On Wed, 06 Nov 2013 16:55:04 +0100 Jonathan Aquilina <jaquil...@eagleeyet.net> said:
> When i say lose i mean more along the lines of ending up being missed or > overlooked. https://phab.enlightenment.org/w/arcanist/ linked from: http://www.enlightenment.org/p.php?p=contribute and when they have been submitted they end up on: https://phab.enlightenment.org/differential/query/c.8PlsTrzNjQ if you look at them you will notice there are things like inline commenting on the patch (just like gerrit) etc. etc. just like gerrit - if people don't use the mechanisms we have in place, then patches may get "lost" on the mailing list. even here with phab patches may go unreviewed for a long time. they end up eternally on the list. unless someone gets assigned to review them explicitly then they are often not noticed. i cleaned out some old ones recently and need to go through more of them soon for release (those that are bug fixes. those that are new features will just wait until efl 1.8 is out). running EVERYTHING through gerrit will just waste our time. realistically what will happen is it will force everyone to ask someone else to review and 99.9% of that will be the rubber-stamp variety. that is what i have seen gerrit devolve into. unless you have a seriously large amount of manpower and time and are willing to slow down your dev process a lot, it becomes nothing more than a rubber-stamping exercise, which in and of itself just adds delay and STILL wastes peoples time in doing the bureaucratic stamping process. there is no point having a pre-review if that is the case (and only makes things worse as if the rubber-stamp comes a little too late, there may be a conflict and the commit has to go back to be re-edited just to get a stamp when it would have been fine if committed directly). if you have a commit - eg on a branch of your own, BEFORE you merge it and WANT people to review before your merge (voluntariily because you're not sure and want more eyeballs) then phab has all of this already. https://phab.enlightenment.org/diffusion/ just go to any commit and click. you can comment on the commit inline like gerrit. just ask people to do this and based on the feedback push in more commits and when ready - merge. that is if you have commit access already. commit access is what you get when "your submissions are more trouble to review before merge than catching them ad-hoc afterwards". ie they are almost all good quality and not bad/malicious etc. and where the value in review before commit is minimal vs the effort to do so. so it means "we trust you". until you have commit access changes can be submitted via phab's review process as above. oh and one other thing. gerrit is a NEW git server written in java. no one here was fond of the idea of having that running at all. not in the slightest. :) > On Wednesday 06 November 2013 15:41:07 Tom Hacohen wrote: > > On 06/11/13 15:35, Jonathan Aquilina wrote: > > > Dont you run the risk of patches falling through the cracks in terms of > > > patches recieved on the mailing list? > > > > We should at some point use patchwork or whatever. But no, there's no > > risk in that, because the official method of sending patches is > > phabricator. Which doesn't lose patches. > > > > -- > > Tom. > > > > > > ---------------------------------------------------------------------------- > > -- November Webinars for C, C++, Fortran Developers > > Accelerate application performance with scalable programming models. Explore > > techniques for threading, error checking, porting, and tuning. Get the most > > from the latest Intel processors and coprocessors. See abstracts and > > register > > http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk > > _______________________________________________ > > enlightenment-devel mailing list > > enlightenment-devel@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > ------------------------------------------------------------------------------ > November Webinars for C, C++, Fortran Developers > Accelerate application performance with scalable programming models. Explore > techniques for threading, error checking, porting, and tuning. Get the most > from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk > _______________________________________________ > enlightenment-devel mailing list > enlightenment-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- The Rasterman (Carsten Haitzler) ras...@rasterman.com ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
