Re: Raphael Hertzog: When to commit into repositories of teams?
On 09/03/2008, Andreas Barth wrote: > as campaigning has started, I would like to know from Raphael Hertzog > his opinion under which circumstances he considers it ok to commit > into revision control repositories of a team where the person leading > the team is active and asks to not commit. Hi, may I ask whether you have examples in mind, and how you would define “is active”? Cheers, -- Cyril Brulebois pgpGO2p7JUKKX.pgp Description: PGP signature
Re: Raphael Hertzog: When to commit into repositories of teams?
* Andreas Barth [Sun, 09 Mar 2008 21:28:52 +0100]: > as campaigning has started, I would like to know from Raphael Hertzog > his opinion under which circumstances he considers it ok to commit into > revision control repositories of a team where the person leading the > team is active and asks to not commit. #436093 for those following along at home. -- Adeodato Simó dato at net.com.org.es Debian Developer adeodato at debian.org Poco después de la guerra civil, un brote de cólera se había llevado a mi madre. La enterramos en Montjuïc el día de mi cuarto cumpleaños. Sólo recuerdo que llovió todo el día y toda la noche, y que cuando le pregunté a mi padre si el cielo lloraba le faltó la voz para responderme. -- Carlos Ruiz Zafón, “La sombra del viento” -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
On Sun, Mar 09, 2008 at 09:28:52PM +0100, Andreas Barth wrote: > as campaigning has started, I would like to know from Raphael Hertzog > his opinion under which circumstances he considers it ok to commit into > revision control repositories of a team where the person leading the > team is active and asks to not commit. ...so much for non-adversarial campaigning, I guess. Cheers, aj -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
aj wrote: [...] > ...so much for non-adversarial campaigning, I guess. Why? So far, this is only adversarial questioning of a candidate. It doesn't necessarily require adversarial campaigning in reply. Or is aba campaigning for one of the other candidates? FWIW, I think each of the candidates have some doubtful debian deeds worth explaining, but there's only one I expect to get all my-way-or-highway-style adversarial if questioned aggressively. With any luck, we'll see in the campaign-only aeons whether that's correct. Oh and I don't really understand the question: why should anyone stop commits into a repository? You could just ask that they're not put on the release management branch without review. Regards, -- MJR/slef My Opinion Only: see http://people.debian.org/~mjr/ Please follow http://www.uk.debian.org/MailingLists/#codeofconduct -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
* Adeodato Simó ([EMAIL PROTECTED]) [080309 22:05]: > * Andreas Barth [Sun, 09 Mar 2008 21:28:52 +0100]: > > > as campaigning has started, I would like to know from Raphael Hertzog > > his opinion under which circumstances he considers it ok to commit into > > revision control repositories of a team where the person leading the > > team is active and asks to not commit. > > #436093 for those following along at home. It could be dpkg as well. Cheers, Andi -- http://home.arcor.de/andreas-barth/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
* MJ Ray ([EMAIL PROTECTED]) [080310 08:13]: > Oh and I don't really understand the question: why should anyone stop > commits into a repository? You could just ask that they're not put on > the release management branch without review. I'm speaking of the release management branch, like it e.g. happened recently with dpkg. Cheers, Andi -- http://home.arcor.de/andreas-barth/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
Hi Andreas, On Sun, 09 Mar 2008, Andreas Barth wrote: > as campaigning has started, I would like to know from Raphael Hertzog > his opinion under which circumstances he considers it ok to commit into > revision control repositories of a team where the person leading the > team is active and asks to not commit. It's sad that you never decided to resolve that dispute when it was live e.g. by accepting the compromise proposed in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=436093#110 I think everything has already been said in the bug log above. I would like to point to this message of Steve Langasek also: http://lists.debian.org/debian-ctte/2007/08/msg00034.html When I discussed privately with you, you only told me it was not ok to commit because I was not the maintainer and never gave any valid reason. From my side, it was really looking much like the few DD who complain of NMU instead of appreciating the help offered by other persons. Explaining before appealing to a higher authority is always a good idea. So as you can see, I accept technical rules regulating commits to a shared repository, it's just that there must be good reasons justifying the overhead. Cheers, -- Raphaël Hertzog Le best-seller français mis à jour pour Debian Etch : http://www.ouaza.com/livre/admin-debian/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
On Mon, 10 Mar 2008, Andreas Barth wrote: > * Adeodato Simó ([EMAIL PROTECTED]) [080309 22:05]: > > #436093 for those following along at home. > > It could be dpkg as well. Can you justify this assertion? I have worked _with_ Guillem Jover and Frank Lichtenheld and _they_ have granted me commit rights to the repository. Cheers, -- Raphaël Hertzog Le best-seller français mis à jour pour Debian Etch : http://www.ouaza.com/livre/admin-debian/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
* Raphael Hertzog ([EMAIL PROTECTED]) [080310 09:46]: > Hi Andreas, > > On Sun, 09 Mar 2008, Andreas Barth wrote: > > as campaigning has started, I would like to know from Raphael Hertzog > > his opinion under which circumstances he considers it ok to commit into > > revision control repositories of a team where the person leading the > > team is active and asks to not commit. > > It's sad that you never decided to resolve that dispute when it was live > e.g. by accepting the compromise proposed in > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=436093#110 So you still think one should just commit to any repository one has technically access to? So e.g. Ians commits to the dpkg repository were ok? And you still think that hijacking a package is ok, and afterwards offering to "oh, I will wait with any commit for 14 days, but will commit it nevertheless" is proper? Cheers, Andi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
* Raphael Hertzog ([EMAIL PROTECTED]) [080310 09:46]: > On Mon, 10 Mar 2008, Andreas Barth wrote: > > * Adeodato Simó ([EMAIL PROTECTED]) [080309 22:05]: > > > #436093 for those following along at home. > > > > It could be dpkg as well. > > Can you justify this assertion? > > I have worked _with_ Guillem Jover and Frank Lichtenheld and _they_ have > granted me commit rights to the repository. Where in my message did I say "buxy has done commits of that kind, and he should explain his actions"? I just asked them what your opinion is, and the recent dpkg flame fest gave me the idea to ask. In that case, you obviously were involved, as you e.g. asked for unaccepts of dpkg, and reverted the commits of Ian to the git repository. So, perhaps you can give me a summary why e.g. Ians commits to dpkg wasn't ok, even though he was trying to get his patches integrated for 6 months, but your commits to dev-ref were ok, even though you didn't contact me (or opened a bug report) and it is explicitly documented you shouldn't just commit? Cheers, Andi -- http://home.arcor.de/andreas-barth/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
On Mon, 10 Mar 2008, Andreas Barth wrote: > * Raphael Hertzog ([EMAIL PROTECTED]) [080310 09:46]: > > On Mon, 10 Mar 2008, Andreas Barth wrote: > > > * Adeodato Simó ([EMAIL PROTECTED]) [080309 22:05]: > > > > #436093 for those following along at home. > > > > > > It could be dpkg as well. > > > > Can you justify this assertion? > > > > I have worked _with_ Guillem Jover and Frank Lichtenheld and _they_ have > > granted me commit rights to the repository. > > Where in my message did I say "buxy has done commits of that kind, and > he should explain his actions"? I just asked them what your opinion is, > and the recent dpkg flame fest gave me the idea to ask. The subject of your mail is "When to commit into repositories of teams?" and you thus imply that I have commited without authorization in the dpkg team repository. That's simply not correct. > In that case, you obviously were involved, as you e.g. asked for > unaccepts of dpkg, and reverted the commits of Ian to the git > repository. I didn't revert anything. Ian only committed in seperate branches. The unaccept has been requested by Guillem and I simply pinged ftpmasters on IRC and communicated the status of the situation to -devel. > So, perhaps you can give me a summary why e.g. Ians commits to dpkg > wasn't ok, even though he was trying to get his patches integrated for 6 > months, but your commits to dev-ref were ok, even though you didn't > contact me (or opened a bug report) and it is explicitly documented you > shouldn't just commit? I wasn't aware of the new review rule and you didn't explain it to me. I said that I will happily provide patches instead to let other people comment in the future. My commits were not ok but you failed to explain me why. And I didn't want to replace you and I'm always ready to work with you. In Ian's case, he was aware before the commit that we didn't want him to do the merge and despite the guidance that I tried to provide him, he stubbornly was blocked to the unconditional merge of his branch as is (that is with a messed up history). Ian hasn't accepted any compromise and hijacked the package and clearly showed that he doesn't want to work with Guillem. I hope you see the difference. Cheers, -- Raphaël Hertzog Le best-seller français mis à jour pour Debian Etch : http://www.ouaza.com/livre/admin-debian/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
On Mon, 10 Mar 2008, Andreas Barth wrote: > * Raphael Hertzog ([EMAIL PROTECTED]) [080310 09:46]: > > Hi Andreas, > > > > On Sun, 09 Mar 2008, Andreas Barth wrote: > > > as campaigning has started, I would like to know from Raphael Hertzog > > > his opinion under which circumstances he considers it ok to commit into > > > revision control repositories of a team where the person leading the > > > team is active and asks to not commit. > > > > It's sad that you never decided to resolve that dispute when it was live > > e.g. by accepting the compromise proposed in > > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=436093#110 > > So you still think one should just commit to any repository one has > technically access to? So e.g. Ians commits to the dpkg repository were > ok? And you still think that hijacking a package is ok, and afterwards > offering to "oh, I will wait with any commit for 14 days, but will > commit it nevertheless" is proper? You're mixing up everything: - I didn't saw the "do not commit" instruction. As a previous contributor I didn't re-read the README-contrib as I was used to work directly in the VCS when Adam Di Carlo was still the official maintainer. - When I committed my changes, you didn't point me to that file and you didn't explain me why you didn't want me to commit. Thus I didn't accept your arbitrary rule (in particular when my involvement in developers-reference predates yours). - In the above compromise, I accept the rule if its purpose is to have review of the patches before they are committed. - That said lack of review is not good but outdated information in the developers-reference is far worse. Thus I suggested that we have a 15 days period where editors can comment and explicitely accept/block a patch (of course a justification needs to be given too). After that any editor can integrate the change by himself it he thinks it's ok. - I also requested to be added back in my editor status as you removed me without my permission. While I had not contributed for a long time, I was still subscribed to the PTS and I was reachable by e-mail. There's nothing else Andreas. Please forgive me if I hurt you while doing this but you have not been very cooperative either. Cheers, -- Raphaël Hertzog Le best-seller français mis à jour pour Debian Etch : http://www.ouaza.com/livre/admin-debian/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
On Mon March 10 2008 02:50:51 Raphael Hertzog wrote: > In Ian's case, he was aware before the commit that we didn't want him to > do the merge and despite the guidance that I tried to provide him, he > stubbornly was blocked to the unconditional merge of his branch as is > (that is with a messed up history). Ian hasn't accepted any compromise and > hijacked the package and clearly showed that he doesn't want to work > with Guillem. Raphael, Where is the dpkg team policy that requires Ian to rebase? We have asked many times and the only policy[0] you have offered in your defense contains recommendations for the developer (Ian), not mandates, and clearly shows that the developer chooses how many times to rebase - whether that be zero, one, or more times. Is is not the case that you and Guillem have violated your own team policy for roughly six months by blocking Ian's changes? --Mike Bird [0] http://wiki.debian.org/Teams/Dpkg/GitUsage -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
* Raphael Hertzog ([EMAIL PROTECTED]) [080310 10:51]: > On Mon, 10 Mar 2008, Andreas Barth wrote: > > * Raphael Hertzog ([EMAIL PROTECTED]) [080310 09:46]: > > > On Mon, 10 Mar 2008, Andreas Barth wrote: > > > > * Adeodato Simó ([EMAIL PROTECTED]) [080309 22:05]: > > > > > #436093 for those following along at home. > > > > > > > > It could be dpkg as well. > > > > > > Can you justify this assertion? > > > > > > I have worked _with_ Guillem Jover and Frank Lichtenheld and _they_ have > > > granted me commit rights to the repository. > > > > Where in my message did I say "buxy has done commits of that kind, and > > he should explain his actions"? I just asked them what your opinion is, > > and the recent dpkg flame fest gave me the idea to ask. > > The subject of your mail is "When to commit into repositories of teams?" > and you thus imply that I have commited without authorization in the dpkg > team repository. That's simply not correct. Eh, I would think taking a class in reading mails prior to running for DPL would be a good thing for you. My mail implies that I want to have an opinion from you on that. It doesn't imply personal experience with such behaviour. But I think that it is important for any DPL candidate to be able to be a conflict solver, and - as the recent dpkg case has shown - conflicts over commits do happen, and can get quite bad. I don't have any doubts with the other two candidates, but as you have played an active role in such conflicts at least twice (on different sides of the conflicts), I would be interessted in your anwers. As I have learned, you didn't really want to answer that. > > So, perhaps you can give me a summary why e.g. Ians commits to dpkg > > wasn't ok, even though he was trying to get his patches integrated for 6 > > months, but your commits to dev-ref were ok, even though you didn't > > contact me (or opened a bug report) and it is explicitly documented you > > shouldn't just commit? > > I wasn't aware of the new review rule and you didn't explain it to me. I asked you on IRC to stop committing for various reasons, one of it being a pending conversion of the dev-ref to another format, but you refused to do so, and you told me that you will commit anyways even if I object. > In Ian's case, he was aware before the commit that we didn't want him to > do the merge and despite the guidance that I tried to provide him, he > stubbornly was blocked to the unconditional merge of his branch as is > (that is with a messed up history). So the difference is that Ian contacted the dpkg-people before, but you didn't contact the dev-ref-people before, and for this reason, you were allowed to commit, but Ian was not? Interessting point of view. Cheers, Andi -- http://home.arcor.de/andreas-barth/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
On Mon, March 10, 2008 11:07, Mike Bird wrote: > Where is the dpkg team policy that requires Ian to rebase? > We have asked many times Exactly who is "we" here? Please name me three of your most recent contributions to Debian that justify you making demands on the project c.q. Raphaël. thanks, Thijs -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Raphael Hertzog: When to commit into repositories of teams?
On Mon, 10 Mar 2008, Andreas Barth wrote: > But I think that it is important for any DPL candidate to be able to > be a conflict solver, and - as the recent dpkg case has shown - > conflicts over commits do happen, and can get quite bad. In both cases I have made proposals to go forward but the other side have deliberatly ignored them. I can make nothing against stubborness. > As I have learned, you didn't really want to answer that. My answer is simple: it was wrong for me to commit in the developers-reference repository. But the fact that you didn't provide a convincing justification of the rule before you escalated it up to the technical committee is the reason why I took it as a personal bias of you against me instead of some rational decision. > > I wasn't aware of the new review rule and you didn't explain it to me. > > I asked you on IRC to stop committing for various reasons, one of it > being a pending conversion of the dev-ref to another format, but you > refused to do so, and you told me that you will commit anyways even if I > object. You used that argument twice at 12 months of interval, I was truly under the impression that you made it up to block further contribution from me. I was following the PTS and the bug report related to the XML conversion had not seen any activity since a very long time (Jul 2006!): http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=374220 I think I have brought all the facts in this thread. My behaviour wasn't perfect and never will be, but I have made sincere attempts to solve the conflict and you simply ignored them. The fact that you brought the story under a strange question for a DPL debate proves that you're not yet able to forgive me. I'm sorry for that. Cheers, -- Raphaël Hertzog Le best-seller français mis à jour pour Debian Etch : http://www.ouaza.com/livre/admin-debian/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]