Re: RFR: webalizer - web server log analysis program
On 01/15/2011 03:54 AM, gregor herrmann wrote:
> I don't agree here; I've seen and been involved in quite a few
> maintainer handovers that worked perfectly without involving the BTS.
>
> Of course if the new maintainer needs a sponsor, the sponsor will
> want to see some proof of the old maintainer's admission.
Which is what I am asking here!!! Jonathan talked about a private email.
I wont
sponsor an upload that would take-over the package just based on that.
> Just out of curiosity:
>> I am very familiar with Git, it's just that normally, we use upstream-sid /
>> debian-sid,
>>
> I'm rather unfamiliar with git but sometimes I stumble over packages
> maintained in git anway; and I haven't seen upstream-sid and
> debian-sid branches yet. Who/where (team, tool, ...) is this schema
> used?
>
I first tried to do my packaging trying to integrate with the work of the
pkg-php team. I maintain lots of php-* (PEAR packages), and that is what
Raphael Geissert and others advised me to do. Given the way I was replied
to (rather con-descendent) I thought I was a silly guy not to know that
fact,
and I thought it was a general habits inside Debian. According to what I
have just read here, there's no real rules, and (like often in Debian), the
reality is just a big mess.
Anyway, using a "pristine-tar" branch and a "debian" branch seems not
enough to me, whatever naming scheme we choose, because we need to
track changes for all flavor of Debian.
Using "stable" or "old-stable" doesn't seem really a good naming scheme,
because when a new version of Debian is released, the meaning changes.
So using lenny, squeeze, sid, rather than old-stable, stable, testing,
unstable,
seems more accurate to me. Which leads again to use:
- upstream-sid / debian-sid
- upstream-squeeze / debian-squeeze
- upstream-lenny / debian-lenny
as branch names, which avoids any kind of possible confusion (and doesn't
forces you to act upon a new Debian release).
Cheers,
Thomas Goirand (zigo)
Reply to: