| 1 |
On Sat, 2008-02-02 at 09:45 +0000, Duncan wrote: |
| 2 |
> > How about just some elog "If you use make install, emerge --noreplace |
| 3 |
> > debianutils" in the kernel's postinst or something. |
| 4 |
> |
| 5 |
> Well, that doesn't help those who get their kernels elsewhere, which IMO |
| 6 |
> are the ones most likely to have their own scripted solutions using make |
| 7 |
> install and thus debianutils in the first place, thus the general |
| 8 |
> announcement suggestion, but a kernel-2.eclass einfo as proposed is a |
| 9 |
> start and certainly won't hurt. =8^) (Also, as I've said, folks with |
| 10 |
> such site-customized scripts are precisely the ones advanced enough they |
| 11 |
> /should/ catch it and be well able to make the necessary changes |
| 12 |
> themselves, regardless of warning, but should != will, and a GMN/GWN/ |
| 13 |
> upgrade-guide warning is a nice touch in any case.) |
| 14 |
|
| 15 |
Well, we can hold off adding it to the actual tree until after the next |
| 16 |
GMN, so that's not really an issue. |
| 17 |
|
| 18 |
My main question was if there was any opposition, and it appears that |
| 19 |
there is none, so long as information is provided to our users. I'm in |
| 20 |
total agreement. |
| 21 |
|
| 22 |
If I don't hear any complaints, I'd like to go ahead and get a GMN |
| 23 |
article together. |
| 24 |
|
| 25 |
Do we really need a front-page news item for this? I don't think the |
| 26 |
impact would be terribly great, as I suspect this affects a very small |
| 27 |
number of people whom are likely to be more advanced users. I think |
| 28 |
adding the info to GMN/gentoo-dev-announce/gentoo-user should suffice. |
| 29 |
If someone thinks we need to do a front-page article, we can do that, |
| 30 |
too. |
| 31 |
|
| 32 |
-- |
| 33 |
Chris Gianelloni |
| 34 |
Release Engineering Strategic Lead |
| 35 |
Games Developer |
Archives