New port with maintainer ports@FreeBSD.org [was: Question about maintainers]

[Babak Farrokhi]

Pav,


On 7/31/05, Pav Lucistnik <pav at freebsd.org> wrote:
> >
> > Let me give you an example: I am the maintainer for www/eventum. The
> > current version in ports tree was 1.5.4 so I submitted the patch for
> > 1.5.5 (ports/84297) and now version 1.6.0 is out but the patch is not
> > submitted despite I was the maintainer myself.
> 
> Submitted on Friday, complained about it on Saturday?

My bad. You are right. I really didn't notice the date while I
originally had ports/83960 in mind to give as example.

> 
> We just don't have a _manpower_ to process all incoming PRs in 24 hours
> after arrival timeframe. Your simple update of eventum took me 20
> minutes to fix, test and commit. I have it pretty extensively automated.
> New ports usually takes longer, depending on how much committer have to
> fix. Trust me, some submissions are really useless.
> 
> Should we just kill those? I believe not.
> 
> Now take a look at http://www.oook.cz/bsd/prstats/busters-ports.html
> those are numbers of closed PRs in past 3 months. That gives me
> 8 PRs/day. In practice I'm spending some four hours a day on it.
> I got a paid fulltime job I have to do, and I also got some non-FreeBSD
> pasttime activities, commonly called "a life".

I am a ports freak and I always have a freshports.org rss feed to
track ports changes as well as submitting updates and patches. It is
obvious that ports team is doing a very good job with testing and
fixing submissions. I personally learn too many things when you clean
up the code before submission and I would like to thank you for the
good job. (e.g. ports/84297). You are teaching people to submit
cleaner codes and hopefully become commiters to help the project. I
hope I can help someday.
 
> 
> The point here is to view the situation from the standpoint of
> committers. No one here can be doing this full-time. And people are
> doing this for fun, don't forget.
> 
> Now a lot of committers spend a lot of time maintaining their own ports,
> which are often complex and heavinly used.
> 
> What we could really use be some dedicated people with a lot of free
> time and a good skill in ports. Those people are hard to find.
> 
> > Another example: I submitted patch to update editors/vim to patchlevel
> > 79, now this version is vulnerable to arbitrary command execution
> > according to CAN-2005-2368. So I submitted the patchlevel 85
> > (ports/84145) and also notified security-team at . But the port is still
> > awaiting approval.
> 
> Well yes, it was three days old when you urged this at secteam. They
> decided to wait on maintainer instead of rushing it in, as the
> vulnerability is not that severe.
> 
> > There is really something wrong with the port management process.
> > People's work is not being respected. So how do I get encouraged to
> > submit my patches?
> 
> Now you cut yourself with double-edged sword. You want us to respect
> your submission by not respecting O'Brien's maintainership?
> 
There is no point to bypass the maintainer approval except in
emergency cases which does not apply to my submission to this port.
However I only contacted secteam to notify them of the vulnerability
and speed up the process (like I already did for ports/83006). I just
thought maybe its a good idea to reduce the maintainer-timeout for
security patches.

> --
> Pav Lucistnik <pav at oook.cz>
>               <pav at FreeBSD.org>
> 
> And now something completely different.
> 
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.1 (FreeBSD)
> 
> iD8DBQBC7K3TntdYP8FOsoIRAlu8AKCYgPe1kdvxYDbASRbeeQ4bHjMBUgCgkU5m
> DXa66FdpBLkITcUydWM709E=
> =pU33
> -----END PGP SIGNATURE-----
> 
> 
> 


-- 
Babak Farrokhi
email: babak at farrokhi.net
web: http://farrokhi.net/