On 26/06/2008 11:58, Christian MICHON wrote: > On Thu, Jun 26, 2008 at 12:18 PM, Mike Williams > <[EMAIL PROTECTED]> wrote: >> Sorry to rain on the git love fest going on but ... >> >> There are other options. There is no killer reason to use git over the >> other systems, and the need for a Cygwin base system for windows does >> raise the bar for Windows users - not everyone will want to have to >> install and maintain a Unix like command system just to use a source >> control system. It would be nice to have a reasoned review of the >> options for Vim development. > > there's no need for cygwin, since there's msysgit (standalone package).
Indeed, but it does still include the Unix command line environment. That is two tools to learn for some. I'm just saying ... > and you can survive with only git-gui in vim particular case as I mentioned. > so the people afraid of command lines won't even notice. Hmm, Vim developers afraid of command lines. ;-) > the global idea of this thread is to simplify maintenance on vim. > >> Mercurial is widely used, with a possible plus of svn like command >> interface. Bzr is also good, if not so widely used. IIRC there has >> been a recent SVN release that supports disconnected working (one of the >> big selling points of DVCSs) so perhaps that should not be discounted >> just yet. > > explain to me why installing a bloated interpreter just to use > mercurial is any better than cygwin then ? ;-) Touche. But neither Bazaar or Mercurial require a full install of Python. FWIW the latest standalone installers for Mercurial and Bazzar are ~2.9MB and ~5.8MB respectively. The latest msysGit is ~8.8MB. Now, about this bloat ... > people have a clear and common misconception here: that git is still > difficult to use/understand. it's not true anymore. git-gui is all > that is neeeded. As I said before, things are changing fast in the world of DVCSs. There is a fast rate developemtn and no one has a clean lead over another (and yes git was renowned for years as having poor documentation - the old "the code is the documentation" approach). As ever, don't be blinded by the length of the check box feature list - you wont use most of them anyway. > it may look like a "git love fest", but it's clear that today > maintaining a project with CVS/SVN is clearly time poorly invested. DVCSs are nothing but another tool. They work better in a common popular mode of development, but centralised VCSs still have a place in this world, just not as many as they used to! > I'll prove my point once my vim repo is up. Don't need to convince me, I am happily using a DVCS. The issue is a change in working practice - and no one likes change at first. All those ad hoc tools and practices you have built up no longer apply. And then finding it doesn't support the way you used to work is seen as a reason to not change. The insight is realising there is a new way of working to achieve the same ends, if not achieve more, and that can take a little time. Oh, and you have go to want to change. TTFN Mike -- I am Pentium of Borg , Division is futile , You will be Approximated. --~--~---------~--~----~------------~-------~--~----~ You received this message from the "vim_dev" maillist. For more information, visit http://www.vim.org/maillist.php -~----------~----~----~----~------~----~------~--~---
