| 1 |
begin quote |
| 2 |
On Tue, 3 Feb 2004 09:09:57 -0500 |
| 3 |
Kurt Lieber <klieber@g.o> wrote: |
| 4 |
|
| 5 |
> > a) 4 / year , retention 1 year... |
| 6 |
> > |
| 7 |
> > can we start with 2/year ? |
| 8 |
<snip> |
| 9 |
|
| 10 |
> I feel it is important to align these release procedures with the rest |
| 11 |
> of our release procedures. Since that is currently a quarterly release |
| 12 |
> schedule, I'd prefer to keep this quarterly as well. |
| 13 |
|
| 14 |
well, that might be a point, however it could also overload us. |
| 15 |
|
| 16 |
|
| 17 |
> > b) Updates and update distribution: |
| 18 |
> > Updates should be distributed -separately- |
| 19 |
|
| 20 |
> I don't necessarily agree with this, but it's simple enough to provide |
| 21 |
> a"2004.1-stable.tbz2" snapshot that people can use to have a totally |
| 22 |
> unchanging tree. Then updates could continue to be distributed via |
| 23 |
> the rsync mirror tree. |
| 24 |
|
| 25 |
|
| 26 |
Well, the point is the act to separate the updates from the "release" in |
| 27 |
oder to ease all external maintainers work. The important thing to note |
| 28 |
here that this sort of "stable" release is to make it simpler for |
| 29 |
downstream administrators. in that regard they are likely to handpatch |
| 30 |
a bundle of the systems, and will therefore want a separate outlet for |
| 31 |
the updates, so they don't merge into the baseline, and thereby create |
| 32 |
additional problems for maintainance. |
| 33 |
|
| 34 |
|
| 35 |
Yes, this is quite important and is actually a requirement for at least |
| 36 |
our organization. (Reason we cannot use either Gentoo or Debian at work) |
| 37 |
|
| 38 |
//Spider |
| 39 |
|
| 40 |
-- |
| 41 |
begin .signature |
| 42 |
This is a .signature virus! Please copy me into your .signature! |
| 43 |
See Microsoft KB Article Q265230 for more information. |
| 44 |
end |
Archives