Re: GNOME DVCS Survey Results
On Tue, Jan 06, 2009 at 11:09:27AM +0100, Josselin Mouette wrote: Le mardi 06 janvier 2009 à 17:21 +0900, James Henstridge a écrit : I'd hope that any DVCS would get a larger user base than current list of active Subversion committers: anyone who contributes patches via mailing lists or bugzilla could use a DVCS the same way as a core developer. The same could go for anyone who does distribution packaging. That???s foolish. We are certainly not going to clone 208 repositories for all GNOME packages we maintain. Especially when the tools to manage Debian packages with DVCSes are still inferior to those we have for the current svn+tarball scheme. Hmm, can't say i agree with that. I maintain Debian packages in SVN, git and bzr. And i must say of those three systems i definately prefer git for maintaining things. Maybe you should have another look at debians DVCS tools? Sjoerd -- Paradise is exactly like where you are right now ... only much, much better. -- Laurie Anderson ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
As I haven't really contributed anything to GNOME in a year or so, I've been keeping out of the debate, but: On 07/01/2009, at 9:00 AM, Gustavo J. A. M. Carneiro wrote: Welcome to the open source world. Generally open source developers are not limited to GNOME, and they eventually learn 2-3 revision control systems. I mean, they don't need to learn a lot of commands of each RCS, just the basics: 1- Checkout/clone module; 2- Update; 3- Create a diff of changes, redirecting to a patch file. I also agree that many contributors do those tasks (especially non- coders like translators), and it would be fairly simple to have a small script that does those tasks for all the SCMs that GNOME hosted if it chose to host multiple SCMs. Something like gnome-scm checkout git:http://host.xz/path/to/repo.git/, gnome-scm checkout bzr:http://bzr-project.example.com/foo.stable/ , gnome-scm update and gnome-scm diff would be a good start. This obviously wouldn't do anything fancy, but for all the people who don't have about SCMs and just want to make a small change it would work. Since it would presumably delegate actual work to the underlying SCM tools, it would also be an easy transition for anyone who wanted to start doing more powerful things, by allowing them to use the real tools on their checkout. If GNOME hosted multiple SCMs then it would mean that developers who work on a large number of projects might have to learn both git and bzr. Which I can't see as a huge issues because anything they maintain will be presumably in their favourite SCM, anything they only do a small amount of work on will only require the basics on the SCM, and anyone who does serious work on multiple projects probably won't have a problem learning another SCM. Is knowing both git and bzr a huge issue? I imagine that most serious developers here already know quite a few, at least CVS, SVN, Git/BZR and probably others. James (who is ducking back outside the flamethrowing-testing area) ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
James Livingston schrieb: As I haven't really contributed anything to GNOME in a year or so, I've been keeping out of the debate, but: On 07/01/2009, at 9:00 AM, Gustavo J. A. M. Carneiro wrote: Welcome to the open source world. Generally open source developers are not limited to GNOME, and they eventually learn 2-3 revision control systems. I mean, they don't need to learn a lot of commands of each RCS, just the basics: 1- Checkout/clone module; 2- Update; 3- Create a diff of changes, redirecting to a patch file. I also agree that many contributors do those tasks (especially non-coders like translators), and it would be fairly simple to have a small script that does those tasks for all the SCMs that GNOME hosted if it chose to host multiple SCMs. Something like gnome-scm checkout git:http://host.xz/path/to/repo.git/, gnome-scm checkout bzr:http://bzr-project.example.com/foo.stable/;, gnome-scm update and gnome-scm diff would be a good start. This obviously wouldn't do anything fancy, but for all the people who don't have about SCMs and just want to make a small change it would work. Since it would presumably delegate actual work to the underlying SCM tools, it would also be an easy transition for anyone who wanted to start doing more powerful things, by allowing them to use the real tools on their checkout. You want git and bzr support for moap :) Stefan If GNOME hosted multiple SCMs then it would mean that developers who work on a large number of projects might have to learn both git and bzr. Which I can't see as a huge issues because anything they maintain will be presumably in their favourite SCM, anything they only do a small amount of work on will only require the basics on the SCM, and anyone who does serious work on multiple projects probably won't have a problem learning another SCM. Is knowing both git and bzr a huge issue? I imagine that most serious developers here already know quite a few, at least CVS, SVN, Git/BZR and probably others. James (who is ducking back outside the flamethrowing-testing area) ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Tue, Jan 6, 2009 at 2:57 PM, Elijah Newren new...@gmail.com wrote: It's a shame that hackers who contribute to GNOME projects which don't use svn.gnome.org were excluded. (I was told their opinions didn't matter. {shrug} that's fine, so long as nobody tries to represent this survey as what GNOME hackers think) How would you draw the line? Who should be included and who shouldn't? And how do we contact them all? I think doing a survey of any group other than those with svn commit access would be practically unmanageable...and far more likely to be suspected of non-representative-ness [New word!]. I'd hope that any DVCS would get a larger user base than current list of active Subversion committers: anyone who contributes patches via mailing lists or bugzilla could use a DVCS the same way as a core developer. The same could go for anyone who does distribution packaging. At the same time, there are reasons not to value their input to the same extent as core developers: 1. some proportion of them will continue to work with patches no matter what VCS is in use. 2. it is difficult to predict what proportion will use DVCS features. 3. those without svn access probably don't hack on the code as much As it is, there isn't much way to tell whether the answers from this group mirror those of committers or not. James. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Hi everyone, Well, let me first say that I'm a bit disappointed at where this thread has gone so far. Some of the words passed around we could do without starting a brand new year. I want to encourage everyone to stay on topic, so we can actually do something. Now, I didn't say so we can make decision, because let me make it clear if it wasn't, this thread is not about making decisions. This thread is about giving those making the decision input they need to consider. Who makes the decision? Those who actually have to implement, oversee, and maintain any change. And needless to say, no change will happen unless someone is willing to JF do it. From what I understand, so far there has been one proposal, from Olav, who proposed that he and John implement John's idea of implementing a git-serve plugin for the bzr repo server. I think many people pointed out the major flaws in that scenario. I want to stress that we should not make the same mistakes that we blame our downstream distros for making, namely: developing hacks in house instead of working them out upstream. Moreover, we should not get ourselves in another homegrown mess that will need to be maintained for the years to come. Look at our bugzilla situation now. git-serve will be a bigger issue. So, here is what I'm bringing to the table: I'm volunteering to work with interested fd.o admins and other volunteers to switch GNOME to git. I need to first go check to see if I can secure enough time to lead this, and if I can get enough peoples' attention to make this happen in a timely manner. I will make an official proposal when I figure those out. Emmanuele Bassi wrote: [0] I'm reasonably sure it has some. not as the one proposed to avoid pissing off somebody somewhere because we want to be inclusive -- no, lemme rephrase that: we are *fucking afraid of committment*. seriously: an abstraction over DVCS? what have we become? are we *ever* going make *any* decision about *anything*? this is actually a larger issue with the GNOME community: we are being afraid. Thanks Emmanuele for bringing this up. It deserves more than a postscript spot. This is a real issue we need to overcome. At some point, we need to understand that we can't make everyone happy. We should aim for making everyone happy-enough. And it helps if individuals also understand that it's not about any single person getting what he wants. That's exactly why we did the survey to get an idea of what the community at large prefers. Cheers, and a late Happy New Year, behdad ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Andrew Cowie wrote: On Sat, 2009-01-03 at 22:46 -0500, Behdad Esfahbod wrote: GNOME contributors with an SVN account who had an SSH key installed on their account were invited to fill in the survey. [It is NOT my intention to get all negative here; I understand - and accept - that projects make decisions and not everyone is happy with them. Luckily this decision ultimately is one I can ignore. Nevertheless, I have been asked by a number of people to write to this thread with why I am so dissatisfied. I do appreciate the effort people made, even if I feel that the way the whole survey exercise was conducted it was impossible for Git to lose] I'm offended by this statement. What do you mean by the way the whole survey exercise was conducted it was impossible for Git to lose? Some comments: ++ It's a shame that hackers who contribute to GNOME projects which don't use svn.gnome.org were excluded. (I was told their opinions didn't matter. {shrug} that's fine, so long as nobody tries to represent this survey as what GNOME hackers think) If anyone represents it as what GNOME hackers think, it's you. What I told you was that because the switch does not affect those other people. Yes, if your contributors are NOT committing to GNOME SVN, their opinions doesn't matter. Neither does my mom's opinion matter in this case. Nothing wrong with that. ++ It was also a shame that I (one who does happen to have a GNOME svn account) was not able to complete the survey either because it crashes Epiphany when you i) vote for bzr and ii) withhold your vote from git, hg, and svn. (When I asked if it might be possible to fix the survey so that GNOME's web browser didn't crash, I was told known bug and too bad, you have to express a preference for Git and Mercurial even if you don't want to. Strange take on democracy. I am rather accustomed to the idea that declining to express a preference for something is an acceptable form of voting. Whatever) The bug became known only after the survey was started. I had three options: - Recall the survey, fix it, start over. - Fix the survey and let those who were affected by it already feel left out. - Continue as is. I chose the last option. The reason you had to rank all options was not the crasher however. The crash was in fact caused by the dialog window asking you to rank all items. As for why it required ranking all options, because of the release-team, sysadmin team, board, and other select individuals who saw the survey before it went live, none noticed this tiny issue. But of course you can theorize that it was done to make sure git can't lose. And indeed when I responded to you explaining this (more briefly, agreed), you chose not to reply. I explicitly did not want to chose Git or Mercurial, because I knew exactly what was going to happen. I've heard it several times already in #gnome-hackers and elsewhere: so it seems the people who prefer Bazaar like Git as their second choice, so surely it's ok to go with that. Great! Decision made No. The rest of the survey was irrelevant. It was quite evident that the object of the exercise was to allow people to say lots of people said Git was either their first or second choice which sounds very impressive, and was exactly the one thing I did NOT want to support. So it crashed my browser. Nice. Nice theory, yes. I designed it such that it crashes the browsers of those who didn't choose git as their favorite. In reality though, you are pissed off because you wanted to vote strategically and couldn't. ++ We chose the Bazaar decentralized version control system for our GNOME project even before the people behin GNOME's centralized code hosting made the courageous and monumental decision to switch from CVS to Subversion. Since GNOME didn't offer any way for us to host our 'mainline' branch on any official sounding resource, {shrug} we didn't. And so we don't. And that's actually the only issue that matters so far as I can tell. No one can force us to stop using Bazaar. People who work at places like Immendio who are using Git to hack on GTK+ cannot be forced to stop git either. And I wouldn't want them to. They're happy with their tool. We're happy with ours. When CVS was the only interchange (actually, that's not true, since the real interchange for most projects is attachments to Bugzilla of all things), then indeed GNOME switching to Subversion was a big deal. But in the era of distributed version control, the next step really matters little. Whatever GNOME _infrastructure_ offers next in terms of hosting is really quite irrelevant, since quite anyone can host their own projects and publish their own branches with nothing more than a vanilla web server. If the choice had happened to be Bazaar, then we probably would
Re: GNOME DVCS Survey Results
On Mon, 2009-01-05 at 16:41 +, Gustavo J. A. M. Carneiro wrote: Heh, thanks a lot. This looks nice. Nicer than the one in gnulib that Rui Tiago pointed out. Although I must say not as nice as my 'gnulog' bazaar log formatter plugin.. ;-) But I guess good enough that I'd be comfortable replacing a hand written ChangeLog with autogenerated one. Regards, It's just my first draft though, I intend to improve it over time, patches welcome :) And it could maybe be moved to some better location. Edward ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
From what I understand, so far there has been one proposal, from Olav, who proposed that he and John implement John's idea of implementing a git-serve plugin for the bzr repo server. I think many people pointed out the major flaws in that scenario. I want to stress that we should not make the same mistakes that we blame our downstream distros for making, namely: developing hacks in house instead of working them out upstream. Moreover, we should not get ourselves in another homegrown mess that will need to be maintained for the years to come. Look at our bugzilla situation now. git-serve will be a bigger issue. I just wanted to point out that the git-serve plugin for bzr is *not* an in house hack, or a hack at all, it is part of the bzr-git plugin. For people who are not aware of it, Bazaar is mainly an object model, while Git is mainly a repository format coupled with a network protocol. What's wrong with mapping an object model to an actual format/protocol ? Concerning the asserted flaws in John's proposal, the only valid point, is that it will need testing as the implementation is not very mature yet, all the other assertions are just plain wrong. Regards, -- Ali ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On di, 2009-01-06 at 04:53 -0500, Behdad Esfahbod wrote: Ali Sabil wrote: Concerning the asserted flaws in John's proposal, the only valid point, is that it will need testing as the implementation is not very mature yet, And that's a HUGE issue, mind you. Here's another problem I'd like to see pointed out. While the hybrid solution might work fine within the GNOME infrastructure, what happens if an outside contributor clones a repository, hacks on and publishes his code on his own server. Remember, we are talking about DVCSes, this *will* happen, it should happen. What if this user publishes his branch using bzr (which works fine with the GNOME servers). How will I merge this branch, if I'm using git? It looks to me that with the git+bzr proposal, we're being forced to learn both systems anyway. We can control what's going on on our own servers, but not what happens outside. So don't tell us it's equivalent. It's not. And let's not forget the broad confusion that will arise from the fact that part of the developers advice you to use git and other parts advice you to use bzr. I am exceedingly less convinced that the advantages of supporting both outweigh the disadvantages (looking at the usage patterns, not the technical merits). R -- Ruben Vermeersch (rubenv) http://www.savanne.be ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Josselin Mouette wrote: Le mardi 06 janvier 2009 à 17:21 +0900, James Henstridge a écrit : I'd hope that any DVCS would get a larger user base than current list of active Subversion committers: anyone who contributes patches via mailing lists or bugzilla could use a DVCS the same way as a core developer. The same could go for anyone who does distribution packaging. That’s foolish. We are certainly not going to clone 208 repositories for all GNOME packages we maintain. Especially when the tools to manage Debian packages with DVCSes are still inferior to those we have for the current svn+tarball scheme. Just because you don't do it doesn't mean that the idea is foolish. There certainly has been interest from multiple distros to move to a DVCS-based workflow. behdad ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Le mardi 06 janvier 2009 à 13:10 +1100, Andrew Cowie a écrit : Regardless, GNOME is not switching to anything. If GNOME infrastructure is going to offer Git hosting, that's lovely for people who chose to use Git as their version control system. {shrug} fine. If GNOME infrastructure concurrently disables their Subversion hosting and/or people stop pushing their changes there, then that's perhaps a bit worse, because it means people in all three systems (+ svn makes four) will lose the easy way they have of collaborating. But again, whatever. This is the point that makes me the most uneasy about a switch. Currently, people can use whatever tool they know best (be it svn, svk, bzr, git, hg or darcs) on top of a svn repository. That makes git users angry since git-svn doesn’t work so well. So instead of improving the said tool, git users want the repository to switch to git instead. Which means it will only be accessible with git (and with bzr using bzr-git). Users of other tools will just be screwed. It’s not as if it mattered to me that much; I’m using tools much crappier than git on a daily basis, and I’m sure I’ll be able to deal with it. But if you really think that casual contributors or translators will contribute as easily as they can today, I can only say “bwahahaha”. I find svn to be still the easiest to use of that list, and while bzr and hg have made laudable efforts in terms of usability, bringing them on par to or better than svn in several areas, the same can’t be said of git. I find it also interesting to see developers of a project for which usability is a prime goal choose the less usable VCS tool at their disposal. Cheers, -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Hi, Behdad Esfahbod wrote: What I told you was that because the switch does not affect those other people. Yes, if your contributors are NOT committing to GNOME SVN, their opinions doesn't matter. Neither does my mom's opinion matter in this case. Nothing wrong with that. This is only true if you don't believe that future contributions to the GNOME base are independent of the RCS we decide to host on gnome.org (should we decide to host only one). If you can identify people contributing to GNOME projects outside gnome.org who would commit directly to gnome.org if their preferred RCS were available, then there is an opportunity cost to saying that their opinion doesn't matter - if you don't consider their opinion, they will never be committing directly to gnome.org (maybe that's OK, maybe it's not). The question which we could ask in this case is: how many people are contributing regularly to GNOME projects, and don't have an account on GNOME SVN? And why don't they? Another question: what is the cost associated with hosting multiple DVCSes? I understand of course that most people don't even have one SCM tool installed on their system. But for those not using a DVCS, moving from 0 to 1 is the big move, and I am not sure that there is a huge cost to having both git and bzr hosting on gnome.org, for official GNOME projects (especially if those who want to use a git client for bzr-hosted projects may do so). snip Yeah, whatever. I can't care less about what you prefer because you are already using it and are happy about it. Keep your personal feelings out of this thread then and everyone will be the happier. If others need their opinions expressed, I think they should do that themselves. Behdad - it's disappointing to see you doing the same thing which you expressed disappointment at this morning. This contributed nothing wasn't a very nice reply. Can you follow your own advice stick to the topic (which is, as you said, giving input to the people who will be doing the work - and to my mind, saying that there are GNOME projects not hosted on gnome.org because of the revision control system is relevant to that discussion). Cheers, Dave. -- Dave Neary GNOME Foundation member dne...@gnome.org ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Dave Neary wrote: Hi, Behdad Esfahbod wrote: What I told you was that because the switch does not affect those other people. Yes, if your contributors are NOT committing to GNOME SVN, their opinions doesn't matter. Neither does my mom's opinion matter in this case. Nothing wrong with that. This is only true if you don't believe that future contributions to the GNOME base are independent of the RCS we decide to host on gnome.org (should we decide to host only one). You are of course right. I probably should also have added that list of SVN users is the only affordable way we could do a closed survey. Simply because that's the only list we have. The only other realistic alternative would have been a public survey with no authentication at all, and I hope we agree that that would have not been more representative than the survey we did. If you can identify people contributing to GNOME projects outside gnome.org who would commit directly to gnome.org if their preferred RCS were available, then there is an opportunity cost to saying that their opinion doesn't matter - if you don't consider their opinion, they will never be committing directly to gnome.org (maybe that's OK, maybe it's not). The question which we could ask in this case is: how many people are contributing regularly to GNOME projects, and don't have an account on GNOME SVN? And why don't they? How do we do that objectively? I can name 10 people that push code to fdo instead of GNOME because of their VCS preference. The next guy can name 10 that push to Launchpad for the same reason. The promise of the survey was to give us solid numbers. How do we do it for anything other than current account holders? Another question: what is the cost associated with hosting multiple DVCSes? I understand of course that most people don't even have one SCM tool installed on their system. But for those not using a DVCS, moving from 0 to 1 is the big move, and I am not sure that there is a huge cost to having both git and bzr hosting on gnome.org, for official GNOME projects (especially if those who want to use a git client for bzr-hosted projects may do so). snip Yeah, whatever. I can't care less about what you prefer because you are already using it and are happy about it. Keep your personal feelings out of this thread then and everyone will be the happier. If others need their opinions expressed, I think they should do that themselves. Behdad - it's disappointing to see you doing the same thing which you expressed disappointment at this morning. This contributed nothing wasn't a very nice reply. Can you follow your own advice stick to the topic (which is, as you said, giving input to the people who will be doing the work - and to my mind, saying that there are GNOME projects not hosted on gnome.org because of the revision control system is relevant to that discussion). Guilty as charged. I took it personal too. My bad. Cheers, behdad Cheers, Dave. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
2009/1/6 Josselin Mouette j...@debian.org: Le mardi 06 janvier 2009 à 17:21 +0900, James Henstridge a écrit : I'd hope that any DVCS would get a larger user base than current list of active Subversion committers: anyone who contributes patches via mailing lists or bugzilla could use a DVCS the same way as a core developer. The same could go for anyone who does distribution packaging. That's foolish. We are certainly not going to clone 208 repositories for all GNOME packages we maintain. Especially when the tools to manage Debian packages with DVCSes are still inferior to those we have for the current svn+tarball scheme. As I mentioned later on in the post you quoted, not everyone in these groups will use a DVCS even if it is available: the existing patch based work flows should work equally well compared to the status quo. That said, there are definitely members of these groups who will use a DVCS if it is available. James. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Hi, Behdad Esfahbod wrote: Dave Neary wrote: snip This is only true if you don't believe that future contributions to the GNOME base are dependent of the RCS we decide to host on gnome.org (should we decide to host only one). You are of course right. I probably should also have added that list of SVN users is the only affordable way we could do a closed survey. Simply because that's the only list we have. The only other realistic alternative would have been a public survey with no authentication at all, and I hope we agree that that would have not been more representative than the survey we did. I think it's perfectly reasonable to consider people with GNOME SVN accounts for a survey. I don't think it's correct to characterise that sample as the only people affected by the change. How you find out about the opinion of others potentially affected by the change, and the extent to which you consider their opinion, is a separate question. The question which we could ask in this case is: how many people are contributing regularly to GNOME projects, and don't have an account on GNOME SVN? And why don't they? How do we do that objectively? I can name 10 people that push code to fdo instead of GNOME because of their VCS preference. The next guy can name 10 that push to Launchpad for the same reason. The promise of the survey was to give us solid numbers. How do we do it for anything other than current account holders? I don't believe there's a way to objectively do that. We have to live with a bit of subjectivity, I'm afraid. A start would be to see how many official GNOME modules have moved their development off gnome.org - a bunch of GNOME maintainers are using either bzr or git-svn these days - and ask them if there are any regular contributors who don't have/want/need an account on gnome.org. Cheers, Dave. -- Dave Neary GNOME Foundation member dne...@gnome.org ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
2009-01-06 klockan 08:21 skrev Max Kanat-Alexander: Git really has no API, you just run the commands and get the output. *Subversion* actually had the best API when I was writing VCI, FWIW. Git and CVS had the worst API, in terms of integration. There may be better modules available now, though--I wrote VCI in 2007. Just an additional: Bazaar has a really nice Python API, but I understand your software was written in Perl, so it couldn't be (easily) used. — Wouter signature.asc Description: Digital signature ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On 01/06/2009 02:09 AM, Ruben Vermeersch wrote: What if this user publishes his branch using bzr (which works fine with the GNOME servers). How will I merge this branch, if I'm using git? It looks to me that with the git+bzr proposal, we're being forced to learn both systems anyway. We can control what's going on on our own servers, but not what happens outside. So don't tell us it's equivalent. It's not. And let's not forget the broad confusion that will arise from the fact that part of the developers advice you to use git and other parts advice you to use bzr. I am exceedingly less convinced that the advantages of supporting both outweigh the disadvantages (looking at the usage patterns, not the technical merits). I was on the fence about this, especially because I prefer bzr to git. But Ruben is right on the money. Implementing DVCS should not raise the barrier for entry for contributors. It should not make it *harder* for maintainers to accept patches/branches. If 70% of GNOME goes one way and 30% goes another way, it will suck for everybody. We need one official VCS for GNOME. The hybrid solution really only makes sense if that one VCS is bzr, which (sadly) seems to go against popular opinion. So assuming we have the manpower to do the git conversion and the tools to make it rock (viewvc, giorious, whtaever), let's JFDI. Man I'm tempted to make a pun about committing right now... Sandy ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Tue, 2009-01-06 at 03:46 -0500, Behdad Esfahbod wrote: So, here is what I'm bringing to the table: I'm volunteering to work with interested fd.o admins and other volunteers to switch GNOME to git. I need to first go check to see if I can secure enough time to lead this, and if I can get enough peoples' attention to make this happen in a timely manner. I will make an official proposal when I figure those out. Given the nature of the GNOME repositories, ie 1 repository per module, and the fact that we have a list of core modules for the various suites, it could be useful to do this one by one using a few volunteers project. In that case I volunteer the one I maintain, namely raw-thumbnailer and niepce[1] to switch them to git. It would allow to have the infrastructure setup and tested while still not blocking GNOME itself from its development pace. Just my 2¢ Hub [1] and for that last one, actually I'd love to fix its history as that SVN export/import totally foobared it, long story short, because subversion is deficient in disallowing import/export without shell access. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Tue, Jan 06, 2009, Behdad Esfahbod wrote: From what I understand, so far there has been one proposal, from Olav, who proposed that he and John implement John's idea of implementing a git-serve plugin for the bzr repo server. I wonder whether you received interesting ideas in the survey itself? As insane as it might sound, I personally wouldn't mind each module picking its VCS. I think common tasks which random contributors need to achieve can be documented for all VCS-es (checkout, do some changes and commit or create a patch). Just like various modules are using various programming languages or even build systems. -- Loïc Minier ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Tue, 2009-01-06 at 21:01 +0100, Loïc Minier wrote: On Tue, Jan 06, 2009, Behdad Esfahbod wrote: From what I understand, so far there has been one proposal, from Olav, who proposed that he and John implement John's idea of implementing a git-serve plugin for the bzr repo server. I wonder whether you received interesting ideas in the survey itself? As insane as it might sound, I personally wouldn't mind each module picking its VCS. I think common tasks which random contributors need to achieve can be documented for all VCS-es (checkout, do some changes and commit or create a patch). Just like various modules are using various programming languages or even build systems. Each app picking its VCS seems better than the proposed system with both bzr and git. Because with the proposal you can pick any vcs you like as a user, but if you didn't pick the one the maintainer used then he and the other developers can't pull from you and you're left out on your own development island. So, all modules would anyway need to marks out what the prefered vcs for it is and all developers would have to learn both. Of course, all that could be avoided if we just decided on one... ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Loïc Minier wrote: On Tue, Jan 06, 2009, Behdad Esfahbod wrote: From what I understand, so far there has been one proposal, from Olav, who proposed that he and John implement John's idea of implementing a git-serve plugin for the bzr repo server. I wonder whether you received interesting ideas in the survey itself? All the survey responses (including all the comments, but excluding svn user names) are out there for everyone to browse. As insane as it might sound, I personally wouldn't mind each module picking its VCS. I think common tasks which random contributors need to achieve can be documented for all VCS-es (checkout, do some changes and commit or create a patch). Just like various modules are using various programming languages or even build systems. This idea was discussed when the DVCS thing was brought up first. I think Olav was against it for sysadmin workload reasons. Every VCS-facing service needs to be implemented for all the supported systems and given that we don't seem to have enough man hours to do one, doing multiple ones just seems unreachable. behdad ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
2009/1/6 Alexander Larsson al...@redhat.com: On Tue, 2009-01-06 at 21:01 +0100, Loïc Minier wrote: On Tue, Jan 06, 2009, Behdad Esfahbod wrote: From what I understand, so far there has been one proposal, from Olav, who proposed that he and John implement John's idea of implementing a git-serve plugin for the bzr repo server. I wonder whether you received interesting ideas in the survey itself? As insane as it might sound, I personally wouldn't mind each module picking its VCS. I think common tasks which random contributors need to achieve can be documented for all VCS-es (checkout, do some changes and commit or create a patch). Just like various modules are using various programming languages or even build systems. Each app picking its VCS seems better than the proposed system with both bzr and git. Because with the proposal you can pick any vcs you like as a user, but if you didn't pick the one the maintainer used then he and the other developers can't pull from you and you're left out on your own development island. So, all modules would anyway need to marks out what the prefered vcs for it is and all developers would have to learn both. Am I the only one crying on how bad and confusing is this going to be for newcomers? One of the most obvious ways to contribute to free software these days it to do it on the people's most used apps, which are the desktop apps (translator as an example). There are guys out there who doesn't even get the point of VCS and their first approach to them is going to be GNOME itself. Think about them trying to browse for information on how to create my first patch, and this section saying you have to figure out which project are you gonna pick, and then, learn to use it. To be honest, I think that this discussion would just go away if we had Tortoise like apps integrated with jhbuild for Nautilus where you have a common set of graphical tools to do the most common work for 90% of the VCS users. Then, and only then, we could stop worrying about which VCS do we choose, since we won't have to fiddle with any command line (for god's sake, we're in 2009 already). Of course, all that could be avoided if we just decided on one... ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list -- Un saludo, Alberto Ruiz ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Tue, 2009-01-06 at 21:30 +, Alberto Ruiz wrote: 2009/1/6 Alexander Larsson al...@redhat.com: On Tue, 2009-01-06 at 21:01 +0100, Loïc Minier wrote: On Tue, Jan 06, 2009, Behdad Esfahbod wrote: From what I understand, so far there has been one proposal, from Olav, who proposed that he and John implement John's idea of implementing a git-serve plugin for the bzr repo server. I wonder whether you received interesting ideas in the survey itself? As insane as it might sound, I personally wouldn't mind each module picking its VCS. I think common tasks which random contributors need to achieve can be documented for all VCS-es (checkout, do some changes and commit or create a patch). Just like various modules are using various programming languages or even build systems. Each app picking its VCS seems better than the proposed system with both bzr and git. Because with the proposal you can pick any vcs you like as a user, but if you didn't pick the one the maintainer used then he and the other developers can't pull from you and you're left out on your own development island. So, all modules would anyway need to marks out what the prefered vcs for it is and all developers would have to learn both. Am I the only one crying on how bad and confusing is this going to be for newcomers? One of the most obvious ways to contribute to free software these days it to do it on the people's most used apps, which are the desktop apps (translator as an example). There are guys out there who doesn't even get the point of VCS and their first approach to them is going to be GNOME itself. Think about them trying to browse for information on how to create my first patch, and this section saying you have to figure out which project are you gonna pick, and then, learn to use it. Welcome to the open source world. Generally open source developers are not limited to GNOME, and they eventually learn 2-3 revision control systems. I mean, they don't need to learn a lot of commands of each RCS, just the basics: 1- Checkout/clone module; 2- Update; 3- Create a diff of changes, redirecting to a patch file. With those three operations alone is enough to cover 90% of all open source developers needs when they are contributing to a project to which they have no commit privileges. And those two operations alone are dead easy to learn for a number of VCSs. Developers with commit privileges need to know just a few more commands: 1- commit 2- push (if applicable) And finally maintainers need to know a few more commands, like branching, merging, and tagging. But they only need to learn those for the modules they maintain. To be honest, I think that this discussion would just go away if we had Tortoise like apps integrated with jhbuild for Nautilus where you have a common set of graphical tools to do the most common work for 90% of the VCS users. Then, and only then, we could stop worrying about which VCS do we choose, since we won't have to fiddle with any command line (for god's sake, we're in 2009 already). Disagreed. Command line interface is often more productive than any GUI for certain things. I don't use any IDE for source code editing, and I would certainly hate to have to use Nautilus for commits. Not to say that GUIs aren't useful for some things. I have seen nice GUIs for bazaar, git, and mercurial. But they are most useful for visualizing the repository tree, rather than operations that change the repository, it seems to me. Regards, -- Gustavo J. A. M. Carneiro g...@inescporto.pt gust...@users.sourceforge.net The universe is always one step beyond logic -- Frank Herbert ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Emmanuele Bassi wrote: [0] I'm reasonably sure it has some. not as the one proposed to avoid pissing off somebody somewhere because we want to be inclusive -- no, lemme rephrase that: we are *fucking afraid of committment*. seriously: an abstraction over DVCS? what have we become? are we *ever* going make *any* decision about *anything*? this is actually a larger issue with the GNOME community: we are being afraid. Seriously. If git via bzr server had been listed as an option on the survey, I would have ranked it dead last (after just bzr, and after stay with svn). It seems pretty clear that the git-over-bzr solution doesn't make the git users any happier than git-over-svn does, so let's not pretend that it's any different from doing just bzr. So, given that we seem to have sysadmin resources to do bzr, but we don't have sysadmin resources to do git, the question we really need to answer is do you prefer migrating to bzr, or sticking with svn?. -- Dan ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On 1/5/09, Johannes Schmid j...@jsschmid.de wrote: 6. Check all the documentation stuff on live.gnome.org that needs to be updated. That is really important because not everybody is familiar with git. There should also be a short introduction to git somewhere on the wiki. And some announcements should probably be made... And perhaps explain the benefits and cool stuff, if we are moving to !svn, we should take advantage of the new cool stuff introduced... that's where something like Federico's proposal to use gitorious fit. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Mon, Jan 5, 2009 at 5:22 PM, Olav Vitters o...@bkor.dhs.org wrote: On Sun, Jan 04, 2009 at 06:34:47PM -0500, David Zeuthen wrote: On Mon, 2009-01-05 at 00:18 +0100, Olav Vitters wrote: I am not evading. Stop trying to make this personal. I don't care about CoC, I don't like you're talking to me. Please. Stop trying to make this look like it's personal and like I'm assaulting you. Because I didn't. And I resent the accusation. You ask Is it *really* so hard to understand that this whole git-serve is a terrible idea. As I don't find it terrible, the statement makes me feel like I'm considered an idiot for disagreeing. Then when I try to point that out, I get Oh, you chose not to quote that... anyway, I'm dropping this. Guys please both exchange a series of yo mama jokes in private then have a beer and shake your hands. It's obvious some of us get excited as it *is* a beauty contest we're participating in even if we paint it as a survey. It's normal and it's fine, just don't let personal taste win over GNOME. We're mostly engineers and we're here to build bike sheds, not to paint them. :) I really really like git but I couldn't care less if we switched to any of the things I picked over svn ;) -- Patryk Zawadzki ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On ma, 2009-01-05 at 12:32 -0500, Diego Escalante Urrelo wrote: On 1/5/09, Johannes Schmid j...@jsschmid.de wrote: 6. Check all the documentation stuff on live.gnome.org that needs to be updated. That is really important because not everybody is familiar with git. There should also be a short introduction to git somewhere on the wiki. And some announcements should probably be made... And perhaps explain the benefits and cool stuff, if we are moving to !svn, we should take advantage of the new cool stuff introduced... that's where something like Federico's proposal to use gitorious fit. Yes, rather than fighting over the backend storage format, let's focus on making our developer experience better. Having a gitorious instance for all of GNOME (our own github, powered by free open-source software [1]), will make collaboration much easier, as well as cover infrastructure problems. [1]: http://gitorious.org/projects/gitorious -- Ruben Vermeersch (rubenv) http://www.savanne.be ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Mathias Hasselmann wrote: Am I missing something? Make build.gnome.org work with the new setup; I plan to write some requirements (nothing fancy, and stuffs that will most probably be also required elsewhere). There may also be some other infrastructure systems that would require some porting (as far as I am concerned, library-web will not require changes). Frederic ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Josselin Mouette wrote: BTW, do we have the resources to migrate the repository to the SVN 1.5 format? It looks like, independently from other decisions, a quick and easy way to improve the situation – and to improve it right now, not in 2010. Olav announced it on October 29th: http://mail.gnome.org/archives/devel-announce-list/2008-October/msg4.html Cheers, Frederic ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Hi, On Sun, Jan 4, 2009 at 7:51 AM, Olav Vitters o...@bkor.dhs.org wrote: Anyway, I'd rather add John Carr to the sysadmin team. I plan to make a proposal to switch GNOME to a DVCS where Git works using Johns suggestion. Then other sysadmins[1] can suggest whatever proposal they want. These proposals can be investigated on merit and then a one can be chosen (chosen as in: go ahead and try if this would work, not go ahead blindly; everything must be tested before a cutover). [1] or whomever. Although I don't see how that would work. While I'm sure John will at least be able to get basic functionality working, and the project has a certain amount of cool geek factor, taking John's proposal as a path forward concerns many in the community for a variety of reasons[*1]. (In fact, I bet such an option would rank lower than any native vcs option had it been included in the survey.) I'd like to help with another path forward, namely native git repositories since I believe that is what most of the community wants. As you said, it isn't clear how it could work for non-sysadmins to come up with clear proposal strategies and implementations. Are there others on the sysadmin team who are willing to work on such a transition? If so, how can I help? Elijah [*1] Reasons I've seen or can think of off the top of my head: * As James H. mentioned on John's blog, you'd likely end up with the intersection of the features of the two version control systems rather than improving things. * John's project does not have a large community behind it and supporting it. In fact, it may end up with a bus factor of 1[*2]. Even if it increases, it doesn't have the kind of large community that, say, git-svn has. In general, it's unsettling to many to adopt a project without a large community behind it. * John's bridge would have to be updated whenever either the bzr or git formats changed (in particular, bzr has changed repository formats several times and even promotes it's ability to seamlessly change repository formats as an advantage), or whenever the network protocols changed (including protocol extensions, such as the git push tell-me-more extension). * It would introduce extra lag between when new features become available, since the bridge would need to be updated for each such change. * There's no guarantee bzr and git will change in ways that will make them remain compatible, so we run the risk of accepting (additional) feature losses as time goes on. It may be a small risk, but we simply don't know and have no way of knowing. * All software has bugs. John's bridge can't be exempt, and particularly as new and not-yet-tested software, it's more of a risk. Will that mean data loss? Loss of features? Inability to perform certain operations? While the bugs are being investigated and fixed, what do maintainers do? Use bzr since it's the official format? I think John's pretty clever and that we would likely avoid most such issues -- but there's no guarantee and this is something that affects developers daily work. * I believe bzr proponents even admit that bzr is still slow for network operations. John's bridge would essentially add another layer on top of that. [*2] http://en.wikipedia.org/wiki/Bus_factor ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Le lundi 05 janvier 2009 à 09:51 -0500, Dan Winship a écrit : It seems pretty clear that the git-over-bzr solution doesn't make the git users any happier than git-over-svn does, so let's not pretend that it's any different from doing just bzr. So, given that we seem to have sysadmin resources to do bzr, but we don't have sysadmin resources to do git, the question we really need to answer is do you prefer migrating to bzr, or sticking with svn?. BTW, do we have the resources to migrate the repository to the SVN 1.5 format? It looks like, independently from other decisions, a quick and easy way to improve the situation – and to improve it right now, not in 2010. -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Hi! Am Montag, den 05.01.2009, 16:23 +0100 schrieb Mathias Hasselmann: First of all I want to thank Behdad and the participants of the survey for giving us numbers. Second I want to complain about the direction this discussion takes. No idea why that many people become personal. This is really unpleasant. Third of all: What so complicated about this migration? As far as I see, the migration consists of the following steps, please tell me if I am too naive: 1. Identify admin scripts that must be ported from svn to git. So far I only know new-svn-repos. 2. Identify commit hooks which have to be ported. Should only global hooks be ported, or would the migration team also be responsible for porting module specific hooks? 3. Actually port the commit hooks. 4. Create snapshots of all SVN repositories using git-svn. 5. Now finish one repository after another: 1. Mark the SVN repository as read-only. 2. Run a final git-svn rebase. 3. Maybe strip git-svn information. 4. Install commit hooks. 5. Test the new git repository. 6. Make the new git repository public. Am I missing something? 6. Check all the documentation stuff on live.gnome.org that needs to be updated. That is really important because not everybody is familiar with git. There should also be a short introduction to git somewhere on the wiki. And some announcements should probably be made... Regards, Johannes signature.asc Description: Dies ist ein digital signierter Nachrichtenteil ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Mon, Jan 5, 2009 at 1:18 PM, Felipe Contreras felipe.contre...@gmail.com wrote: 2009/1/5 Ali Sabil ali.sa...@gmail.com: On Sun, Jan 4, 2009 at 11:33 PM, Olav Vitters o...@bkor.dhs.org wrote: On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. Sorry for not being clear in my explanations. Basically, as Olav pointed out, it is about having Bazaar on the server, with a git-serve plugin allowing it to fulfill the git client requests as well as the bzr client requests. The following scenarios will be possible: (bzr repo) - (git serve plugin) - network --- (git client) (bzr repo) - (bzr serve) - network --- (bzr client) both bzr and git will operate fully, nothing will be partially supported, since the bazaar repository format is a superset of the git repo format (ie. it stores more metadata). I talked about hg, just to highlight that the solution is quite future proof, because you can certainly apply the same solution to allow hg clients to access the repository. First of all, who is going to develop and maintain the git serve plugin? Whoever does it I bet the end result won't be as good as the native git. Emulators tend to behave differently from the native counterpart. Second, as David mentioned; what would happen in the case the git protocol is updated and backward compatibility is removed? We will need to wait until the git serve plugin is updated, possibly rewritten. Third, every repository format has advantages and drawbacks. So far it looks like the git repository format works for most people, what is the need to avoid it? Fourth, we should not re-invent the wheel, people use either bzr or git, and not both for a reason; depending on a theoretical git serve plugin is just asking for trouble. The way I understood the proposal, bazaar would be the official dvcs and a usable- albeit officially unsupported- git wrapper would be provided. Assuming that a future version of git doesn't introduce incompatibilities, the approach has the advantage of being an easy solution which works for all git and bazaar users. If a future version of git _is_ incompatible, the official bazaar access would be totally unaffected. That said, according to the survey most people use git. Most of those users don't care about bazaar access at all, but might be slightly irritated if there are any quirks with the git wrapper. If you'd like to try to make everyone happy then the wrapper approach has it's advantages. If you'd rather make a small group slightly annoyed and a bigger group totally happy then go with git. Fifth, if the majority of the GNOME community prefers git, why degrade the git experience with an emulation? It makes much more sense for the bzr minority to emulate bzr experience with bzr-git if so they desire. -- Felipe Contreras ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list -Natan ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
2009/1/5 Ali Sabil ali.sa...@gmail.com: On Sun, Jan 4, 2009 at 11:33 PM, Olav Vitters o...@bkor.dhs.org wrote: On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. Sorry for not being clear in my explanations. Basically, as Olav pointed out, it is about having Bazaar on the server, with a git-serve plugin allowing it to fulfill the git client requests as well as the bzr client requests. The following scenarios will be possible: (bzr repo) - (git serve plugin) - network --- (git client) (bzr repo) - (bzr serve) - network --- (bzr client) both bzr and git will operate fully, nothing will be partially supported, since the bazaar repository format is a superset of the git repo format (ie. it stores more metadata). I talked about hg, just to highlight that the solution is quite future proof, because you can certainly apply the same solution to allow hg clients to access the repository. First of all, who is going to develop and maintain the git serve plugin? Whoever does it I bet the end result won't be as good as the native git. Emulators tend to behave differently from the native counterpart. Second, as David mentioned; what would happen in the case the git protocol is updated and backward compatibility is removed? We will need to wait until the git serve plugin is updated, possibly rewritten. Third, every repository format has advantages and drawbacks. So far it looks like the git repository format works for most people, what is the need to avoid it? Fourth, we should not re-invent the wheel, people use either bzr or git, and not both for a reason; depending on a theoretical git serve plugin is just asking for trouble. Fifth, if the majority of the GNOME community prefers git, why degrade the git experience with an emulation? It makes much more sense for the bzr minority to emulate bzr experience with bzr-git if so they desire. -- Felipe Contreras ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Mon, 2009-01-05 at 11:23 +0100, Edward Hervey wrote: On Sun, 2009-01-04 at 20:32 +, Gustavo J. A. M. Carneiro wrote: On Sat, 2009-01-03 at 22:46 -0500, Behdad Esfahbod wrote: In December I ran a distributed version control system survey for GNOME. From the survey opening page: Thank you for taking the GNOME DVCS Survey. This survey is run on behalf of the GNOME Foundation board of directors, release team, and sysadmin team. The GNOME project is planning a possible move from SVN to a distributed version control system in 2009. The contenders for the system to use are bzr, git, and hg. The aim of the survey is to help us better understand familiarity and preferences of our active contributor base regarding the future version control system for GNOME. The survey results will be informational and will be sent to foundation-list and desktop-devel-list upon completion. GNOME contributors with an SVN account who had an SSH key installed on their account were invited to fill in the survey. A total of 1083 account holders were invited, and 579 filled in the survey. The survey results are now available to the public: http://www.gnome.org/~behdad/dvcs-survey/ Elijah Newren did an initial analysis of the data. His analysis also includes the survey questions and answers. Find it at: http://blogs.gnome.org/newren/2009/01/03/gnome-dvcs-survey-results/ If you analyze the results, please reply to this thread and also leave a comment on my blog post linking to your analysis: http://mces.blogspot.com/2009/01/gnome-dvcs-survey.html Just in case I am forced to switch to git in the future (being open minded here, although I prefer bazaar), does someone have any advice how to generate a nice GNU style ChangeLog (like what emacs produces) from git commit logs? I know that some projects like Cairo auto-generate ChangeLog already, but the default git changelog format is too detailed/ugly IMHO. Heya, I wrote a python script for PiTiVi (now that we've switch to git) that does just that and which we run at (pre-)release-time to generate the ChangeLog file: http://git.pitivi.org/?p=pitivi.git;a=blob_plain;f=docs/makeChangelog.py;hb=HEAD Heh, thanks a lot. This looks nice. Nicer than the one in gnulib that Rui Tiago pointed out. Although I must say not as nice as my 'gnulog' bazaar log formatter plugin.. ;-) But I guess good enough that I'd be comfortable replacing a hand written ChangeLog with autogenerated one. Regards, -- Gustavo J. A. M. Carneiro g...@inescporto.pt gust...@users.sourceforge.net The universe is always one step beyond logic -- Frank Herbert ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 20:32 +, Gustavo J. A. M. Carneiro wrote: On Sat, 2009-01-03 at 22:46 -0500, Behdad Esfahbod wrote: In December I ran a distributed version control system survey for GNOME. From the survey opening page: Thank you for taking the GNOME DVCS Survey. This survey is run on behalf of the GNOME Foundation board of directors, release team, and sysadmin team. The GNOME project is planning a possible move from SVN to a distributed version control system in 2009. The contenders for the system to use are bzr, git, and hg. The aim of the survey is to help us better understand familiarity and preferences of our active contributor base regarding the future version control system for GNOME. The survey results will be informational and will be sent to foundation-list and desktop-devel-list upon completion. GNOME contributors with an SVN account who had an SSH key installed on their account were invited to fill in the survey. A total of 1083 account holders were invited, and 579 filled in the survey. The survey results are now available to the public: http://www.gnome.org/~behdad/dvcs-survey/ Elijah Newren did an initial analysis of the data. His analysis also includes the survey questions and answers. Find it at: http://blogs.gnome.org/newren/2009/01/03/gnome-dvcs-survey-results/ If you analyze the results, please reply to this thread and also leave a comment on my blog post linking to your analysis: http://mces.blogspot.com/2009/01/gnome-dvcs-survey.html Just in case I am forced to switch to git in the future (being open minded here, although I prefer bazaar), does someone have any advice how to generate a nice GNU style ChangeLog (like what emacs produces) from git commit logs? I know that some projects like Cairo auto-generate ChangeLog already, but the default git changelog format is too detailed/ugly IMHO. Heya, I wrote a python script for PiTiVi (now that we've switch to git) that does just that and which we run at (pre-)release-time to generate the ChangeLog file: http://git.pitivi.org/?p=pitivi.git;a=blob_plain;f=docs/makeChangelog.py;hb=HEAD Edward ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
First of all I want to thank Behdad and the participants of the survey for giving us numbers. Second I want to complain about the direction this discussion takes. No idea why that many people become personal. This is really unpleasant. Third of all: What so complicated about this migration? As far as I see, the migration consists of the following steps, please tell me if I am too naive: 1. Identify admin scripts that must be ported from svn to git. So far I only know new-svn-repos. 2. Identify commit hooks which have to be ported. Should only global hooks be ported, or would the migration team also be responsible for porting module specific hooks? 3. Actually port the commit hooks. 4. Create snapshots of all SVN repositories using git-svn. 5. Now finish one repository after another: 1. Mark the SVN repository as read-only. 2. Run a final git-svn rebase. 3. Maybe strip git-svn information. 4. Install commit hooks. 5. Test the new git repository. 6. Make the new git repository public. Am I missing something? * Steps one and two have to be done by the current SVN admins. * Step three is a programming task and therefore can be done by each GNOME hacker knowing the programming languages used. * Step four rounds automatically and just needs some watching. * Step five could be done in parallel. So is it really true, that we don't have the man power to do this migration? I cannot believe this. Ciao, Mathias -- Mathias Hasselmann mathias.hasselm...@gmx.de http://taschenorakel.de/mathias/, http://www.openismus.com/ ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 04, 2009 at 06:34:47PM -0500, David Zeuthen wrote: On Mon, 2009-01-05 at 00:18 +0100, Olav Vitters wrote: I am not evading. Stop trying to make this personal. I don't care about CoC, I don't like you're talking to me. Please. Stop trying to make this look like it's personal and like I'm assaulting you. Because I didn't. And I resent the accusation. You ask Is it *really* so hard to understand that this whole git-serve is a terrible idea. As I don't find it terrible, the statement makes me feel like I'm considered an idiot for disagreeing. Then when I try to point that out, I get Oh, you chose not to quote that... anyway, I'm dropping this. I'm not saying you're assaulting me (way too strong). I just don't like the tone. but suggest offline if needed (in hindsight, probably should've done that right away). Only replying on list to make clear to everyone that I didn't think you were assaulting. -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Ali Sabil a écrit : That's not what John's proposal is about ! John wants to use the bzr format as a repository format, and add a git-serve plugin to bzr to be able to talk to the git clients. In other words, you will be able to access the same data using either bzr, git or hg. Well, if people say git and your answer is bzr with a git frontend, then why don't we stick with svn on the server and let people use git-svn if they please ? When I want to ride my bike, I don't install a new pedal system on my car : I just use my bike! Snark PS: notice that my last comment in the survey was already something like this : if switching is too costly or people can't agree on what to switch to, then let's keep svn and let people use git-svn, hg-svn, bzr-svn, etc-svn. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Mon, Jan 05, 2009 at 07:00:52AM -0700, Elijah Newren wrote: I'd like to help with another path forward, namely native git repositories since I believe that is what most of the community wants. As you said, it isn't clear how it could work for non-sysadmins to come up with clear proposal strategies and implementations. Are there others on the sysadmin team who are willing to work on such a transition? If so, how can I help? Don't know if there are other sysadmins who'd work on this. I've cc'ed gnome-sysadmin so that people can answer themselves instead of me guessing. I'll let John reply on all other questions. -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sat, 2009-01-03 at 22:46 -0500, Behdad Esfahbod wrote: GNOME contributors with an SVN account who had an SSH key installed on their account were invited to fill in the survey. [It is NOT my intention to get all negative here; I understand - and accept - that projects make decisions and not everyone is happy with them. Luckily this decision ultimately is one I can ignore. Nevertheless, I have been asked by a number of people to write to this thread with why I am so dissatisfied. I do appreciate the effort people made, even if I feel that the way the whole survey exercise was conducted it was impossible for Git to lose] Some comments: ++ It's a shame that hackers who contribute to GNOME projects which don't use svn.gnome.org were excluded. (I was told their opinions didn't matter. {shrug} that's fine, so long as nobody tries to represent this survey as what GNOME hackers think) ++ It was also a shame that I (one who does happen to have a GNOME svn account) was not able to complete the survey either because it crashes Epiphany when you i) vote for bzr and ii) withhold your vote from git, hg, and svn. (When I asked if it might be possible to fix the survey so that GNOME's web browser didn't crash, I was told known bug and too bad, you have to express a preference for Git and Mercurial even if you don't want to. Strange take on democracy. I am rather accustomed to the idea that declining to express a preference for something is an acceptable form of voting. Whatever) I explicitly did not want to chose Git or Mercurial, because I knew exactly what was going to happen. I've heard it several times already in #gnome-hackers and elsewhere: so it seems the people who prefer Bazaar like Git as their second choice, so surely it's ok to go with that. Great! Decision made No. The rest of the survey was irrelevant. It was quite evident that the object of the exercise was to allow people to say lots of people said Git was either their first or second choice which sounds very impressive, and was exactly the one thing I did NOT want to support. So it crashed my browser. Nice. ++ We chose the Bazaar decentralized version control system for our GNOME project even before the people behin GNOME's centralized code hosting made the courageous and monumental decision to switch from CVS to Subversion. Since GNOME didn't offer any way for us to host our 'mainline' branch on any official sounding resource, {shrug} we didn't. And so we don't. And that's actually the only issue that matters so far as I can tell. No one can force us to stop using Bazaar. People who work at places like Immendio who are using Git to hack on GTK+ cannot be forced to stop git either. And I wouldn't want them to. They're happy with their tool. We're happy with ours. When CVS was the only interchange (actually, that's not true, since the real interchange for most projects is attachments to Bugzilla of all things), then indeed GNOME switching to Subversion was a big deal. But in the era of distributed version control, the next step really matters little. Whatever GNOME _infrastructure_ offers next in terms of hosting is really quite irrelevant, since quite anyone can host their own projects and publish their own branches with nothing more than a vanilla web server. If the choice had happened to be Bazaar, then we probably would have moved our principle copy of our 'mainline' branch there. That would have been nice but otherwise is inconsequential since hosting the primary 'mainline' somewhere else costs us nothing, and I long since offered other people accounts to publish their own branches there too. But since it's going to be Git, well, it offers nothing for us. If the choice had been the other way around, then Git people would simply continue to host their branches somewhere else as they already area. Again, no change. This is ultimately why the whole debate is a bit pointless. Regardless, GNOME is not switching to anything. If GNOME infrastructure is going to offer Git hosting, that's lovely for people who chose to use Git as their version control system. {shrug} fine. If GNOME infrastructure concurrently disables their Subversion hosting and/or people stop pushing their changes there, then that's perhaps a bit worse, because it means people in all three systems (+ svn makes four) will lose the easy way they have of collaborating. But again, whatever. ++ I will close by saying that switching to Bazaar was an unbelievable breath of fresh air after so much pain using Git. I wrote about that briefly here: http://research.operationaldynamics.com/blogs/andrew/software/version-control/git-is-like-cvs.html I've been using DVCS systems for a long long time. I have great respect for all the groups who have worked on the 3rd generation tools. Unfortunately I have no sympathy for Git anymore, and am tired of it screwing over people
Re: GNOME DVCS Survey Results
Hi, [Disclaimer: I wasn't involved in the construction or running of the survey, other than the analysis you saw plus some late feedback on the survey questions (I think my feedback was merely to suggest the other answer for contributor types.)] 2009/1/5 Andrew Cowie and...@operationaldynamics.com: On Sat, 2009-01-03 at 22:46 -0500, Behdad Esfahbod wrote: GNOME contributors with an SVN account who had an SSH key installed on their account were invited to fill in the survey. [It is NOT my intention to get all negative here; I understand - and accept - that projects make decisions and not everyone is happy with them. Luckily this decision ultimately is one I can ignore. Nevertheless, I have been asked by a number of people to write to this thread with why I am so dissatisfied. I do appreciate the effort people made, even if I feel that the way the whole survey exercise was conducted it was impossible for Git to lose] Thanks for taking the time to do so. I'm sorry you feel that way about the survey exercise; I was encouraged that people were moving forward and that they had created what I felt was an unbiased survey as possible (Behdad asked for feedback from the foundation board, release team, and others before sending it out, and I think my main comment at the time was that I was happy to see the lack of bias in the survey; my only other comment was the other contributor thing, IIRC.) Some comments: ++ It's a shame that hackers who contribute to GNOME projects which don't use svn.gnome.org were excluded. (I was told their opinions didn't matter. {shrug} that's fine, so long as nobody tries to represent this survey as what GNOME hackers think) How would you draw the line? Who should be included and who shouldn't? And how do we contact them all? I think doing a survey of any group other than those with svn commit access would be practically unmanageable...and far more likely to be suspected of non-representative-ness [New word!]. Also, if users aren't using GNOME svn then why would they care if we switch or not? Shouldn't the survey poll those whom it would affect? ++ It was also a shame that I (one who does happen to have a GNOME svn account) was not able to complete the survey either because it crashes Epiphany when you i) vote for bzr and ii) withhold your vote from git, hg, and svn. (When I asked if it might be possible to fix the survey so that GNOME's web browser didn't crash, I was told known bug and too bad, you have to express a preference for Git and Mercurial even if you don't want to. Strange take on democracy. I am rather accustomed to the idea that declining to express a preference for something is an acceptable form of voting. Whatever) I explicitly did not want to chose Git or Mercurial, because I knew exactly what was going to happen. I've heard it several times already in #gnome-hackers and elsewhere: so it seems the people who prefer Bazaar like Git as their second choice, so surely it's ok to go with that. Great! Decision made No. The rest of the survey was irrelevant. It was quite evident that the object of the exercise was to allow people to say lots of people said Git was either their first or second choice which sounds very impressive, and was exactly the one thing I did NOT want to support. So it crashed my browser. Nice. That sucks. Big time. However, you'll be happy to know that most of my analyzing work was originally performed on an alternative output file that included all partial answers. (It also contained a bit more data, such as svn usernames -- and thus I can verify that your response was included in this file and just did so.) Now the reason this is relevant is that when I got the final data in an alternate format from Behdad, I had already generated all my plots and written my analysis. So I had to regenerate all the plots and compare old and new versions. It turns out the two sets were basically indistinguishable to my eye. I spot checked a couple of my claims in my analysis (e.g. that translators preferred git over svn since it was so close in both data sets), but actually didn't check them all. Thus, you could say that my analysis is more valid (or at least more verified) for the set of users that also includes partial answers like yours. On a related note, it looks like the total number of people who ranked the various systems (taken from the data file including partial answers) are: any: 581 bzr: 585 git: 583 hg: 581 svn: 582 So, yes, it looks like there were more people who left git unranked than bzr -- the difference being two people. However, I'm a bit confused by this statement: No. The rest of the survey was irrelevant. It was quite evident that the object of the exercise was to allow people to say lots of people said Git was either their first or second choice which sounds very
Re: GNOME DVCS Survey Results
On Sun, 04 Jan 2009 17:40:18 -0500 David Zeuthen da...@fubar.dk wrote: Then what happens when a new version of git with a new feature, incompatible with the git-serve kludge, is released? I don't know if you've talked to the git developers, but they're very firmly against adding any new features to the git server. It is supposed to do one thing and one thing only, and that's allow you to clone from it. The server basically consists of a mechanism to tarball a repository, send it over the wire, and untar it on the client side. -Max -- http://www.everythingsolved.com/ Competent, Friendly Bugzilla and Perl Services. Everything Else, too. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
So, I'm not a GNOME contributor, but I am the author of VCI, a Perl module that interacts with version control systems (currently CVS, Svn, Hg, Git, and Bzr), and so I wanted to chime in a bit on this thread. The first thing to understand is that the git server protocol is very simple. When I was writing VCI, I talked to the git developers and asked if they were going to add any more features to the server. They said no, it's only supposed to allow you to clone from it, and that adding any other features would increase the attack surface (a particularly valid concern when writing servers in C). Now, I am a very definite bzr user, so that should be kept in mind, but I also have a lot of experience with various VCSes, so I do have somewhat of an objective viewpoint. On Mon, 5 Jan 2009 07:00:52 -0700 Elijah Newren new...@gmail.com wrote: * As James H. mentioned on John's blog, you'd likely end up with the intersection of the features of the two version control systems rather than improving things. For most operations, no, that's not true. git does not do any actual operations over the wire, it does all of its operations on your local repository, so you wouldn't lose any git functionality, there. * John's bridge would have to be updated whenever either the bzr or git formats changed Unless he uses bzrlib or calls the bzr/git binaries directly, which I imagine he would? * It would introduce extra lag between when new features become available, since the bridge would need to be updated for each such change. Well, there will be no new features in the git server. git never works remotely, it always works locally. So that wouldn't be a problem there. * I believe bzr proponents even admit that bzr is still slow for network operations. It depends. It's more of a latency thing currently than anything else. With a low-latency connection, I find it just as fast as Git or Hg. And I haven't personally experienced the low-latency thing recently, just heard reports. One question that should also be considered in this whole discussion is--what sysadmin is going to maintain the server? If you can't get anybody who will actually *maintain* a Git repository (I mean, you can point out tools or talk about it, but...) for the next 10 years or so, then it'd be pretty hard to move to it. That includes maintaining all the integration around the repository. Remember that Git really has no API, you just run the commands and get the output. *Subversion* actually had the best API when I was writing VCI, FWIW. Git and CVS had the worst API, in terms of integration. There may be better modules available now, though--I wrote VCI in 2007. -Max -- http://www.everythingsolved.com/ Competent, Friendly Bugzilla and Perl Services. Everything Else, too. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Elijah Newren did an initial analysis of the data. His analysis also includes the survey questions and answers. Find it at: http://blogs.gnome.org/newren/2009/01/03/gnome-dvcs-survey-results/ This is pretty decent analysis going on here :) I'd like to remind people of John Carr's recent blog post too, someone mentioned in the survey results actually. JC has been working on bzr with git protocol support, which would fulfil many of the requirements for having a GNOME DVCS. Happy new year everyone :) BR, K ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 04, 2009 at 08:10:21AM -0600, Jason D. Clinton wrote: This is pretty decent analysis going on here :) I'd like to remind people of John Carr's recent blog post too, someone mentioned in the survey results actually. JC has been working on bzr with git protocol support, which would fulfil many of the requirements for having a GNOME DVCS. I'd like to point out that--of the 15 people who regularly use git and bzr--git still won. That isn't a contest. It is a survey. http://www.gnome.org/~shaunm/survey/first-picks-permutations.png It seems to me that a lot of brain power, sysadmin time, and general I am a sysadmin and disagree with your notion that sysadmin time is somehow saved. I'd rather asses such things myself. Further, sysadmin time is not so important. proliferation of Things To Learn for New People(tm) can be saved if the six people (1.04% of respondents) who ranked bzr above git in that graph can just bite the bullet and admit that git won. Can we please It is a survey. It is NOT about 'winning'. just all move on? Further, your explanation is incomplete. As you said, the graph is about people knowing two DVCS systems. I wouldn't say I knew 2. Those 6 are incomplete. Now before you reply: we have a clear need for git to work (ranked 1st 50% of the time, etc). But if you say move on, how do you think a switch is made? Magic? Anyway, I'd rather add John Carr to the sysadmin team. I plan to make a proposal to switch GNOME to a DVCS where Git works using Johns suggestion. Then other sysadmins[1] can suggest whatever proposal they want. These proposals can be investigated on merit and then a one can be chosen (chosen as in: go ahead and try if this would work, not go ahead blindly; everything must be tested before a cutover). [1] or whomever. Although I don't see how that would work. -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 8:51 AM, Olav Vitters o...@bkor.dhs.org wrote: That isn't a contest. It is a survey. Please don't read more in to my email than I intended. There's no need to get defensive. http://www.gnome.org/~shaunm/survey/first-picks-permutations.png It seems to me that a lot of brain power, sysadmin time, and general I am a sysadmin and disagree with your notion that sysadmin time is somehow saved. I'd rather asses such things myself. Further, sysadmin time is not so important. Thank you for voicing your opinion. just all move on? Further, your explanation is incomplete. As you said, the graph is about people knowing two DVCS systems. I wouldn't say I knew 2. Those 6 are incomplete. I highlighted this statistical analysis because those 6 contain the subset of 4 vocal users demanding that we /also/ support bzr. Now before you reply: we have a clear need for git to work (ranked 1st 50% of the time, etc). But if you say move on, how do you think a switch is made? Magic? Please don't be patronizing. I'm not an idiot. Anyway, I'd rather add John Carr to the sysadmin team. I plan to make a proposal to switch GNOME to a DVCS where Git works using Johns suggestion. Then other sysadmins[1] can suggest whatever proposal they want. These proposals can be investigated on merit and then a one can be chosen (chosen as in: go ahead and try if this would work, not go ahead blindly; everything must be tested before a cutover). John's idea is a good one but it patently loses on technical merit. As stated by John here, git will only be support in a degraded, bastardized form because he chose bzr as the repository format: http://blogs.gnome.org/johncarr/2008/12/11/dvcs-for-gnome/#comment-172 Are we really going to go back to the days of CVS where file moves aren't supported? It strikes me that this very vocal minority--John and Robert Carr, Karl Lattimer and Rob Taylor (whom are four of the six people I mentioned above)--are potentially delaying even longer what we've wanted for more than two years, now. It is from these same people that came the suggestion that git users were a rapid, vocal minority. Why are we letting them derail this process? Moving will not be easy, obviously. But doing it John's way will be, in my technical analysis, an order of magnitude more painful. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Not to be hostile, but please don't accuse me of holding anything up or being a vocal minority. I have never spoken out, posted, or blogged about any of the DVCS decisions. I think I said in the survey I would prefer bzr, however I didn't really care at the time (and much less since Discovering git-rebase--interactive) and if I indicated otherwise, that was not intentional. On Jan 4, 2009, at 10:40, Jason D. Clinton m...@jasonclinton.com wrote: On Sun, Jan 4, 2009 at 8:51 AM, Olav Vitters o...@bkor.dhs.org wrote: That isn't a contest. It is a survey. Please don't read more in to my email than I intended. There's no need to get defensive. http://www.gnome.org/~shaunm/survey/first-picks-permutations.png It seems to me that a lot of brain power, sysadmin time, and general I am a sysadmin and disagree with your notion that sysadmin time is somehow saved. I'd rather asses such things myself. Further, sysadmin time is not so important. Thank you for voicing your opinion. just all move on? Further, your explanation is incomplete. As you said, the graph is about people knowing two DVCS systems. I wouldn't say I knew 2. Those 6 are incomplete. I highlighted this statistical analysis because those 6 contain the subset of 4 vocal users demanding that we /also/ support bzr. Now before you reply: we have a clear need for git to work (ranked 1st 50% of the time, etc). But if you say move on, how do you think a switch is made? Magic? Please don't be patronizing. I'm not an idiot. Anyway, I'd rather add John Carr to the sysadmin team. I plan to make a proposal to switch GNOME to a DVCS where Git works using Johns suggestion. Then other sysadmins[1] can suggest whatever proposal they want. These proposals can be investigated on merit and then a one can be chosen (chosen as in: go ahead and try if this would work, not go ahead blindly; everything must be tested before a cutover). John's idea is a good one but it patently loses on technical merit. As stated by John here, git will only be support in a degraded, bastardized form because he chose bzr as the repository format: http://blogs.gnome.org/johncarr/2008/12/11/dvcs-for-gnome/#comment-172 Are we really going to go back to the days of CVS where file moves aren't supported? It strikes me that this very vocal minority--John and Robert Carr, Karl Lattimer and Rob Taylor (whom are four of the six people I mentioned above)--are potentially delaying even longer what we've wanted for more than two years, now. It is from these same people that came the suggestion that git users were a rapid, vocal minority. Why are we letting them derail this process? Moving will not be easy, obviously. But doing it John's way will be, in my technical analysis, an order of magnitude more painful. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Anyway, I'd rather add John Carr to the sysadmin team. I plan to make a proposal to switch GNOME to a DVCS where Git works using Johns suggestion. Then other sysadmins[1] can suggest whatever proposal they want. These proposals can be investigated on merit and then a one can be chosen (chosen as in: go ahead and try if this would work, not go ahead blindly; everything must be tested before a cutover). John's idea is a good one but it patently loses on technical merit. As stated by John here, git will only be support in a degraded, bastardized form because he chose bzr as the repository format: http://blogs.gnome.org/johncarr/2008/12/11/dvcs-for-gnome/#comment-172 Are we really going to go back to the days of CVS where file moves aren't supported? A git move operation is simply git rm git add. By that reckoning i'd either not be able to represent any deletes or any adds because of that god damn impaired Bzr file format! Wow, i'd be so eager to share that idea with the community ;) The big deal here is that git uses a heuristic to say Foo is now called Bar (selectively; its not done for speed in some cases). This is not stored anywhere in the file format, git redetermines it (if it wouldnt be too slow to do so). In Bazaar, its stored in the file format. This means merge doesnt have to consider ancestry to know if 2 files are related, it just knows they are. Solution? We simply have to run that heuristic ourselves so that Bazaar knows 2 files are related at import time. Git support is not degraded here, and Bazaar is no worse off than if you had imported a Git project into Bazaar for the first time (basically merge will work, but won't work *as* well in the rename case). I'm not a complete idiot - if it was going to be a degraded, bastardized form of Git I wouldn't waste my time on it. I suppose I might be an evil genius stalling for Bazaar DS9 to be written (sorry for the very bad joke that probably only i get...). It strikes me that this very vocal minority--John and Robert Carr, Karl Lattimer and Rob Taylor (whom are four of the six people I mentioned above)--are potentially delaying even longer what we've wanted for more than two years, now. It is from these same people that came the suggestion that git users were a rapid, vocal minority. Why are we letting them derail this process? This is not my 1st reply. The first one was fully of angry cow :(). Please dont single people out. I'm happy to have a hand wavy discussion with you 1st person, IRL especially so. Also, KL and RT are innocent. As bkor has stated, there are lots of Git users so any implementation will support you, and support you well. That is a requirement. So any talk of my idea is not Git vs Bazaar, its talk of one way we can move forward. So i dont consider it to be derailing. When mentioning my idea, lets stick to technical problems with it rather than trying to undermine anyone who has looked at it and thinks it is sound and doable. John ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 04, 2009 at 09:40:33AM -0600, Jason D. Clinton wrote: On Sun, Jan 4, 2009 at 8:51 AM, Olav Vitters o...@bkor.dhs.org wrote: That isn't a contest. It is a survey. Please don't read more in to my email than I intended. There's no need to get defensive. It is not defensive. I don't like changing a survey into 'winning' / contest. http://www.gnome.org/~shaunm/survey/first-picks-permutations.png It seems to me that a lot of brain power, sysadmin time, and general I am a sysadmin and disagree with your notion that sysadmin time is somehow saved. I'd rather asses such things myself. Further, sysadmin time is not so important. Thank you for voicing your opinion. just all move on? Further, your explanation is incomplete. As you said, the graph is about people knowing two DVCS systems. I wouldn't say I knew 2. Those 6 are incomplete. I highlighted this statistical analysis because those 6 contain the subset of 4 vocal users demanding that we /also/ support bzr. Yes, but then said 6. That is incomplete. Now before you reply: we have a clear need for git to work (ranked 1st 50% of the time, etc). But if you say move on, how do you think a switch is made? Magic? Please don't be patronizing. I'm not an idiot. You talk about moving on. I don't see anyone who'd do something like that. My reply is that nothing will happen unless someone does something real (not just another thread). Anyway, I'd rather add John Carr to the sysadmin team. I plan to make a proposal to switch GNOME to a DVCS where Git works using Johns suggestion. Then other sysadmins[1] can suggest whatever proposal they want. These proposals can be investigated on merit and then a one can be chosen (chosen as in: go ahead and try if this would work, not go ahead blindly; everything must be tested before a cutover). John's idea is a good one but it patently loses on technical merit. As stated by John here, git will only be support in a degraded, bastardized form because he chose bzr as the repository format: I read his comment not in the same way. Bzr supports more, Git less. However, I will less John answer... as that will be more concrete. http://blogs.gnome.org/johncarr/2008/12/11/dvcs-for-gnome/#comment-172 Are we really going to go back to the days of CVS where file moves aren't supported? Git doesn't do renames; instead applies heuristics. So this is applied. It strikes me that this very vocal minority--John and Robert Carr, Karl Lattimer and Rob Taylor (whom are four of the six people I mentioned above)--are potentially delaying even longer what we've wanted for more than two years, now. It is from these same people that came the suggestion that git users were a rapid, vocal minority. Why are we letting them derail this process? Again, you're limiting it to 6 people. It is not about the six. This is why I responded before. Instead, you use that number again. Even adding people's names, I don't find this useful. I am not going to talk about 'derailing'.. too emotional word. Moving will not be easy, obviously. But doing it John's way will be, in my technical analysis, an order of magnitude more painful. His way is a solution I expect to be implemented in 2009. To be honest, I really wonder if something else would happen that I'd qualify as a good switch. Yes, might be more difficult to implement. This is what can be discussed. (Along with other migration proposals.) -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 2:43 AM, Karl Lattimer k...@qdh.org.uk wrote: Elijah Newren did an initial analysis of the data. His analysis also includes the survey questions and answers. Find it at: http://blogs.gnome.org/newren/2009/01/03/gnome-dvcs-survey-results/ This is pretty decent analysis going on here :) I'd like to remind people of John Carr's recent blog post too, someone mentioned in the survey results actually. JC has been working on bzr with git protocol support, which would fulfil many of the requirements for having a GNOME DVCS. I'd like to point out that--of the 15 people who regularly use git and bzr--git still won. http://www.gnome.org/~shaunm/survey/first-picks-permutations.png It seems to me that a lot of brain power, sysadmin time, and general proliferation of Things To Learn for New People(tm) can be saved if the six people (1.04% of respondents) who ranked bzr above git in that graph can just bite the bullet and admit that git won. Can we please just all move on? My fear is that this effort to keep bzr on life support will cause bzr to show up as a requirement in distcheck for modules maintained by people who are still holding out. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 11:48 AM, John Carr john.c...@unrouted.co.uk wrote: I'm not a complete idiot - if it was going to be a degraded, bastardized form of Git I wouldn't waste my time on it. I suppose I might be an evil genius stalling for Bazaar DS9 to be written (sorry for the very bad joke that probably only i get...). I don't think you're an idiot. I think you're quite smart. Can you please tell us exactly what your words, This is a price that a maintainer pays for using Git and one reason why eventually they might decide to (and have the option to) switch to using Bazaar, mean and to which git features you are planning on this statement applying to encourage people to use bzr? Or do you mean that you taking that sentence back? Also, can you tell us if Canonical is directing you to work on this? ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sat, 2009-01-03 at 22:46 -0500, Behdad Esfahbod wrote: In December I ran a distributed version control system survey for GNOME. From the survey opening page: Thank you for taking the GNOME DVCS Survey. This survey is run on behalf of the GNOME Foundation board of directors, release team, and sysadmin team. The GNOME project is planning a possible move from SVN to a distributed version control system in 2009. The contenders for the system to use are bzr, git, and hg. The aim of the survey is to help us better understand familiarity and preferences of our active contributor base regarding the future version control system for GNOME. The survey results will be informational and will be sent to foundation-list and desktop-devel-list upon completion. GNOME contributors with an SVN account who had an SSH key installed on their account were invited to fill in the survey. A total of 1083 account holders were invited, and 579 filled in the survey. The survey results are now available to the public: http://www.gnome.org/~behdad/dvcs-survey/ Elijah Newren did an initial analysis of the data. His analysis also includes the survey questions and answers. Find it at: http://blogs.gnome.org/newren/2009/01/03/gnome-dvcs-survey-results/ If you analyze the results, please reply to this thread and also leave a comment on my blog post linking to your analysis: http://mces.blogspot.com/2009/01/gnome-dvcs-survey.html Just in case I am forced to switch to git in the future (being open minded here, although I prefer bazaar), does someone have any advice how to generate a nice GNU style ChangeLog (like what emacs produces) from git commit logs? I know that some projects like Cairo auto-generate ChangeLog already, but the default git changelog format is too detailed/ugly IMHO. -- Gustavo J. A. M. Carneiro g...@inescporto.pt gust...@users.sourceforge.net The universe is always one step beyond logic -- Frank Herbert ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On 1/4/09, Gustavo J. A. M. Carneiro g...@inescporto.pt wrote: Just in case I am forced to switch to git in the future (being open minded here, although I prefer bazaar), does someone have any advice how to generate a nice GNU style ChangeLog (like what emacs produces) from git commit logs? I know that some projects like Cairo auto-generate ChangeLog already, but the default git changelog format is too detailed/ugly IMHO. On a side note here, I recalled being against dropping ChangeLogs in projects in favour of commit messages. But now I love it and I realize that my main problem was that with SVN I *needed* the ChangeLog since that was the *only way* I could quickly read the project history, svn log took ages and was ugly. Now with giggle or git command line utils it's trivial to watch differences and read commit messages. On top of that we used to copy the exact same text from the ChangeLog entry into the commit message, which was pretty useless but necessary because of SVN limitations (now this is a vague comment, don't flame me). I'm not saying that ChangeLogs are useless, just commenting a little experience I had. Again, just a side note. greetings! ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
2009-01-04 klockan 15:10 skrev Jason D. Clinton: I'd like to point out that--of the 15 people who regularly use git and bzr--git still won. Two remarks. First remark: In the survey I answered that I do not really know much about git, and that I do not use it often. This has a reason, which I haven't seen anyone take into account: the few times I *did* try to use git it was an utterly frustrating experience, and I gave up pretty much immediately. (In contrast, the bzr experience has been a lot better: many good tutorials, better error messages and help from the command line tool, a friendly and active community, developers who actually *do* care about their users, and a clean, extensible design with great plugins floating around.) Second remark: a survey is never about ‘winning’ or ‘losing’. — Wouter signature.asc Description: Digital signature ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 17:48 +, John Carr wrote: As bkor has stated, there are lots of Git users so any implementation will support you, and support you well. That is a requirement. So any talk of my idea is not Git vs Bazaar, its talk of one way we can move forward. So i dont consider it to be derailing. When mentioning my idea, lets stick to technical problems with it rather than trying to undermine anyone who has looked at it and thinks it is sound and doable. Can you explain what would happen in the event that a future version of git switches to an repository format that isn't compatible with how you want to store data? David ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 3:01 PM, Zeeshan Ali (Khattak) zee...@gmail.com wrote: How about we set-up a task-force of volunteers who would want to help in the move, each volunteer promising at least 3 hours a week? 3 hours is a very small amount of time but I am hoping that we'll be able to gather at least 10 volunteers and together we can do it, even using our spare time. I can commit that much time as long as there's clear delegation of work by--preferably--the sysadmin team. I don't want to sit on a committee that does a lot of deciding and no actual doing. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 3:59 PM, David Zeuthen da...@fubar.dk wrote: On Sun, 2009-01-04 at 17:48 +, John Carr wrote: As bkor has stated, there are lots of Git users so any implementation will support you, and support you well. That is a requirement. So any talk of my idea is not Git vs Bazaar, its talk of one way we can move forward. So i dont consider it to be derailing. When mentioning my idea, lets stick to technical problems with it rather than trying to undermine anyone who has looked at it and thinks it is sound and doable. Can you explain what would happen in the event that a future version of git switches to an repository format that isn't compatible with how you want to store data? It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
2009/1/4 Gustavo J. A. M. Carneiro g...@inescporto.pt: Just in case I am forced to switch to git in the future (being open minded here, although I prefer bazaar), does someone have any advice how to generate a nice GNU style ChangeLog (like what emacs produces) from git commit logs? I know that some projects like Cairo auto-generate ChangeLog already, but the default git changelog format is too detailed/ugly IMHO. Actually, now that I looked, gnulib also has what you request[1]. Rui [1] http://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=blob;f=build-aux/gitlog-to-changelog;h=0a94b9e8ba1e26ad24a535ec57f39797aac78c12;hb=HEAD ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Hi! On Sun, Jan 4, 2009 at 6:06 PM, Olav Vitters o...@bkor.dhs.org wrote: On Sun, Jan 04, 2009 at 09:40:33AM -0600, Jason D. Clinton wrote: On Sun, Jan 4, 2009 at 8:51 AM, Olav Vitters o...@bkor.dhs.org wrote: Moving will not be easy, obviously. But doing it John's way will be, in my technical analysis, an order of magnitude more painful. His way is a solution I expect to be implemented in 2009. No matter how good that sounds, it's still not a solution, it's a workaround to the problem that we don't have (human) resources to do a move to git. To be honest, I really wonder if something else would happen that I'd qualify as a good switch. How about we set-up a task-force of volunteers who would want to help in the move, each volunteer promising at least 3 hours a week? 3 hours is a very small amount of time but I am hoping that we'll be able to gather at least 10 volunteers and together we can do it, even using our spare time. In any case, after looking at the results of the survey we should only look at hybrid/dual proposal like John's when we don't find any way of moving to git in a reasonable amount of time ( 6 months). -- Regards, Zeeshan Ali (Khattak) FSF member#5124 ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 23:01 +0200, Zeeshan Ali (Khattak) wrote: How about we set-up a task-force of volunteers who would want to help in the move, each volunteer promising at least 3 hours a week? 3 hours is a very small amount of time but I am hoping that we'll be able to gather at least 10 volunteers and together we can do it, even using our spare time. Would it be worth investigating whether it's worth having the Foundation pay someone to help with this migration (planning, executing, maybe even hosting etc.)? I mean, the eco-system around git is huge (github and others comes to mind) and growing... I'm pretty sure there's plenty git experts around. David ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 3:27 PM, Olav Vitters o...@bkor.dhs.org wrote: I can commit that much time as long as there's clear delegation of work by--preferably--the sysadmin team. I don't want to sit on a committee that does a lot of deciding and no actual doing. What do you mean with delegation? Which do you mean: (yes, exaggerating) - Hey, do the switch, hopefully it'll work out in the end? - Run this command, then this one, then that More of a, Given this requirement, you find a solution to this specific problem. Report back in a week and ask for help if you get stuck, where solution may involve writing code in the form of post-commit hooks. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
David Zeuthen wrote: On Sun, 2009-01-04 at 17:48 +, John Carr wrote: As bkor has stated, there are lots of Git users so any implementation will support you, and support you well. That is a requirement. So any talk of my idea is not Git vs Bazaar, its talk of one way we can move forward. So i dont consider it to be derailing. When mentioning my idea, lets stick to technical problems with it rather than trying to undermine anyone who has looked at it and thinks it is sound and doable. Can you explain what would happen in the event that a future version of git switches to an repository format that isn't compatible with how you want to store data? Probably just like bzr already went through several repository formats and allowed easy upgrades (just like Subversion repository format changed and it didn't cause any problem for users). I don't think there is a problem here. And also, data would be available in native git format on lots of computers, and could always be pushed to a vanilla git server. Frederic [Disclaimer: this is just my understanding of the proposal, I may be wrong and corrected by anybody.] ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Il giorno dom, 04/01/2009 alle 16.11 -0500, Matthias Clasen ha scritto: It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. BTW, once switched to DVCS, how much disk space I should have in order to perform a full GNOME Desktop build with jhbuild? A WebKit build from git needs ~740MB :-/ People using JHBuild to develop one project against latest code or simply testing the whole desktop don't need the full history for all GNOME Desktop modules bzr allows lightweight checkouts [1]. What about git? [1] http://doc.bazaar-vcs.org/bzr.dev/en/user-guide/index.html#getting-a-lightweight-checkout ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 3:47 PM, Luca Ferretti elle@libero.it wrote: bzr allows lightweight checkouts [1]. What about git? Yes, it does. This is not an issue. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 10:47 PM, Luca Ferretti elle@libero.it wrote: People using JHBuild to develop one project against latest code or simply testing the whole desktop don't need the full history for all GNOME Desktop modules bzr allows lightweight checkouts [1]. What about git? git-clone has a --depth option [0] to perform shallow clones up to a certain number of revisions. Marko [0] http://www.kernel.org/pub/software/scm/git/docs/git-clone.html ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 11:47 PM, Luca Ferretti elle@libero.it wrote: Il giorno dom, 04/01/2009 alle 16.11 -0500, Matthias Clasen ha scritto: It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. BTW, once switched to DVCS, how much disk space I should have in order to perform a full GNOME Desktop build with jhbuild? A WebKit build from git needs ~740MB :-/ How much does it consume if it's a svn checkout? I heard (don't know if it's true or not) git repo usually takes less diskspace then svn checkout. This page seems to support this claim: http://git.or.cz/gitwiki/GitSvnComparsion An SVN working directory always contains two copies of each file: one for the user to actually work with and another hidden in .svn/ to aid operations such as status, diff and commit. In contrast a Git working directory requires only one small index file that stores about 100 bytes of data per tracked file. On projects with a large number of files this can be a substantial difference in the disk space required per working copy. -- Regards, Zeeshan Ali (Khattak) FSF member#5124 ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 22:47 +0100, Frederic Peters wrote: Probably just like bzr already went through several repository formats and allowed easy upgrades (just like Subversion repository format changed and it didn't cause any problem for users). I don't think there is a problem here. I don't find this answer compelling. At all. It also doesn't answer the question. It's not unlikely that a future git repo format is fundamentally incompatible with current or future bzr repo formats. And also, data would be available in native git format on lots of computers, and could always be pushed to a vanilla git server. Someone really got to explain exactly why support for multiple repository formats is desirable. First, it only makes it much harder for users to grasp; we're going to end up with some projects have l.g.o pages / README files / mailing list messages saying use bzr to check out this branch and others saying the same for git. That's *not* desirable; it makes it so much harder for new contributors. Second, it also makes it harder to set up things like jhbuild; either you end up pulling from both git and bzr (from the same underlying repo) or you end up mentally having to translate branch names etc. from one system to another. This is error prone. Third, I could go on with examples, just consider the set of webtools (cgit, annotation, source code searching etc.) we end up with on dvcs.gnome.org; some would be built against bzr, others against git. You get inconsistent branch names, you end up overloading contributors with different concepts and so forth. Finally: We're talking about people's data here. The first rule of holding peoples data is that you don't screw around with it just because. Data integrity matters. Keeping things simple and staying with a *single* kind of hammer (instead of a weird homegrown mutant hammer) helps here. Otherwise we end up with data loss. Frankly, I'm concerned that some people are even considering using such homegrown kludges for holding our GNOME source code. David ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Il giorno dom, 04/01/2009 alle 23.58 +0200, Zeeshan Ali (Khattak) ha scritto: On Sun, Jan 4, 2009 at 11:47 PM, Luca Ferretti elle@libero.it wrote: Il giorno dom, 04/01/2009 alle 16.11 -0500, Matthias Clasen ha scritto: It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. BTW, once switched to DVCS, how much disk space I should have in order to perform a full GNOME Desktop build with jhbuild? A WebKit build from git needs ~740MB :-/ How much does it consume if it's a svn checkout? Note that 740MB is the size of source stuff + build stuff (moreover, but I'm not sure, WebKit duplicates some source file at build time). The size of fresh checkout is 575MB (see attached file for `du -ch` details) 8.0K./.git/refs/heads 4.0K./.git/refs/tags 16K ./.git/refs/remotes/origin 20K ./.git/refs/remotes 40K ./.git/refs 4.0K./.git/branches 8.0K./.git/info 48K ./.git/hooks 456M./.git/objects/pack 4.0K./.git/objects/info 36K ./.git/objects/ef 20K ./.git/objects/0f 16K ./.git/objects/28 60K ./.git/objects/b9 60K ./.git/objects/ad 1000K ./.git/objects/a4 16K ./.git/objects/dc 40K ./.git/objects/9e 16K ./.git/objects/c5 36K ./.git/objects/90 556K./.git/objects/9d 228K./.git/objects/fb 40K ./.git/objects/bd 392K./.git/objects/14 68K ./.git/objects/7d 548K./.git/objects/62 348K./.git/objects/d5 40K ./.git/objects/d6 28K ./.git/objects/51 368K./.git/objects/df 444K./.git/objects/ae 24K ./.git/objects/77 528K./.git/objects/fd 40K ./.git/objects/a5 28K ./.git/objects/63 32K ./.git/objects/eb 268K./.git/objects/27 68K ./.git/objects/6f 552K./.git/objects/a2 24K ./.git/objects/da 360K./.git/objects/a8 308K./.git/objects/bb 72K ./.git/objects/e9 768K./.git/objects/53 20K ./.git/objects/91 20K ./.git/objects/f0 56K ./.git/objects/b7 52K ./.git/objects/13 136K./.git/objects/61 216K./.git/objects/45 364K./.git/objects/d8 332K./.git/objects/2e 132K./.git/objects/c2 228K./.git/objects/00 600K./.git/objects/f2 508K./.git/objects/db 408K./.git/objects/73 1.2M./.git/objects/f4 72K ./.git/objects/67 44K ./.git/objects/2d 224K./.git/objects/83 212K./.git/objects/8e 40K ./.git/objects/b0 220K./.git/objects/65 380K./.git/objects/9a 68K ./.git/objects/81 180K./.git/objects/22 44K ./.git/objects/d2 40K ./.git/objects/2a 88K ./.git/objects/a6 32K ./.git/objects/42 48K ./.git/objects/3b 112K./.git/objects/48 24K ./.git/objects/7b 56K ./.git/objects/76 36K ./.git/objects/9b 508K./.git/objects/6e 152K./.git/objects/fe 268K./.git/objects/c6 216K./.git/objects/8b 48K ./.git/objects/3f 360K./.git/objects/5b 76K ./.git/objects/b5 20K ./.git/objects/37 768K./.git/objects/6b 28K ./.git/objects/f6 352K./.git/objects/39 564K./.git/objects/17 68K ./.git/objects/5e 64K ./.git/objects/1f 360K./.git/objects/ac 36K ./.git/objects/24 32K ./.git/objects/ee 24K ./.git/objects/29 24K ./.git/objects/78 24K ./.git/objects/70 380K./.git/objects/fc 104K./.git/objects/72 64K ./.git/objects/8d 348K./.git/objects/6a 24K ./.git/objects/e6 28K ./.git/objects/bc 24K ./.git/objects/94 100K./.git/objects/59 420K./.git/objects/4b 24K ./.git/objects/ce 40K ./.git/objects/f5 24K ./.git/objects/1b 268K./.git/objects/4a 24K ./.git/objects/d4 36K ./.git/objects/03 220K./.git/objects/e5 28K ./.git/objects/33 40K ./.git/objects/aa 32K ./.git/objects/e1 76K ./.git/objects/f9 488K./.git/objects/ba 24K ./.git/objects/dd 48K ./.git/objects/c0 368K./.git/objects/ab 364K./.git/objects/26 352K./.git/objects/0d 32K ./.git/objects/01 332K./.git/objects/a1 44K ./.git/objects/97 28K ./.git/objects/38 52K ./.git/objects/cb 36K ./.git/objects/6c 92K ./.git/objects/74 24K ./.git/objects/ca 20K ./.git/objects/5a 604K./.git/objects/9c 360K./.git/objects/54 220K./.git/objects/1c 228K./.git/objects/d3 28K ./.git/objects/68 420K./.git/objects/19 32K ./.git/objects/43 264K./.git/objects/c9 316K./.git/objects/89 20K ./.git/objects/95 380K./.git/objects/ec 24K ./.git/objects/4d 332K./.git/objects/5c 280K./.git/objects/d7 112K./.git/objects/58 88K ./.git/objects/e7 24K ./.git/objects/3e 344K./.git/objects/69 28K ./.git/objects/40 16K ./.git/objects/cd 28K ./.git/objects/7c 28K ./.git/objects/cf 24K ./.git/objects/09 28K ./.git/objects/c1 28K ./.git/objects/cc 388K./.git/objects/e4 20K ./.git/objects/e3 56K ./.git/objects/12 36K ./.git/objects/99 336K
Re: GNOME DVCS Survey Results
Jason D. Clinton wrote: On Sun, Jan 4, 2009 at 3:47 PM, Luca Ferretti elle@libero.it wrote: bzr allows lightweight checkouts [1]. What about git? Yes, it does. This is not an issue. I think non-git users already knows that git can do everything™, but they would learn about git ways faster if you pointed to explanations (just like Marko did). Frederic ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On sön, 2009-01-04 at 23:58 +0200, Zeeshan Ali (Khattak) wrote: On Sun, Jan 4, 2009 at 11:47 PM, Luca Ferretti elle@libero.it wrote: Il giorno dom, 04/01/2009 alle 16.11 -0500, Matthias Clasen ha scritto: It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. BTW, once switched to DVCS, how much disk space I should have in order to perform a full GNOME Desktop build with jhbuild? A WebKit build from git needs ~740MB :-/ How much does it consume if it's a svn checkout? I heard (don't know if it's true or not) git repo usually takes less diskspace then svn checkout. This page seems to support this claim: A complete git repo is usually smaller than a complete SVN one (according to common knowlege - as in, I didn't run any benchmarks), but one commonly only checks out the /trunk subdirectory in subversion, while git usually checks out the whole project history, including all branches - it could be a substantial amount of data you don't check out with SVN. ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Matthias Clasen schrieb: On Sun, Jan 4, 2009 at 3:59 PM, David Zeuthen da...@fubar.dk wrote: On Sun, 2009-01-04 at 17:48 +, John Carr wrote: As bkor has stated, there are lots of Git users so any implementation will support you, and support you well. That is a requirement. So any talk of my idea is not Git vs Bazaar, its talk of one way we can move forward. So i dont consider it to be derailing. When mentioning my idea, lets stick to technical problems with it rather than trying to undermine anyone who has looked at it and thinks it is sound and doable. Can you explain what would happen in the event that a future version of git switches to an repository format that isn't compatible with how you want to store data? It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. I totally agree. Sooner or later it will become a nightmare to maintain. - -- Greetings, Sebastian Pölsterl -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAklhNewACgkQ1ygZeJ3lLIeGIACglzAktDqy1eQ6VBsOsak41zSk d6cAnAh9IK1acbtnyufeezRL+TQ9Dgvp =N+VG -END PGP SIGNATURE- ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 11:05 PM, David Zeuthen da...@fubar.dk wrote: On Sun, 2009-01-04 at 22:47 +0100, Frederic Peters wrote: Probably just like bzr already went through several repository formats and allowed easy upgrades (just like Subversion repository format changed and it didn't cause any problem for users). I don't think there is a problem here. I don't find this answer compelling. At all. It also doesn't answer the question. It's not unlikely that a future git repo format is fundamentally incompatible with current or future bzr repo formats. And also, data would be available in native git format on lots of computers, and could always be pushed to a vanilla git server. Someone really got to explain exactly why support for multiple repository formats is desirable. To put it straight: the git repository format is not as awesome as people want to believe. First, it only makes it much harder for users to grasp; we're going to end up with some projects have l.g.o pages / README files / mailing list messages saying use bzr to check out this branch and others saying the same for git. That's *not* desirable; it makes it so much harder for new contributors. That's not what John's proposal is about ! John wants to use the bzr format as a repository format, and add a git-serve plugin to bzr to be able to talk to the git clients. In other words, you will be able to access the same data using either bzr, git or hg. Second, it also makes it harder to set up things like jhbuild; either you end up pulling from both git and bzr (from the same underlying repo) or you end up mentally having to translate branch names etc. from one system to another. This is error prone. Third, I could go on with examples, just consider the set of webtools (cgit, annotation, source code searching etc.) we end up with on dvcs.gnome.org; some would be built against bzr, others against git. You get inconsistent branch names, you end up overloading contributors with different concepts and so forth. Finally: We're talking about people's data here. The first rule of holding peoples data is that you don't screw around with it just because. Data integrity matters. Keeping things simple and staying with a *single* kind of hammer (instead of a weird homegrown mutant hammer) helps here. Otherwise we end up with data loss. Frankly, I'm concerned that some people are even considering using such homegrown kludges for holding our GNOME source code. Comparing the size of the Bazaar unit tests with those of Git, I would certainly choose Bazaar for storing my data. Cheers, -- Ali ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
David Zeuthen wrote: I don't find this answer compelling. At all. It also doesn't answer the question. It's not unlikely that a future git repo format is fundamentally incompatible with current or future bzr repo formats. Just like I noted it was just an understanding of John's proposal; because I feel it has value and thought it was dismissed a bit too fast. But I won't make suppositions again on the technicalities. Frederic ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 23:16 +0100, Robin Sonefors wrote: How much does it consume if it's a svn checkout? I heard (don't know if it's true or not) git repo usually takes less diskspace then svn checkout. This page seems to support this claim: A complete git repo is usually smaller than a complete SVN one (according to common knowlege - as in, I didn't run any benchmarks), but one commonly only checks out the /trunk subdirectory in subversion, while git usually checks out the whole project history, including all branches - it could be a substantial amount of data you don't check out with SVN. I think Zeeshan is talking about working copies, not repositories. An svn working copy stores two complete copies of each file in the repo; one is the one that you see and edit, and one is in .svn/text-base ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 11:16 PM, Robin Sonefors ozam...@flukkost.nu wrote: On sön, 2009-01-04 at 23:58 +0200, Zeeshan Ali (Khattak) wrote: On Sun, Jan 4, 2009 at 11:47 PM, Luca Ferretti elle@libero.it wrote: Il giorno dom, 04/01/2009 alle 16.11 -0500, Matthias Clasen ha scritto: It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. BTW, once switched to DVCS, how much disk space I should have in order to perform a full GNOME Desktop build with jhbuild? A WebKit build from git needs ~740MB :-/ How much does it consume if it's a svn checkout? I heard (don't know if it's true or not) git repo usually takes less diskspace then svn checkout. This page seems to support this claim: A complete git repo is usually smaller than a complete SVN one (according to common knowlege - as in, I didn't run any benchmarks), but one commonly only checks out the /trunk subdirectory in subversion, while git usually checks out the whole project history, including all branches - it could be a substantial amount of data you don't check out with SVN. Well, Actually the quotes from the GitSvnComparsion page are very misleading, it is true that a git working directory needs less space than an svn working directory, it is also true that a git repository is smaller than an svn repository. The main difference is that with git, you *clone* a repository, and then create a working directory out of it, so you need sizeof(repo) + sizeof(git-working-directory) on your hard disk, while with svn, all what you need is sizeof(svn-working-directory). ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 23:20 +0100, Ali Sabil wrote: First, it only makes it much harder for users to grasp; we're going to end up with some projects have l.g.o pages / README files / mailing list messages saying use bzr to check out this branch and others saying the same for git. That's *not* desirable; it makes it so much harder for new contributors. That's not what John's proposal is about ! John wants to use the bzr format as a repository format, and add a git-serve plugin to bzr to be able to talk to the git clients. In other words, you will be able to access the same data using either bzr, git or hg. Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. Finally: We're talking about people's data here. The first rule of holding peoples data is that you don't screw around with it just because. Data integrity matters. Keeping things simple and staying with a *single* kind of hammer (instead of a weird homegrown mutant hammer) helps here. Otherwise we end up with data loss. Frankly, I'm concerned that some people are even considering using such homegrown kludges for holding our GNOME source code. Comparing the size of the Bazaar unit tests with those of Git, I would certainly choose Bazaar for storing my data. I wasn't commenting on bzr vs git storage format; I'm sure either is fine. I was commenting on the fact that someone proposes to inject something like git-serve in the middle; that's what I think is a kludge. David ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 23:33 +0100, Olav Vitters wrote: On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. Then what happens when a new version of git with a new feature, incompatible with the git-serve kludge, is released? Then we're screwed, right? And who gets to pay? We do. We're stuck with an old version of git. Us. The very same people who very clearly said git, not bzr. Is it *really* so hard to understand that this whole git-serve is a terrible idea? David ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 04, 2009 at 11:37:05PM +0100, Patryk Zawadzki wrote: On Sun, Jan 4, 2009 at 11:33 PM, Olav Vitters o...@bkor.dhs.org wrote: On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. The potential problem I see is all of the remote branches will use different DVCS that do not support git + hg + bzr. So eventually all Again: No Hg. of us will be forced to use all three tools in order to merge changes from remote branches (unless we expect *all* people to provide *all* changes as patches in which case I don't see the real gain of switching to a distributed tool). Interesting point. I actually saw it as a benefit (store locally using whatever you like). On GNOME server (personal stuff), doesn't matter. Anyway, if you're going against the maintainer who wants to merge, too bad for you IMO. -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 4, 2009 at 11:33 PM, Olav Vitters o...@bkor.dhs.org wrote: On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. Sorry for not being clear in my explanations. Basically, as Olav pointed out, it is about having Bazaar on the server, with a git-serve plugin allowing it to fulfill the git client requests as well as the bzr client requests. The following scenarios will be possible: (bzr repo) - (git serve plugin) - network --- (git client) (bzr repo) - (bzr serve) - network --- (bzr client) both bzr and git will operate fully, nothing will be partially supported, since the bazaar repository format is a superset of the git repo format (ie. it stores more metadata). I talked about hg, just to highlight that the solution is quite future proof, because you can certainly apply the same solution to allow hg clients to access the repository. cheers, -- Ali ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Hi! It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. I totally agree here! This is simply a problem of QA. If someone writes a system that can serve all possible (D)VCS clients that's fine but this won't happen tommorow and I will need a huge amount of time to be finished and tested. And in addition it's unlikely that such a system will support more than a common subset of the features of the underlying DVCS system. First, be honest, we need to decide which system to use. I have no preference here, the survey says that most current users prefer git. So it sounds reasonable to go that way if it doesn't has to much problematic impact on the infrastructure side. Second, a VCS system is something that just has to work. I doubt many people really care a lot about what system they use as long as it does not cause any problems for them. People familiar to git will easily learn bzr and the bzr-people will learn git. It's not a good idea to make this decision too important and to do flame-wars. Probably all major DVCS fit our needs and it is more a matter of taste. The survey was a good point to find out the taste of the GNOME developers = we should accept that. Regards Johannes signature.asc Description: Dies ist ein digital signierter Nachrichtenteil ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 23:58 +0100, Olav Vitters wrote: On Sun, Jan 04, 2009 at 05:40:18PM -0500, David Zeuthen wrote: Is it *really* so hard to understand that this whole git-serve is a terrible idea? You expect me to reply to this??!? I expected you to reply to the other three mails where I asked the same thing as I did in the mail you replied to. Oh, you chose not to quote that; here it is again: Then what happens when a new version of git with a new feature, incompatible with the git-serve kludge, is released? Then we're screwed, right? And who gets to pay? We do. We're stuck with an old version of git. Us. The very same people who very clearly said git, not bzr. But, alas, you didn't reply to this. You instead hand-waved about something else. I don't think I breached any code of conduct, written or otherwise, by displaying my frustration about how you are evading my question. David ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Mon, Jan 05, 2009 at 12:04:30AM +0100, Johannes Schmid wrote: Hi! It seems pretty clear to me that any 'homegrown' system like this is not suitable as a longterm, stable solution for a project the size of gnome. I totally agree here! This is simply a problem of QA. If someone writes a system that can serve all possible (D)VCS clients that's fine but this That is not what is proposed. won't happen tommorow and I will need a huge amount of time to be finished and tested. And in addition it's unlikely that such a system Proposed solution doesn't take a long time to finish. will support more than a common subset of the features of the underlying DVCS system. [..] Second, a VCS system is something that just has to work. I doubt many people really care a lot about what system they use as long as it does No need to guess, we can look at the survey. -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, 2009-01-04 at 17:40 -0500, David Zeuthen wrote: On Sun, 2009-01-04 at 23:33 +0100, Olav Vitters wrote: On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. Then what happens when a new version of git with a new feature, incompatible with the git-serve kludge, is released? Then we're screwed, right? And who gets to pay? We do. We're stuck with an old version of git. Us. The very same people who very clearly said git, not bzr. Is it *really* so hard to understand that this whole git-serve is a terrible idea? more importantly: is it *really* so hard to understand that if you want a bzr storage for git you should probably propose it upstream instead of writing something ad hoc for GNOME alone? if the idea has any merit[0] then it should be pushed upstream -- even as an optional repository format. ciao, Emmanuele. [0] I'm reasonably sure it has some. not as the one proposed to avoid pissing off somebody somewhere because we want to be inclusive -- no, lemme rephrase that: we are *fucking afraid of committment*. seriously: an abstraction over DVCS? what have we become? are we *ever* going make *any* decision about *anything*? this is actually a larger issue with the GNOME community: we are being afraid. -- Emmanuele Bassi, W: http://www.emmanuelebassi.net B: http://log.emmanuelebassi.net ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Sun, Jan 04, 2009 at 06:05:30PM -0500, David Zeuthen wrote: On Sun, 2009-01-04 at 23:58 +0100, Olav Vitters wrote: On Sun, Jan 04, 2009 at 05:40:18PM -0500, David Zeuthen wrote: Is it *really* so hard to understand that this whole git-serve is a terrible idea? You expect me to reply to this??!? I expected you to reply to the other three mails where I asked the same thing as I did in the mail you replied to. Oh, you chose not to quote that; here it is again: I chose not to quote that yes, as this is getting too personal for me. However, I only get more replies back which I consider of terrible quality. Then what happens when a new version of git with a new feature, incompatible with the git-serve kludge, is released? Then we're screwed, right? And who gets to pay? We do. We're stuck with an old version of git. Us. The very same people who very clearly said git, not bzr. But, alas, you didn't reply to this. You instead hand-waved about something else. I don't think I breached any code of conduct, written or otherwise, by displaying my frustration about how you are evading my question. I am not evading. Stop trying to make this personal. I don't care about CoC, I don't like you're talking to me. Anyway, I've already asked John to respond to your point as he is doing the work. I did that before replying to you. This as I thought he would give the best answer. My answer: well, AFAIK, the communication stuff is very generic, so breakage is unlikely. Further, that is why John becomes a sysadmin. Feel free to rewrite my answer as needed. -- Regards, Olav ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Le lundi 05 janvier 2009 à 00:04 +0100, Johannes Schmid a écrit : I totally agree here! This is simply a problem of QA. If someone writes a system that can serve all possible (D)VCS clients that's fine but this won't happen tommorow No, it already happened and it is called Subversion. This is the only repository format that all major DVCS clients can talk to. -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Mon, 2009-01-05 at 00:18 +0100, Olav Vitters wrote: I am not evading. Stop trying to make this personal. I don't care about CoC, I don't like you're talking to me. Please. Stop trying to make this look like it's personal and like I'm assaulting you. Because I didn't. And I resent the accusation. David ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Le dimanche 04 janvier 2009 à 15:40 -0500, Diego Escalante Urrelo a écrit : On a side note here, I recalled being against dropping ChangeLogs in projects in favour of commit messages. But now I love it and I realize that my main problem was that with SVN I *needed* the ChangeLog since that was the *only way* I could quickly read the project history, svn log took ages and was ugly. Now with giggle or git command line utils it's trivial to watch differences and read commit messages. Only if you have a checkout. But for those who want to look at the history with the web interface, this is a nightmare with both viewvc and gitweb. And for us who regularly access all modules without wanting to checkout all of them, this is a clear regression from what we had with a simple ChangeLog. -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
2009/1/5 Josselin Mouette j...@debian.org: Le dimanche 04 janvier 2009 à 15:40 -0500, Diego Escalante Urrelo a écrit : On a side note here, I recalled being against dropping ChangeLogs in projects in favour of commit messages. But now I love it and I realize that my main problem was that with SVN I *needed* the ChangeLog since that was the *only way* I could quickly read the project history, svn log took ages and was ugly. Now with giggle or git command line utils it's trivial to watch differences and read commit messages. Only if you have a checkout. But for those who want to look at the history with the web interface, this is a nightmare with both viewvc and gitweb. And for us who regularly access all modules without wanting to checkout all of them, this is a clear regression from what we had with a simple ChangeLog. I think that you do get the same experience with gitweb as with giggle. Eg http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=3bfacef412b4bc993a8992217e50f1245f2fd3a6 and http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=3bfacef412b4bc993a8992217e50f1245f2fd3a6 or perhaps this is not what you had in mind? Marko ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On Mon, 2009-01-05 at 00:41 +0100, Josselin Mouette wrote: Le dimanche 04 janvier 2009 à 15:40 -0500, Diego Escalante Urrelo a écrit : On a side note here, I recalled being against dropping ChangeLogs in projects in favour of commit messages. But now I love it and I realize that my main problem was that with SVN I *needed* the ChangeLog since that was the *only way* I could quickly read the project history, svn log took ages and was ugly. Now with giggle or git command line utils it's trivial to watch differences and read commit messages. Only if you have a checkout. But for those who want to look at the history with the web interface, this is a nightmare with both viewvc and gitweb. And for us who regularly access all modules without wanting to checkout all of them, this is a clear regression from what we had with a simple ChangeLog. I'd agree if you could not properly see a project's commit history on a web interface, like with viewvc for svn, but since you can: http://git.clutter-project.org/cgit.cgi?url=clutter/log/ your point is moot. Clutter removed the ChangeLog for good with the switch to a proper (i.e. not svn) revision control system. we generate the ChangeLog for the tarballs from the commit logs, obviously, since you cannot access the revision control history in that case. ciao, Emmanuele / not pimping git - just cgit. -- Emmanuele Bassi, W: http://www.emmanuelebassi.net B: http://log.emmanuelebassi.net ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
Le dimanche 04 janvier 2009 à 23:44 +, Emmanuele Bassi a écrit : I'd agree if you could not properly see a project's commit history on a web interface, like with viewvc for svn, but since you can: http://git.clutter-project.org/cgit.cgi?url=clutter/log/ your point is moot. Sorry, but how is this interface better than that of viewvc? Clutter removed the ChangeLog for good with the switch to a proper (i.e. not svn) revision control system. I also fail to see how this has anything to do with the VCS you use (apart from bashing subversion, which looks quite trendy these days). While svn log is slow, it does the same job, so this is really an interface and usability point, not something related to the VCS engine itself. -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list
Re: GNOME DVCS Survey Results
On 01/04/2009 05:10 PM, Emmanuele Bassi wrote: On Sun, 2009-01-04 at 17:40 -0500, David Zeuthen wrote: On Sun, 2009-01-04 at 23:33 +0100, Olav Vitters wrote: On Sun, Jan 04, 2009 at 05:29:02PM -0500, David Zeuthen wrote: Uh, but that's exactly how I understood the proposal and I believe that the points I made (that you didn't respond to) still stands: That it's crazy to officially want to support git, bzr and hg *at* the same time *from* the same repo. It's just asking for trouble. That isn't true. It is Bzr on server, with Git support. Nothing about Hg, nothing about doing partly Git, partly Bzr. Then what happens when a new version of git with a new feature, incompatible with the git-serve kludge, is released? Then we're screwed, right? And who gets to pay? We do. We're stuck with an old version of git. Us. The very same people who very clearly said git, not bzr. Is it *really* so hard to understand that this whole git-serve is a terrible idea? more importantly: is it *really* so hard to understand that if you want a bzr storage for git you should probably propose it upstream instead of writing something ad hoc for GNOME alone? if the idea has any merit[0] then it should be pushed upstream -- even as an optional repository format. ciao, Emmanuele. [0] I'm reasonably sure it has some. not as the one proposed to avoid pissing off somebody somewhere because we want to be inclusive -- no, lemme rephrase that: we are *fucking afraid of committment*. seriously: an abstraction over DVCS? what have we become? are we *ever* going make *any* decision about *anything*? this is actually a larger issue with the GNOME community: we are being afraid. Exactly. The idea that our gnome vcs infrastructure would be run by some homegrown abstraction layer is rather scary (regardless of how well it's written and how talented the developers are). The fact that some people are acting like it's a reasonable solution for a project the size of GNOME is even scarier. jonner ___ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list