On 9/19/07, Yuri Baburov <[EMAIL PROTECTED]> wrote: > I'm willing to add schema-evolution into django.
Sweet! I can't wait to see the code! However, I don't see what this has to do with commit access. The right way to develop a feature of this size is as a third-party project -- see django-openid, django-registration, django-tagging, etc. Register a project on Google Code, and hack away. > So I started updating the code yesterday. It is not required for this > project to be integrated into Django itself anymore, but I think good > solution of schema-evolution should be placed in the django project to > never become outdated like it is now. OK... so how exactly will rolling a feature into core prevent it from becoming outdated? That's the huge problem with merging something big like schema-evolution in: we have to be sure that the developer(s) won't disappear and leave the rest of us to maintain some huge amount of code. Recognize that it's going to take a *long* time for something of that level to get merged, and become comfortable with the fact that it may never happen at all. Every single core developer would need to understand and approve of the code, and we'd need to find a "champion" who we can trust to provide long-term maintenance. Those are high hurdles to cross, I know, but them's the breaks. > Also I realized today that I don't want to write patches to django. > Mostly because this process is very inefficient to people from outside > - I get no feedback on tickets unless I raise the question in this > conference. And you can't commit current patches, marked > 'ready-to-checkin', for months. In Trac, I see four tickets filed by you: #4497, #5018, #5032, and #5273. Two were marked wontfix: * #4497 is actually really a duplicate (Malcolm marked it incorrectly) of #5532. * #5018 contains a good idea, but was marked wontfix because the implementation is suboptimal. One (#5032) is a duplicate. The last one, #5273, is a tiny, nearly trivial design change, which has been open for three weeks. While I'm sorry we didn't get to it earlier, three weeks is hardly a painful amount of time to wait for a "None" link in the admin. As far as I can find, no ticket reported by you has been open for "months", and in fact only a single one of your tickets is still open. So while I'm truely sorry you feel this way -- really, I am -- I'm having trouble seeing exactly why you're so frustrated. > How do I become committer in this situation? :) Contribute code. Don't get angry when tickets are marked wontfix. Contribute more code. Be patient when tickets take a long time to get checked in. Contribute code. Recognize that we're going to critique your coding style, sometimes harshly. Adapt, and contribute more code. Every added commiter (i.e. everyone except me and Adrian) had contributed dozens (if not hundreds) of lines of code before we even *thought* about adding them. Some contributed large swaths of code in the form of magic-removal or similar efforts; some contributed many, many small patches. None of them, you'll notice, publically called out the core developers or insulted other members of the community. You might be surprised to know that a number of our those commiters were also unhappy with the pace of Django's development before they got the commit bit. In each case, they were polite enough to email me and/or Adrian privately. If there's a connection between politeness and commit access, it's not an accident. Jacob --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django developers" group. To post to this group, send email to django-developers@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/django-developers?hl=en -~----------~----~----~----~------~----~------~--~---
