used by other
developers.  I didn't mean to imply public builds for testing by
non-developers.  I mean the difference between what a developer does and
what a developer releases.  That isn't only OLPC originated code, that's
also the choice of what RPMs to accept from outside.  Accepting lots of
RPMs at once is the same as making lots of code change.

Why don't you have private build streams?  That's what I can do with
debxo, for instance ... build on my desktop, test on an XO, and then
avoid releasing anything to the public until I've verified that what
I've changed actually works.

Why can't the build system be replicated so that each developer can test
their change before releasing it?  What is it about the build system that
prevents it?  I thought the build system was just a set of downloads and
put-it-together processes.

-- James Cameron mailto:qu...@us.netrek.org http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-12 Thread Michael Stone
On Fri, Dec 12, 2008 at 03:27:21PM +1100, James Cameron wrote:
G'day Chris,

I'll give a partial line of reasoning response ... this is not complete,
I'm short of time.

Where I said public, I meant developer builds that can be used by other
developers.  I didn't mean to imply public builds for testing by
non-developers.  I mean the difference between what a developer does and
what a developer releases.  That isn't only OLPC originated code, that's
also the choice of what RPMs to accept from outside.  Accepting lots of
RPMs at once is the same as making lots of code change.

Why don't you have private build streams?  That's what I can do with
debxo, for instance ... build on my desktop, test on an XO, and then
avoid releasing anything to the public until I've verified that what
I've changed actually works.

Why can't the build system be replicated so that each developer can test
their change before releasing it?  What is it about the build system that
prevents it?  I thought the build system was just a set of downloads and
put-it-together processes.

It can, it has been, and no one seems to care. See 

   http://wiki.laptop.org/go/Build_system
   http://wiki.laptop.org/go/Building_custom_images
   http://wiki.laptop.org/go/Puritan
   
http://wiki.laptop.org/go/User:Mstone/Commentaries/Releases_2#Practical_Matters

for some historical records of the discussions.

Michael
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread Daniel Drake
On Thu, Dec 11, 2008 at 5:06 AM, Mikus Grinbergs [EMAIL PROTECTED] wrote:
 The Journal entries do not show the object icon - so they cannot be
 launched from.

 This SEVERE error has now lasted through five Joyride builds.

joyride builds are automatic. 5 builds is about 1 day. Please be a
little more patient.

I assume from your language that this is a regression, i.e. it worked
six joyride builds ago? This wasn't clear from your original mail, but
is valuable info to have. Can you recall the most recent working
version?

 I'm posting here, rather than write a ticket, as a way to get
 quickest to those who can do something about it.  Or should users
 who have real XOs (rather than emulators) stop participating ?

I'd suggest tickets for all bugs because they are harder to get lost
(and soon will be triaged into appropriate priority). Then, for the
important ones, feel free to raise extra attention on the mailing
list.

But even if you do both, you may have to wait more than 1 day for a fix.

cheers,
Daniel
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread Mikus Grinbergs
 I assume from your language that this is a regression, i.e. it worked
 six joyride builds ago? This wasn't clear from your original mail, but
 is valuable info to have. Can you recall the most recent working
 version?

The last time the Joyride Journal did not have the problem was build 
2581 from 12/07.  The problem showed up in build 2583 from 12/09. 
(Building of 2582 failed, without producing an image.)

mikus

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread Mikus Grinbergs
 I am just making a new snapshot release of sugar since there were some
 deps of sugar-toolkit on the sugar package. What is helpful when writing
 those emails is always to have a quick look at the logs.
 
 For the impatient ones:
 http://koji.fedoraproject.org/koji/taskinfo?taskID=993286

Disclaimer:  META-discussion follows

Putting 'patience' into context:  For an example of me getting 
impatient - having Journal unexpectedly stop working in Joyride 
(with no one seeming to care).  For an example of me remaining 
patient, this list (of where F9 is apparently at a later level than 
F10) has been the output of Joyride's 'yum check-update' for ages:

 iproute.i386 2.6.27-1.fc9 
 nspr.i3864.7.3-1.1.fc9
 pyabiword.i386   0.6.1-4.olpc3
 python-telepathy.noarch  0.15.3-1.olpc3   
 sugar-artwork.i386   0.83.1-2.olpc3   
 sugar-base.i386  0.83.1-2.olpc3   
 sugar-journal.noarch 100-1.olpc3  
 xorg-x11-drv-evdev.i386  2.0.8-1.fc9  


It is likely that the regression in Journal was caused by releasing 
sugar-toolkit without simultaneously releasing something which that 
package interacted with.  But how is the person who gets the Joyride 
build to know what he needs to wait for?  For taskID=993286 ??

mikus

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread Eben Eliason
On Thu, Dec 11, 2008 at 3:57 PM, Mikus Grinbergs mi...@bga.com wrote:
 I am just making a new snapshot release of sugar since there were some
 deps of sugar-toolkit on the sugar package. What is helpful when writing
 those emails is always to have a quick look at the logs.

 For the impatient ones:
 http://koji.fedoraproject.org/koji/taskinfo?taskID=993286

 Disclaimer:  META-discussion follows

 Putting 'patience' into context:  For an example of me getting
 impatient - having Journal unexpectedly stop working in Joyride
 (with no one seeming to care).  For an example of me remaining
 patient, this list (of where F9 is apparently at a later level than
 F10) has been the output of Joyride's 'yum check-update' for ages:

 iproute.i386 2.6.27-1.fc9
 nspr.i3864.7.3-1.1.fc9
 pyabiword.i386   0.6.1-4.olpc3
 python-telepathy.noarch  0.15.3-1.olpc3
 sugar-artwork.i386   0.83.1-2.olpc3
 sugar-base.i386  0.83.1-2.olpc3
 sugar-journal.noarch 100-1.olpc3
 xorg-x11-drv-evdev.i386  2.0.8-1.fc9


 It is likely that the regression in Journal was caused by releasing
 sugar-toolkit without simultaneously releasing something which that
 package interacted with.  But how is the person who gets the Joyride
 build to know what he needs to wait for?  For taskID=993286 ??

Well, there are a few factors here:

1) Joyide is the development build stream.  It doesn't by any means
promise stability, in general.
2) It's early in the release cycle, so things are even more likely to
break in fairly big ways.
3) As was mentioned, a few days to a week is really quite short, all
things considered.

We don't suggest that people who depend on day to day use of their XO
use the joyride builds all the time. However, we certainly appreciate
everyone who does help test it, since it's clearly best to catch any
and all regressions as early as possible, as large or small as they
might be.  For that, I'm sure we all thank you!

Of course, writing a ticket is /always/ the correct thing to do in
such circumstances, so that we can adequately triage the bug, and
ensure it doesn't get lost. Even if we don't get it fixed within a day
or two, this will ensure that it does get fixed. Thanks!

- Eben


 mikus

 ___
 Devel mailing list
 Devel@lists.laptop.org
 http://lists.laptop.org/listinfo/devel

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


2588 - Journal unusable

2008-12-11 Thread Mikus Grinbergs
With 2590 the Joyride Journal problem of entries not displaying an 
icon (so those Activities can be re-launched) has been corrected.

Thank you,  mikus




___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread david
On Thu, 11 Dec 2008, Eben Eliason wrote:

 On Thu, Dec 11, 2008 at 3:57 PM, Mikus Grinbergs mi...@bga.com wrote:

 It is likely that the regression in Journal was caused by releasing
 sugar-toolkit without simultaneously releasing something which that
 package interacted with.  But how is the person who gets the Joyride
 build to know what he needs to wait for?  For taskID=993286 ??

 Well, there are a few factors here:

 1) Joyide is the development build stream.  It doesn't by any means
 promise stability, in general.
 2) It's early in the release cycle, so things are even more likely to
 break in fairly big ways.
 3) As was mentioned, a few days to a week is really quite short, all
 things considered.

 We don't suggest that people who depend on day to day use of their XO
 use the joyride builds all the time. However, we certainly appreciate
 everyone who does help test it, since it's clearly best to catch any
 and all regressions as early as possible, as large or small as they
 might be.  For that, I'm sure we all thank you!

when I was reporting errors a few weeks ago I got the response we're 
early in the process, don't bother to report anything until we have it 
stabilized a bit, we'll tell you when to start testing. I asked within 
the last week or so if it was time to start testing yet and was told no.

David Lang

 Of course, writing a ticket is /always/ the correct thing to do in
 such circumstances, so that we can adequately triage the bug, and
 ensure it doesn't get lost. Even if we don't get it fixed within a day
 or two, this will ensure that it does get fixed. Thanks!

 - Eben


 mikus

 ___
 Devel mailing list
 Devel@lists.laptop.org
 http://lists.laptop.org/listinfo/devel

 ___
 Devel mailing list
 Devel@lists.laptop.org
 http://lists.laptop.org/listinfo/devel

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread quozl
On Thu, Dec 11, 2008 at 04:24:32PM -0500, Eben Eliason wrote:
 1) Joyide is the development build stream.  It doesn't by any means
 promise stability, in general.
 2) It's early in the release cycle, so things are even more likely to
 break in fairly big ways.

It is as if joyride is being used as part of developers' edit, compile
and test sequence.  Wouldn't it be more appropriate to do this sequence
privately before a developer releases their changes to the public?

Here's how it would happen ... the developer would build a new RPM,
install it on a test unit, test that it works, *before* publishing the
RPM for the joyride to pick up.  If any failures occur due to
integration against other developer changes, then stronger dependencies
should be added.

I'm also waiting for the joyride builds to be usable before I consume
download data resources on them.

-- 
James Cameronmailto:qu...@us.netrek.org http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread Chris Ball
Hi James,

Thanks for these thoughts.  They seem like worthwhile goals, but I don't
think we're likely to get there in practise.  In specific:

It is as if joyride is being used as part of developers' edit,
compile and test sequence.  Wouldn't it be more appropriate to do
this sequence privately before a developer releases their changes
to the public?

So, first of all, this developers - public distinction is bogus;
these aren't public builds, they're developer builds, and the only
reason this seems odd is that most companies aren't doing their
development in public.  We _do not have_ a private build stream to use
to stage builds in before letting you see them, and we don't want one.

Here's how it would happen ... the developer would build a new RPM,
install it on a test unit, test that it works, *before* publishing
the RPM for the joyride to pick up.  If any failures occur due to
integration against other developer changes, then stronger
dependencies should be added.

Many of the people supplying RPMs that we end up using in Joyride are
not writing them for the XO -- they might be Fedora updates, or even
upstream Sugar changes that are tested under jhbuild and not the XO.
We're going to have to deal with this as part of accepting that we
(OLPC-the-company) don't write 90% of the code we ship, and don't
have influence over the people who do write it, most of whom we'll
never meet.

Finally, I think it's a little unfair (but not in an offended kind of
way) to expect us to have stable or even working Joyride builds at the
moment: we just moved to Fedora 10, and it's always going to be the case
that a bunch of stuff breaks and you have to go through the breakage
piece by piece; this isn't a case where we'd have a working build if
only some developer had tested her changes before throwing them over
the wall into Joyride.

I'm also waiting for the joyride builds to be usable before I
consume download data resources on them.

All that said, the desire to *have* a working build is a very reasonable
thing, and that's why we're trying hard not to trick anyone into
thinking that Joyride contains that build.  If you have the time and
bandwidth to help shake out the bugs, we'd love your help; if you don't,
that's fine too and you should stay on 8.2.x for another few weeks.

Hope that helps!

- Chris.
-- 
Chris Ball   c...@laptop.org
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-11 Thread James Cameron
G'day Chris,

I'll give a partial line of reasoning response ... this is not complete,
I'm short of time.

Where I said public, I meant developer builds that can be used by other
developers.  I didn't mean to imply public builds for testing by
non-developers.  I mean the difference between what a developer does and
what a developer releases.  That isn't only OLPC originated code, that's
also the choice of what RPMs to accept from outside.  Accepting lots of
RPMs at once is the same as making lots of code change.

Why don't you have private build streams?  That's what I can do with
debxo, for instance ... build on my desktop, test on an XO, and then
avoid releasing anything to the public until I've verified that what
I've changed actually works.

Why can't the build system be replicated so that each developer can test
their change before releasing it?  What is it about the build system that
prevents it?  I thought the build system was just a set of downloads and
put-it-together processes.

-- 
James Cameronmailto:qu...@us.netrek.org http://quozl.netrek.org/
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


2588 - Journal unusable

2008-12-10 Thread Mikus Grinbergs
The Journal entries do not show the object icon - so they cannot be
launched from.

This SEVERE error has now lasted through five Joyride builds.

I'm posting here, rather than write a ticket, as a way to get 
quickest to those who can do something about it.  Or should users 
who have real XOs (rather than emulators) stop participating ?

mikus



___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel


Re: 2588 - Journal unusable

2008-12-10 Thread david
On Thu, 11 Dec 2008, Mikus Grinbergs wrote:

 The Journal entries do not show the object icon - so they cannot be
 launched from.

 This SEVERE error has now lasted through five Joyride builds.

 I'm posting here, rather than write a ticket, as a way to get
 quickest to those who can do something about it.  Or should users
 who have real XOs (rather than emulators) stop participating ?

we were told to stop participating a dozen or so joyrides ago, they will 
tell us when we should start testing again.

David Lang
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel