Обсуждение: Contributor listing policy
Folks, *speaking* of being open about policies, I'm requesting comments on the Core Team's draft contributor listing policy: http://developer.postgresql.org/index.php/ContributorListings Here's what's changing from current practice: 1) period for dropping people who have not been around is set to 2 years. No explicit period was set before. 2) non-code contributors now get listed in the "Contributors" section, but are not eligible for the "Major Developers" section. This is a compromise between previous practice (not listing non-code contributors at all) and what some people would like to see ("Major Contributors" with non-code contributors); I figure we'll revisit this policy in a year or so. 3) this will require a member of the WWW team to add "year contributed" and "comments" non-displaying fields to the web tool we manage the contributor listings with. How long would that take to add? -- --Josh Josh Berkus PostgreSQL @ Sun San Francisco
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 7 Mar 2008 15:14:08 -0800 Josh Berkus <josh@agliodbs.com> wrote: > Folks, > > *speaking* of being open about policies, I'm requesting comments on > the Core Team's draft contributor listing policy: > http://developer.postgresql.org/index.php/ContributorListings > > Here's what's changing from current practice: > 1) period for dropping people who have not been around is set to 2 > years. No explicit period was set before. > > 2) non-code contributors now get listed in the "Contributors" > section, but are not eligible for the "Major Developers" section. There is no major developer section. There are: Core Major Contributors Contributors This particular change was discussed and debated publicly and the patch submitted by me and approved, months ago. It is also why the contributor list is not under developers anymore. It's under community. > This is a compromise between previous practice (not listing non-code > contributors at all) and what some people would like to see ("Major > Contributors" with non-code contributors); I figure we'll revisit > this policy in a year or so. The idea that a line of code is more important than the organization of an army (users) is incorrect. Without one the other is pointless. > > 3) this will require a member of the WWW team to add "year > contributed" and "comments" non-displaying fields to the web tool we > manage the contributor listings with. How long would that take to > add? > Not long. I could do it fairly easily. Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL political pundit | Mocker of Dolphins -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH0fKIATb/zqfZUUQRAohyAJ9tIog12RxEBBlaXQjEs88+26mtKgCfURZZ EwlGK3wCOda9STjQwZzsNvY= =E7I7 -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 7 Mar 2008 17:57:25 -0800 "Joshua D. Drake" <jd@commandprompt.com> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Fri, 7 Mar 2008 15:14:08 -0800 > Josh Berkus <josh@agliodbs.com> wrote: > > > Folks, > > > > *speaking* of being open about policies, I'm requesting comments on > > the Core Team's draft contributor listing policy: > > http://developer.postgresql.org/index.php/ContributorListings > > > > Here's what's changing from current practice: > > 1) period for dropping people who have not been around is set to 2 > > years. No explicit period was set before. > > > > 2) non-code contributors now get listed in the "Contributors" > > section, but are not eligible for the "Major Developers" section. Of note on this is: Editing the contributor listings will be carried out by the Core Team. This is outside the scope of core purpose as they portray themselves. Which is as a steering committee/release manager. Although members of core could certainly be part of the process, I think that like everything else in the infrastructure contributor management should be lead by a team of individuals, a committee if you will. Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL political pundit | Mocker of Dolphins -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH0gehATb/zqfZUUQRAkuhAJoD8hFH11fPqkWE8ytKbzr1cuT77gCfZlm7 lx0TjDJB6ZYzAlt1+qx0uKQ= =L3aN -----END PGP SIGNATURE-----
Josh,
> This particular change was discussed and debated publicly and the patch
> submitted by me and approved, months ago. It is also why the
> contributor list is not under developers anymore. It's under community.
Actually, looking at the archives, there wasn't much discussion when you
submitted the patch; I doubt that most people realized the changes it made.
I know that I wasn't clear on it, noticing only the change in sort order, and
certainly the rest of the core team (or hackers) didn't discuss it.
Contributor listings are not a matter of *only* WWW team's discretion. In
fact, traditionally, contents of the developer listings have been determined
by the core team; if we're going to change that and give sole authority to
WWW maintainers, then I think we ought to have an *explicit* discussion about
that.
> > This is a compromise between previous practice (not listing non-code
> > contributors at all) and what some people would like to see ("Major
> > Contributors" with non-code contributors); I figure we'll revisit
> > this policy in a year or so.
>
> The idea that a line of code is more important than the organization of
> an army (users) is incorrect. Without one the other is pointless.
I'm hardly one to argue that non-code contributors aren't important. However,
I also don't see a reason to dramatically change the listings all at once;
why not do it in stages, with non-code contributors listed in the
"contributors" section this year, and under "major" maybe next year? Given
that non-code contributors currently aren't listed *at all* despite the
change in headings, I still see it moving in the right direction.
Futher, I don't feel that I -- as the core team member current preparing the
names for the contributor listings -- have a good handle on the difference
between a "major" and "minor" non-code contributor. *I* would like a year to
feel out some good practices.
The core team felt this was a reasonable compromise; Magnus felt it was a
reasonable compromise. So far, you're the only one to think it's not a
reasonable way to go, so I'd like to hear from some other people as well as
you.
--
Josh Berkus
PostgreSQL @ Sun
San Francisco
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sun, 9 Mar 2008 12:11:57 -0700 Josh Berkus <josh@agliodbs.com> wrote: > Actually, looking at the archives, there wasn't much discussion when > you submitted the patch; I doubt that most people realized the > changes it made. I know that I wasn't clear on it, noticing only the > change in sort order, and certainly the rest of the core team (or > hackers) didn't discuss it. Perhaps you should read again there are actually two threads and there was a very long discussion. That every single core member spoke up at least once except for Jan. The following is the thread on my patch which changed the wording from Developers to Contributors and moved its home under /community/. http://archives.postgresql.org/pgsql-www/2007-12/msg00022.php And then a conversation you initiated on actually updating the list last November. http://archives.postgresql.org/pgsql-www/2007-11/msg00432.php > > Contributor listings are not a matter of *only* WWW team's > discretion. In fact, traditionally, contents of the developer > listings have been determined by the core team; As I understand it, traditionally Robert handled that with Core oversight. > if we're going to > change that and give sole authority to WWW maintainers, then I think > we ought to have an *explicit* discussion about that. > I am not sure where you got this. I never suggested in anyway it was the sole responsibility or discretion of the WWW team. > > The idea that a line of code is more important than the > > organization of an army (users) is incorrect. Without one the other > > is pointless. > > I'm hardly one to argue that non-code contributors aren't important. > However, I also don't see a reason to dramatically change the > listings all at once; why not do it in stages, with non-code > contributors listed in the "contributors" section this year, and > under "major" maybe next year? Given that non-code contributors > currently aren't listed *at all* despite the change in headings, I > still see it moving in the right direction. The reason non code contributors aren't listed is the list has been basically unmaintained for over a year. There are also code contributors that aren't listed, Stefan being one of the most glaring examples. Secondly, using your reasoning it's o.k. to overlook the contributions made by non code contributors, "because it's easier". It would be easier just to remove the whole list. If you are going to start making judgments on the value of someones contributions, you must do it in full. You can not, in good conscious be arbitrary or selective. > > The core team felt this was a reasonable compromise; Magnus felt it > was a reasonable compromise. So far, you're the only one to think > it's not a reasonable way to go, so I'd like to hear from some other > people as well as you. > Well I would like hear it from more than just you. Magnus hasn't said a word publicly, neither has Dave, Jan, Marc, Tom, Bruce, Peter or the several dozen of other major contributors that are or are not listed. Currently you and I are the only one's speaking about this. Secondly - -www is hardly the forum for this as this is not about "the website" but about attribution to the contributors to the community which should take place in a more populous forum such as -general. Robert (who I believe is actually person in charge of this list) has stated: A major contributor is differentiated from other contributors based on 1) longevity in the community, 2) number of areas they work on, 3) signifigance of the contributions that have been made. http://archives.postgresql.org/pgsql-www/2007-11/msg00488.php Based on the above policy, code has nothing to do with it, the size of contribution does. The fact that non core contributors haven't been listed appears to be an oversight more than anything. One that is long overdue to be fixed. Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL political pundit | Mocker of Dolphins -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH1FR4ATb/zqfZUUQRAkGpAJ43oCgc0B7NoNmSw2Ux46MbhSsKIACfd0Ga Izxu+8mo+piPhspfoq/drHI= =t/es -----END PGP SIGNATURE-----
"Joshua D. Drake" <jd@commandprompt.com> writes: > Josh Berkus <josh@agliodbs.com> wrote: >> Actually, looking at the archives, there wasn't much discussion when >> you submitted the patch; I doubt that most people realized the >> changes it made. I know that I wasn't clear on it, noticing only the >> change in sort order, and certainly the rest of the core team (or >> hackers) didn't discuss it. > Perhaps you should read again there are actually two threads and there > was a very long discussion. That every single core member spoke up at > least once except for Jan. The following is the thread on my patch > which changed the wording from Developers to Contributors and moved its > home under /community/. > http://archives.postgresql.org/pgsql-www/2007-12/msg00022.php Why is any of this discussion happening on pgsql-www? Surely it is not the webteam's charter to determine who is or is not a contributor. If, as is evidently the case, you are unwilling to defer to core's opinions on the matter, you need to muster support for your position on some more widely-read mailing list. >> In fact, traditionally, contents of the developer >> listings have been determined by the core team; > As I understand it, traditionally Robert handled that with Core > oversight. I don't particularly know who made the webpage edits, but it's always been core's determination who is or is not listed. > The reason non code contributors aren't listed is the list has been > basically unmaintained for over a year. Traditionally core has reconsidered the list shortly after each major release. We were in fact in process of doing that now. The long delay since the last changes is a direct result of the slipped 8.3 schedule. regards, tom lane
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sun, 09 Mar 2008 20:59:07 -0400 Tom Lane <tgl@sss.pgh.pa.us> wrote: > > http://archives.postgresql.org/pgsql-www/2007-12/msg00022.php > > Why is any of this discussion happening on pgsql-www? Surely it is > not the webteam's charter to determine who is or is not a contributor. > Agreed :) > If, as is evidently the case, you are unwilling to defer to core's > opinions on the matter, you need to muster support for your position > on some more widely-read mailing list. Yes, I brought that up in reply :). I said, it should be discussed on pgsql-general or other populated list. JoshB is the one that started this on -www. > > >> In fact, traditionally, contents of the developer > >> listings have been determined by the core team; > > > As I understand it, traditionally Robert handled that with Core > > oversight. > > I don't particularly know who made the webpage edits, but it's always > been core's determination who is or is not listed. > O.k. > > The reason non code contributors aren't listed is the list has been > > basically unmaintained for over a year. > > Traditionally core has reconsidered the list shortly after each major > release. We were in fact in process of doing that now. The long > delay since the last changes is a direct result of the slipped 8.3 > schedule. Fair enough, I assume -hackers is not the place to have this discussion. Should this be on -general? Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL political pundit | Mocker of Dolphins -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH1JZaATb/zqfZUUQRArdsAKCbOF5X9c1Q/XpD0nCvfCb4JIUozgCfW3PO OlcUm/ucENsKeU0t7LeE0LI= =ukGB -----END PGP SIGNATURE-----
On Sun, Mar 09, 2008 at 02:19:50PM -0700, Joshua D. Drake wrote: > > Contributor listings are not a matter of *only* WWW team's > > discretion. In fact, traditionally, contents of the developer > > listings have been determined by the core team; > > As I understand it, traditionally Robert handled that with Core > oversight. Yes, that is correct. Basically, Robert drew up the changes he wanted to do, and sent them by -core for approval. > > > The idea that a line of code is more important than the > > > organization of an army (users) is incorrect. Without one the other > > > is pointless. > > > > I'm hardly one to argue that non-code contributors aren't important. > > However, I also don't see a reason to dramatically change the > > listings all at once; why not do it in stages, with non-code > > contributors listed in the "contributors" section this year, and > > under "major" maybe next year? Given that non-code contributors > > currently aren't listed *at all* despite the change in headings, I > > still see it moving in the right direction. > > The reason non code contributors aren't listed is the list has been > basically unmaintained for over a year. There are also code > contributors that aren't listed, Stefan being one of the most glaring > examples. That's because it hasn't yet been updated for 8.3. > > The core team felt this was a reasonable compromise; Magnus felt it > > was a reasonable compromise. So far, you're the only one to think > > it's not a reasonable way to go, so I'd like to hear from some other > > people as well as you. > > > > Well I would like hear it from more than just you. Magnus hasn't said a > word publicly, neither has Dave, Jan, Marc, Tom, Bruce, Peter or the Well, let me say it publically then, if the reference isn't enough. I think it's a reasonable compromise to get started with, that we can continue building on. It was also presented as the solution that -core agreed on. I'm sure that if Josh actually lied about that, someone would've spoken up quite fast. But I strongly doubt that Josh would claim to present the "view of the core team" if the discussion hadn't taken place. //Magnus
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 10 Mar 2008 10:33:53 +0100 Magnus Hagander <magnus@hagander.net> wrote: > It was also presented as the solution that -core agreed on. I'm sure > that if Josh actually lied about that, someone would've spoken up > quite fast. But I strongly doubt that Josh would claim to present the > "view of the core team" if the discussion hadn't taken place. I am in no way suggesting that Josh was lying. Joshua D. Drake > > //Magnus > - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL political pundit | Mocker of Dolphins -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH1VS0ATb/zqfZUUQRAp3WAJ4mXmx5d9bI312YSTImRkHNifCp2QCeKgQV yjCmydSKHI2GV6dzM3p3mWw= =4nKT -----END PGP SIGNATURE-----
Magnus, > It was also presented as the solution that -core agreed on. I'm sure that > if Josh actually lied about that, someone would've spoken up quite fast. > But I strongly doubt that Josh would claim to present the "view of the core > team" if the discussion hadn't taken place. Heh. As if I could get away with that -- I'd have until list lag caught up to get blasted. I guess one of the questions here is "who owns the contributor listings?". It's not a question we've ever dealt with specifically before, and it's unclear on even what *mailing list* would be involved in discussing them. It seems like we'd need to involve half or more of the lists. For the last 3 years, nobody has discussed this because Robert just did it and submitted the list to Core, which approved it. Now Robert is tired of the work, and what was implicit needs to become explicit. The reason I'm putting forward that Core ought to be ultimately responsible is threefold: 1) Core is a central point of contact which is supposed to know what's going on in the various disconnected mailing lists, and as such is our only existing "central" coordinating group; 2) The seven Core team members place in the listings isn't going to change, and thus we can argue about who should be where without statutory personal bias; 3) Core does conventionally deal with other issues around contributor status, such as CVS access, release notes, and (in extreme cases) banning. Barring Core handling it, we'd have to form a separate committee, and somehow pick people who would be both representative and relatively impartial. That seems like it would increase the amount of work involved in getting the listings updated siginificantly, to the point where they might not get updated at all. Given that the only identified real problem (listings not being updated frequently enough) is not solved by forming a separate committee, why not take the easiest path, at least until another concrete problem is identified? --Josh Berkus
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 10 Mar 2008 10:15:32 -0700 Josh Berkus <josh@agliodbs.com> wrote: > Magnus, > > > It was also presented as the solution that -core agreed on. I'm > > sure that if Josh actually lied about that, someone would've spoken > > up quite fast. But I strongly doubt that Josh would claim to > > present the "view of the core team" if the discussion hadn't taken > > place. > > Heh. As if I could get away with that -- I'd have until list lag > caught up to get blasted. > > I guess one of the questions here is "who owns the contributor > listings?". It's not a question we've ever dealt with specifically > before, and it's unclear on even what *mailing list* would be > involved in discussing them. It seems like we'd need to involve half > or more of the lists. As I suggested, Tom double suggested, I agreed and then I talked to Bruce on the phone about this thread has been moved to a more populous list. Please see the new thread on -advocacy. Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL political pundit | Mocker of Dolphins -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH1XCvATb/zqfZUUQRAoeUAJ9pGi04KVRxSHom4onc/OO0b8jjngCcC6wa LuLQ1udTaCnWilAGCArKhLA= =pG4o -----END PGP SIGNATURE-----
Josh, > Fair enough, I assume -hackers is not the place to have this > discussion. Should this be on -general? I don't think so -- General is mostly newbies seeking help, not contributors. Hackers would be the most applicable list, followed by this list (already taken care of) and Advocacy for non-code contributors. However, such a discussion is liable to take weeks. You need to ask yourself whether the changes you want are worth spending weeks sheperding an e-mail discussion, and then longer before the listings actually get updated. --Josh
On Monday 10 March 2008 13:15, Josh Berkus wrote:
> Magnus,
>
> > It was also presented as the solution that -core agreed on. I'm sure that
> > if Josh actually lied about that, someone would've spoken up quite fast.
> > But I strongly doubt that Josh would claim to present the "view of the
> > core team" if the discussion hadn't taken place.
>
> Heh. As if I could get away with that -- I'd have until list lag caught
> up to get blasted.
>
> I guess one of the questions here is "who owns the contributor
> listings?". It's not a question we've ever dealt with specifically
> before, and it's unclear on even what *mailing list* would be involved
> in discussing them. It seems like we'd need to involve half or more of
> the lists.
>
> For the last 3 years, nobody has discussed this because Robert just did
> it and submitted the list to Core, which approved it. Now Robert is
> tired of the work, and what was implicit needs to become explicit.
>
To be clear, I didn't get tired of the work, I actually enjoyed the work
(making sure others get recognition for thier efforts was a highlight for
me). What I did get tired of was that everytime I went to update the list
some yaywho would pipe up with yet another scheme to redraw the entire
listing.
--
Robert Treat
Build A Brighter LAMP :: Linux Apache {middleware} PostgreSQL
On Sunday 09 March 2008 20:59, Tom Lane wrote: > "Joshua D. Drake" <jd@commandprompt.com> writes: > > Josh Berkus <josh@agliodbs.com> wrote: > >> Actually, looking at the archives, there wasn't much discussion when > >> you submitted the patch; I doubt that most people realized the > >> changes it made. I know that I wasn't clear on it, noticing only the > >> change in sort order, and certainly the rest of the core team (or > >> hackers) didn't discuss it. > > > > Perhaps you should read again there are actually two threads and there > > was a very long discussion. That every single core member spoke up at > > least once except for Jan. The following is the thread on my patch > > which changed the wording from Developers to Contributors and moved its > > home under /community/. > > > > http://archives.postgresql.org/pgsql-www/2007-12/msg00022.php > > Why is any of this discussion happening on pgsql-www? Surely it is not > the webteam's charter to determine who is or is not a contributor. > > If, as is evidently the case, you are unwilling to defer to core's > opinions on the matter, you need to muster support for your position > on some more widely-read mailing list. > > >> In fact, traditionally, contents of the developer > >> listings have been determined by the core team; > > > > As I understand it, traditionally Robert handled that with Core > > oversight. > > I don't particularly know who made the webpage edits, but it's always > been core's determination who is or is not listed. > > > The reason non code contributors aren't listed is the list has been > > basically unmaintained for over a year. > > Traditionally core has reconsidered the list shortly after each major > release. We were in fact in process of doing that now. The long delay > since the last changes is a direct result of the slipped 8.3 schedule. > This seems to discount my efforts in the whole process. My theory was if I ever left it up to core, it would never get updated. Never is a long time, so I suspect I'll be proven wrong in the long run, but so far I am correct. -- Robert Treat Build A Brighter LAMP :: Linux Apache {middleware} PostgreSQL
On Wed, Mar 12, 2008 at 11:56 PM, Robert Treat <xzilla@users.sourceforge.net> wrote: > > This seems to discount my efforts in the whole process. My theory was if I > ever left it up to core, it would never get updated. Never is a long time, so > I suspect I'll be proven wrong in the long run, but so far I am correct. I think we were all waiting for you to do your thing again :-) -- Dave Page EnterpriseDB UK Ltd: http://www.enterprisedb.com PostgreSQL UK 2008 Conference: http://www.postgresql.org.uk