Am Dienstag, den 19.08.2008, 09:12 -0400 schrieb Mark Ramm: > We have a new server for TurboGears.org, and Alberto asked if we could > host ToscaWidgets.org there as well. > > This seemed like a good idea to everybody responsible to the server, > but I raised the idea of switching TG2 to mercurial in a private > e-mail to the parties involved. > > Obviously this is something that we should all talk about on the > mailing list before we even think seriously about making a switch in > our core development processes. So, here's my personal thinking so > far: > > 1) Alberto's got a neat setup on the toscawidgets site, where > mercurial, trac, and everything are all part of a single system with a > single username/password. > 2) Alberto's setup would allow us to host trac sites for projects like > silverplate, or tg.ext.geo with their own mercurial repositories very > easily. > 3) Mecurial's main advantages are easy branching/merging and the > ability to do commits offline > 4) Pylons, Beaker, Routes, WebHelpers, and many other upstream > projects already use mercurial. > > Chris Arent has brought up a couple of potential issues which we need > to think through: > > 1) Documentation would need to change > 2) We would have to help people learn mercurial > 3) Mercurial would need to be easy to install on all our target platforms > > I think that there are binary installers for OS X and windows, and > RPM's and debs for all the major linux platforms. And mercurial is > python, so if you have all the nessisary stuff easy_install might even > work too, though it appears from Chris's e-mail to the list that it > might not always work. > I had to install mercurial on all my slaves. It was quite easy on: WinXp,Win2k3,Vista,Debian,Ubuntu,Solaris. I had no problems but I only used it for checkouts, so I don't know anything about merging, commits or something like that. > There's a pretty good introduction to distributed version control at: > > http://betterexplained.com/articles/intro-to-distributed-version-control-illustrated/ > > I'm sure there are some other issues surrounding the move to a more > distributed model. In fact, to prime the pump of that discussion, > here's a very old blog post by Ian Bicking explainining why he's not > 100% sold on the distributed model. > > http://blog.ianbicking.org/distributed-vs-centralized-scm.html >
--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears Trunk" 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/turbogears-trunk?hl=en -~----------~----~----~----~------~----~------~--~---
