[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#754910: cgmanager_0.20-1_amd64.changes REJECTED



[editing to only reply to parts]

On Thu, Jul 17, 2014 at 02:30:26AM +0800, Thomas Goirand wrote:
> I'm sorry if this sounds not so cool, but I'm not sure I get this.
> Ansgar wrote that there should be a -dev package (on which Daniel wrote
> back that he thought it'd be micro-packaging, which is something that
> the FTP masters have for a long time advocate against), and now you're
> advising that dropping the -dev package could be a solution. So what is
> it that Daniel should be doing exactly, in a way that the FTP masters
> would accept?

From Ansgar:

| If it is a private library, then there should be no -dev package. If you
| include a -dev package, other packages may use the shared library and it
| is no longer a private library (even though no package currently might
| do so).
| 
| So currently I don't think the included shared library is actually a
| private library. Please either a, don't include a -dev package, or b,
| build a proper shared library package (and please use lib* and lib*-dev
| in that case).

    <53BA6738.9090103@debian.org>

> I've by the way clearly told Daniel that he was wrong for not opening an
> ITP (he really is). However, one of the reasons we're having ITPs, is so
> that others can oppose to it.

No, it's so we don't duplicate work.

> say Serge Hallyn should be the maintainer of cgmanager, just because he
> opened an ITP. That's not how ITP should work in Debian, and that's not
> how they do.

Currently, neither are the maintainer. I asked them to resolve it before
one of them did become the maintainer. ITPs are mutexes, not ownership,
which is why I didn't let Serge's package through.

FTR, I didn't see these issues on his package, and after a review, it might
have turned out to be fine for the archive.

> So, talking about "race condition" as you put it here, is kind of an
> overstatement to me.

No. It's not. You file an ITP to avoid duplicating work. You didn't file
an ITP (or have dba do it), so work was duplicated. Now we're dealing
with it. That's a race condition. Lesson learned?

> Well, I can *advise* him to do that, but there's nothing in our by-laws
> that forces him to do it (for all of your 3 points). So it wont be
> mandatory for me sponsoring packages, or for the FTP masters to accept
> them as much as I know.

No, you just waste contributor time like this.

> By the way, ITP are a courtesy, and I've never read that they were
> mandatory (have you seen FTP masters rejecting packages on that ground?
> I haven't...).

No, and we havn't, but they waste everyone's time by not filing them.
Just like spitting on the sidewalk isn't illegal, but very rude to
others.


Cheers,
  Paul

-- 
 .''`.  Paul Tagliamonte <paultag@debian.org>  |   Proud Debian Developer
: :'  : 4096R / 8F04 9AD8 2C92 066C 7352  D28A 7B58 5B30 807C 2A87
`. `'`  http://people.debian.org/~paultag
 `-     http://people.debian.org/~paultag/conduct-statement.txt

Attachment: signature.asc
Description: Digital signature


Reply to: