Ilias Lazaridis wrote: > Christopher Lenz wrote: > >> ... >> The Genshi migration will not be backported to 0.10.x. Genshi will >> only be available in versions of Trac >= 0.11. >> >> Once 0.11 is available (even in unreleased form), and the Genshi >> branch has been merged, plugins can start using Genshi, but they will >> not work in Trac 0.9 or 0.10. Plugins still using ClearSilver will >> continue to work as long as ClearSilver is installed. >> >> Backporting the changes required for Genshi would (IMHO) be too much >> work, and we can really find better uses of developer time. >> > > ...like developing in paralell for 0.10 and 0.11 ? > > If I understand your releases right, 0.10.X will be for long time (6 to > 12 months) the official "latest stable release". >
That's a valid point IMHO. I really don't want to have to work on 0.10 (Clearsilver) and 0.11 (Genshi) in parallel for about a year. If it nevertheless goes that way, I personally won't spend much time on 0.10-stable (not that there would be necessarily a need for, as 0.10 is "almost perfect" as we all now ;) ). My take would be to make the coexistence period of Clearsilver and Genshi *as short as possible*, by having a 0.11 release feature-wise more or less equivalent to 0.10, but using Genshi. If we want to implement all the features currently scheduled for 0.11, I guess it'll take us another 10 months (approx.). Now, rather than post-pone all the 0.11 features to 0.12, we could think about creating intermediate milestones, 0.11pre1, 0.11pre2 etc. which will be "feature" milestones, e.g. 0.11pre1 would be "0.10 + genshi", 0.11pre2 would be the integration of WorkFlow, 0.11pre3 migration to setuptools, etc. I think this will improve on the current situation. During 0.10dev, a lot of people ended up using it by following 'trunk', because trunk remained mostly usable during a large part of the development. That's fine of course, and we don't want to change that. However, a lot of people could have been "scared" of using the latest development version, and even if they were interested in, didn't necessarily know which changeset to use, or what was the state of affairs on the trunk. Having "preX" milestones would help clarify that (we close a "pre" milestone and tag it when we now that the trunk is stable), and potentially gain us more beta testers. > If I develope plugins, I need to develop them twice (0.10, then porting > to 0.11). Most possibly, developers will develope with CS, to run under > both, 0.10 and 0.11 (which will delay the "total" Genshi migration). > New plugins should be written for Genshi, I think and old ones could be ported right now. Having new "Genshi-only" plugins will motivate users to upgrade. (he, I'm talking as if 0.10 was dead old, it's not even released yet ;) ) > So, wouldn't it be better to integrate the Genshi modifications in an > non-intrusive way into 0.10, thus developers migrate immediately to > Genshi? 0.11 would then be able to _remove_ clearsilver dependency (as > everything relevant will have been ported until then). > > So, ideally, the "Genshi Enabling Additions" should find their way soon > into 0.10.x - I think everyone will agree here. > No, I'm afraid that this would only make the 0.10 development line last longer, which I think would be counter-productive. Again, if people have enough motivation to go for 0.11 and if we make that possible by having very early "pre" releases, then the coexistence period will be short. -- Christian [1] - http://groups.google.com/group/trac-dev/msg/feea22fb61b06cc6 --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Trac Development" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/trac-dev -~----------~----~----~----~------~----~------~--~---
