| 1 |
On Mon, 4 Feb 2002, Einar Karttunen wrote: |
| 2 |
|
| 3 |
> I don't think so. The work would just go into specifying keywords for the |
| 4 |
> ebuilds. If we have categories it is easier to create local categories |
| 5 |
> and organise ebuilds by criterion the original author didn't think of. |
| 6 |
> For example lynx could be put into console-apps and web-browsers, but |
| 7 |
> this may be done without touching the ebuild at all. So someone may create |
| 8 |
> an category without consulting the authors of the respective ebuilds. |
| 9 |
> This would remove the dependency of managing categories and ebuilds |
| 10 |
> and distribute the work. Also new experimental ebuilds could be added |
| 11 |
> quickly into portage and integrated into official categories when they |
| 12 |
> prove successful. Gentoo installations could be copied by creating |
| 13 |
> a custom category containing all installed applications and then |
| 14 |
> merging it on another server. |
| 15 |
> |
| 16 |
|
| 17 |
You've got a point there, I didn't think about that but you've pointed to |
| 18 |
the other side of the coin. Evaluating the pro's and con's I think you're |
| 19 |
right and we should not autogenerate those categories. |
| 20 |
|
| 21 |
Paul |
| 22 |
|
| 23 |
-- |
| 24 |
___ |
| 25 |
/~~~\ | Paul de Vrieze |
| 26 |
| O-O | | Student of information management and technology |
| 27 |
| _ | | Mail: Paul@××××××××.net |
| 28 |
\___/ | Homepage: http://www.devrieze.net |
Archives