| 1 |
On Sun, 10 Apr 2005 23:57:12 +0200 Christian Parpart <trapni@g.o> |
| 2 |
wrote: |
| 3 |
| > SVN uses transactions and |
| 4 |
| > changesets. These make a heck of a lot more sense if they're done on |
| 5 |
| > a per project basis. |
| 6 |
| |
| 7 |
| reason? |
| 8 |
|
| 9 |
Because you can pull out a meaningful and relevant changeset without |
| 10 |
having to arse around with path prefixes. |
| 11 |
|
| 12 |
| > Unlike with CVS, this makes a big difference -- SVN |
| 13 |
| > revision IDs are actually meaningful, |
| 14 |
| |
| 15 |
| SVN repository IDs represent the state of the whole repository at a |
| 16 |
| given time, nothing more or less. |
| 17 |
|
| 18 |
Not repo IDs. Revision IDs. |
| 19 |
|
| 20 |
| Hmm... besides, the ASF is just having a single repository for all |
| 21 |
| their public projects (with about 1000+ contributors) w/o any |
| 22 |
| problems. |
| 23 |
|
| 24 |
So we should make the same mistakes as them? Sure, a single repo would |
| 25 |
be usable, but multiple repos would be a heck of a lot better. |
| 26 |
|
| 27 |
-- |
| 28 |
Ciaran McCreesh : Gentoo Developer (Vim, Fluxbox, shell tools) |
| 29 |
Mail : ciaranm at gentoo.org |
| 30 |
Web : http://dev.gentoo.org/~ciaranm |
Archives