Re: [gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Sebastian Pipping
On 06/09/2011 03:37 PM, Rich Freeman wrote:
 do we need some kind of policy around membership on special
 project teams. QA and Devrel are the most obvious examples, Infra might
 be another.

in my eyes we do.  too much power to be unregulated.

what does it take to get this rolling?



sebastian



Re: [gentoo-dev] Reviving webapp-config

2011-06-10 Thread Sebastian Pipping
Questions:

 - What does reviving mean in detail?
 A re-write?  A somewhat compatible re-write?
 Getting back to maintaining the current code?
 Why did you choose how you did?

 - Have you spoken to Andreas Nüsslein who worked on a
   re-write in context of an earlier GSoC?

Best,




Sebastian



Re: [gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Dane Smith
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 06/10/11 07:44, Sebastian Pipping wrote:
 On 06/09/2011 03:37 PM, Rich Freeman wrote:
 do we need some kind of policy around membership on special
 project teams. QA and Devrel are the most obvious examples, Infra might
 be another.
 
 in my eyes we do.  too much power to be unregulated.
 
 what does it take to get this rolling?

Part of me thinks this is a good idea for the simple reason that some
people seem to have issues with QA/DevRel. Perhaps if the lead were
appointed by council there would be less raging every time the team
tried to do anything. But then again, we can't all play nice over a
freaking ChangeLog, even with Council making the call, so I somehow
doubt this will fix *anything at all*

I do not want to see elections for 3 team leads. As far as I'm
concerned, it's way too much of a hassle. Having said that, council
appointments for the team lead may not be a bad idea. Maybe it will
work. I doubt it, but I'm open to try it. My fear there is I don't want
to see team leads changing every year just because there is a new
council. If the team is working well together, there is no sense in
fubaring that merely because we can.

Perhaps do council appointments if the lead steps down / if the team
calls for a re-appointment (there would need to be rules for this part.
I don't want to see a new appointee merely because the lead upset one
person. Perhaps if more than 50% of the team or like 10 other developers
are asking for a new lead or some such foo.)

Also, while I like the idea of cleaning out those teams once in a
while for inactive members, I'm not a huge fan of a new lead coming in
and removing people from the team just because.

Lastly, given that it will be the lead for a given team, I think that
team should have the ability to pick their candidates to go to
council, and maybe just give Council the vote on who gets it. Or, have
council appoint people they think are fit, and the team can vote from
there. Either way I think would work alright.

Just my 2 cents.
Regards,
- -- 
Dane Smith (c1pher)
Gentoo Linux Developer -- QA / Crypto / Sunrise / x86
RSA Key: http://pgp.mit.edu:11371/pks/lookup?search=0x0C2E1531op=index
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.17 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=8ajo
-END PGP SIGNATURE-



Re: [gentoo-dev] Reviving webapp-config

2011-06-10 Thread Jesus Rivero (Neurogeek)
On Fri, Jun 10, 2011 at 7:49 AM, Sebastian Pipping sp...@gentoo.org wrote:
 Questions:

  - What does reviving mean in detail?
     A re-write?  A somewhat compatible re-write?
     Getting back to maintaining the current code?
     Why did you choose how you did?

  - Have you spoken to Andreas Nüsslein who worked on a
   re-write in context of an earlier GSoC?


IIRC, last year's GSoC featured a project involving the rewrite of
webapp-config. Andreas was the student, but I don't recall right now
who the mentor was. Maybe this would be a good start!

 Best,




 Sebastian



Best regards,

-- 
Jesus Rivero (Neurogeek)
Gentoo Developer



Re: [gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Francisco Blas Izquierdo Riera (klondike)
I was thinking of writting this in private, but I bet it will do more
good if I do it public.

I'm 22 (most of you could call me a kid) and a reasonably recent new
developer and I'm sad having to ask you, am I the only one seeing
childishness on your actions, and this yours implies at least Samuli,
Mike and Diego and probably many others.

You are discussing and reveling for a stupid file, even worse, its not
even code and solutions to automate the process have been proposed so
you are discussing over nothing.

If you guys want to rebel I can give you many good reasons which are
meaningful that a few lines of something that's not even code:
* Mike your country considers freedom of speech a restrictable right,
maybe you should fight and rebel against that instead of a stupid file.
* Samuli, extremist right wing parties are gaining power in your
country, I think this is a way better reason to rebel than a stupid file.
* Diego, Berlusconi a way better reason to be outraged I think.

I know this will serve for nothing you are going to keep discussing over
pride? 10 minutes of your live? instead of fighting what really should
matter you. It is up to you, meanwhile I'll keep fighting for the camped
people in Spain instead of some random piece of documentation.

Have a nice day.



signature.asc
Description: OpenPGP digital signature


Re: [gentoo-dev] Reviving webapp-config

2011-06-10 Thread Matthew Summers
On Fri, Jun 10, 2011 at 6:49 AM, Sebastian Pipping sp...@gentoo.org wrote:
 Questions:

  - What does reviving mean in detail?
     A re-write?  A somewhat compatible re-write?

I am not necessarily interested in a complete re-write unless its
really warranted due to support for new features. First things first,
webapp-config needs to have the open bugs addressed and current
functionality should be supported.

     Getting back to maintaining the current code?

There is some good code in there, so I think at minimum some of that
will be maintained moving forward. Regardless, the current codebase
should be audited, bugs fixed, and a new design spec incorporating the
new features desired needs to be written in collaboration with various
stake holders.


     Why did you choose how you did?

I do not understand this sentence, but will try to explain the choices
a bit. Having used webapp-config for 7 years or so, I had grown fond
of it for managing things like drupal (back in the day) and for
handling static media for my company projects. As time progressed, it
seemed reasonable to extend the functionality of w-c to do fancier
things like change tracking or byte-compiling python modules outside
of site-packages for multi-instance and potentially multi-versioned
deployments of the various pythonic webapps I manage. It seems
reasonable to roll this sort of functionality into the existing tool.
I am not intending to make this something specific to python webapps,
just to be clear, but include tools to handle webapps written in
ruby/rails, perl, etc.

Additionally, Gentoo's infra team uses w-c for a few things and its a
nice tool that mostly integrates well into cfengine (afaik). Whatever
the case with automation, it does make management tasks easier
(perhaps IMO).

Further, there are substantial applications, like Moodle, that would
benefit from a more robust deployment toolkit within Gentoo.

So, in general, from both a distribution and a professional
perspective its quite clear (at least to me), that this tool has an
important role in Gentoo and therefore needs to be revived.


  - Have you spoken to Andreas Nüsslein who worked on a
   re-write in context of an earlier GSoC?

I have not. I was unaware of this project until it was mentioned in
replies to this thread. I have started reading the code. Perhaps there
are some elements of Andreas' code we can incorporate into the w-c
codebase. I need to dig into the code far more than my cursory glance.


 Best,




 Sebastian



Good questions Sebastian, thanks.

Matthew W. Summers
Gentoo Foundation Inc.



Re: [gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Francisco Blas Izquierdo Riera (klondike)
El 10/06/11 17:33, Francisco Blas Izquierdo Riera (klondike) escribió:
 * Diego, Berlusconi a way better reason to be outraged I think.
Small clarification here: I'm not comparing Diego with Berlusconi AFAIK
he isn't a corrupt underage fucking politician, I'm pointing him
Berlusconi ruling Italy is a quite good reason to fight against compared
to adding some lines in a file.



signature.asc
Description: OpenPGP digital signature


Re: [gentoo-dev] Reviving webapp-config

2011-06-10 Thread Sebastian Pipping
On 06/10/2011 05:38 PM, Matthew Summers wrote:
 Why did you choose how you did?
 
 I do not understand this sentence, 

I intended to write as you did, sorry.  If that's still bad English: I
wanted to hear about your rationale, which you have explained by now.
Thanks.


 [..] this tool has an
 important role in Gentoo and therefore needs to be revived.

I wished people were thinking like that about genkernel :-)

Best,



Sebastian



Re: [gentoo-dev] ChangeLog generation - pros and cons (council discussion request)

2011-06-10 Thread Fabian Groffen
On 09-06-2011 18:06:02 +0200, Andreas K. Huettel wrote:
   If typos matter then they matter to everybody, and if they don't then
   we should not care.  QA in Gentoo should be a consistent experience.
  
  while the last sentence is true, the first is not.  if a minority of
  people care about typos, and/or they rarely fix said typos, then the
  logical answer is that their opinion loses out.  it doesnt mean that
  everyone agrees.
  -mike
 
 you will surely agree then- if a minority of people thinks ebuild
 removals should not go into the changelog, their opinion just loses
 out. Right?
 
 Anyway. You and Samuli are at the moment just wasting everyone's time
 here. 

Please don't mix threads.  This thread is just about what we should do
with ChangeLogs (if, and if, how to generate them) not about a
particular preference of documenting things or not.

So far, this thread was reasonably made of constructive, and sort of
objective replies.  It would be nice, and probably most useful for the
council, if we could keep it this way.



-- 
Fabian Groffen
Gentoo on a different level



Re: [gentoo-dev] profiles/ChangeLog is too large

2011-06-10 Thread Fabian Groffen
On 09-06-2011 21:39:43 +0200, Ulrich Mueller wrote:
 The profiles/ChangeLog file has grown to more than 700 kbytes. Would
 it make sense to remove ancient (say, pre-2010) entries from it?

Perhaps you can generalise the discussion into when you do consider a
changelog entry to (still) be relevant?
I guess figuring out something here is useful both now as well as in the
future.


-- 
Fabian Groffen
Gentoo on a different level



Re: [gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Rich Freeman
On Fri, Jun 10, 2011 at 9:05 AM, Dane Smith c1p...@gentoo.org wrote:
 Perhaps do council appointments if the lead steps down / if the team
 calls for a re-appointment (there would need to be rules for this part.
 I don't want to see a new appointee merely because the lead upset one
 person. Perhaps if more than 50% of the team or like 10 other developers
 are asking for a new lead or some such foo.)

 ...

 Lastly, given that it will be the lead for a given team, I think that
 team should have the ability to pick their candidates to go to
 council, and maybe just give Council the vote on who gets it. Or, have
 council appoint people they think are fit, and the team can vote from
 there. Either way I think would work alright.

I also dislike the general election idea, for the reasons you state.

I think the ideal process is something like:

1.  Teams put forth recommendations for who THEY would like to see as
the lead, perhaps with more than one choice.
2.  The Council is free to pick any lead they like, and change that
lead any time they like.
3.  However, the Council is encouraged that unless there is a big
reason not to do so, they just accept or choose from the team's
nominations, and only do so annually.

I don't like the concept of the council only getting to ratify a
decision already made by the team.  This will just lead to more
bickering on the lists about the wrong people being on the team or
whatever and the fox being in charge of the henhouse or whatever.

The Council has a mandate, because they are elected.  You can disagree
with the Council, but you can't argue that their decisions don't have
SOME kind of backing simply because they have been selected by the dev
community as a whole.  By giving the Council ultimate authority (and
accountability) that mandate then is conferred upon the team leads for
QA, Devrel, etc.

This is not unlike how any business or similar concern is run.  Teams
usually know best how they should be run, but they still fall under
the board or whatever and as long as they're doing a good job boards
generally just rubber-stamp their recommendations.  When things go
wrong, then the board takes a more active role, even to the point of
completely overriding the team if that is what it takes to fix things
- but usually they just put somebody in charge that they feel will
handle things.

Government isn't a good example as it tends to be dominated by
cronyism, and I think there is general agreement that this is NOT how
we want things to work.  The council should not generally fiddle with
every little thing QA does, or whatever, but they can step in when the
issue is serious.

Rich



Re: [gentoo-dev] Gentoo package statistics -- GSoC 2011

2011-06-10 Thread Sebastian Pipping
On 06/08/2011 04:36 PM, Vikraman wrote:
 * Repository, Keyword, Useflags (plus,minus,unset), Counter, Size,
   and Build   time for each installed package

How many operations do you expect for a submissions with 1000 packages
on SQL level?  Will that be around 1000 inserts?

Best,



Sebastian



[gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Markos Chandras
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512

On 06/09/2011 08:54 PM, Piotr Jaroszyński wrote:
 On 9 June 2011 15:44, Diego Elio Pettenò flamee...@gmail.com wrote:
 Given the lead is held responsible for the behaviour of the team's
 member in respect to the QA work, I don't think it is unexpected of the
 lead to remove those people who have shown no intention to collaborate.

 Beside, I warned both of you that you had to follow policy or get out,
 neither of you even _bothered_ replying to my request.
 
 On 04/18/2010 03:31 PM, Diego Elio “Flameeyes” Pettenò wrote:
 Diego, you were nominated as well.  Do you accept?

 Muahhahaha — no I don't think it would be a good idea. It would almost
 certainly end up with a pissing-off contest between me and council or me
 and devrel depending on their position
 
 Guess you were right after all.
 

I am sorry but this is not a way for a leader to treat the members of
his team. I am retiring myself from QA as well. Do note that 4 members
have already gone from QA. This cannot be a coincidence.

- -- 
Regards,
Markos Chandras / Gentoo Linux Developer / Key ID: B4AFF2C2
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.17 (GNU/Linux)
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=xzyy
-END PGP SIGNATURE-



Re: [gentoo-dev] Gentoo package statistics -- GSoC 2011

2011-06-10 Thread Vikraman
On Sat, Jun 11, 2011 at 01:10:36AM +0200, Sebastian Pipping wrote:
 On 06/08/2011 04:36 PM, Vikraman wrote:
  * Repository, Keyword, Useflags (plus,minus,unset), Counter, Size,
and Build time for each installed package
 
 How many operations do you expect for a submissions with 1000 packages
 on SQL level?  Will that be around 1000 inserts?
 

One insert for each package entry, and one insert for every useflag.

 Best,
 
 
 
 Sebastian
 

-- 
Vikraman


signature.asc
Description: PGP signature


[gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Diego Elio Pettenò
Il giorno sab, 11/06/2011 alle 01.48 +0100, Markos Chandras ha scritto:
 
 I am sorry but this is not a way for a leader to treat the members of
 his team. I am retiring myself from QA as well. Do note that 4 members
 have already gone from QA. This cannot be a coincidence.
 

For those who wouldn't get qa@g.o mail, Tomáš has also requested
retirement from qa, in
http://bugs.gentoo.org/show_bug.cgi?id=368097#c66 :

 Markos for one you and vapier are the reason why i left the team because
 otherwise i would have to kick both of you out.

So I'd like to know how you can even pretend to count four members when:

 - one retired because other members of the team decided to do as they
like;
 - two has been kicked out for playing along just if they can also make
the rules (as soon as a rule was enacted that they didn't like they
decided to ignore it, even under request to either not do so or be
removed from QA);
 - you decided to retire because I applied a warning I had sent in
advance (on April 30th).

Or maybe are you counting Sven who I haven't heard from in a year or
two, and whose autorepoman mail has stopped coming ... probably before
you ever became a developer?

Maybe I scream in private, but what you three (keeping Tomáš out of
this) are doing is crying in public because you're no longer allowed to
poop in the sandbox you should keep clean.

Do note that it was even your words:

 I am sorry but having elections every few months is not a solution.
 First we need to clean up the team, then become team, then have
 elections.

Which is exactly what I'm going to do: I'm going to make sure that the
team is on the same page: policies has to be followed, or they need to
be changed. Which doesn't look like either of them (nor you I guess)
want to do. I'm pretty sure I didn't ask much to the team beside
actually following what the council decided.

Finally, I'd like to point out that neither my character nor my actions
have changed the slightest since the mail that Peper quoted — yet I was
elected as team leader; it looks like though people wanted me to scream
at anyone else beside them — too easy that way.

And the only two people in the team who bothered to cast a vote (Luca
and Christian), seems not to have an issue with me keeping this way.

So, this might hurt your feelings, but I'm not really sorry to see you
leave. As I said before I had been disappointed when people I had a high
esteem of decided that rules shouldn't apply to them.

-- 
Diego Elio Pettenò — Flameeyes
http://blog.flameeyes.eu/




Re: [gentoo-dev] Re: [gentoo-commits] gentoo commit in xml/htdocs/proj/en/qa: index.xml

2011-06-10 Thread Ulrich Mueller
 On Fri, 10 Jun 2011, Donnie Berkholz wrote:

 I like the idea of appointed leads instead of holding elections, it
 fits into my big picture of less bureaucracy and more meritocracy
 (see my email re running for council on -project). Appointments
 would be made by the next level up from the lead. So project leads
 would come from the council, subproject leads from the project, etc.

What problem are you trying to solve?

There maybe some point doing this for projects that have elevated
powers (like QA or devrel), but I think for normal projects our
current system works well enough.

Ulrich