Re: cvs commit: ports/www/lusca-head/files patch-ab patch-ac

From: Gabor Kovesdan <>
Date: Sun, 06 Dec 2009 22:24:36 +0100
Adrian Chadd escribió:
> I'm sorry for the drive-by commits like this - I had tested the port
> locally and deployed it in production; I didn't realise there were so
> many issues until after the commit.
> I plan on bumping the port revision once I've established that all of
> the post-install issues are fixed. I think they are now.
> I apologies for the noise,
Adrian, there's always a normal mistake factor, we all make mistakes. 
But for example, looking at your first two follow-up commits after the 
additon of the port, it seems that there were also very trivial 
mistakes, which probably could have been found by someone with 
experience in the ports development. The problem is that you obviously 
don't have that experience and you take it perfectly natural to commit 
to something you are not quite competent in without any review or 
approval request. Even ports committers send a call for reviewers when a 
port is unusually complex. It is a very important difference between 
ports and src that the ports tree isn't branched but supposed to be 
stable, so we do our best to keep it actually as stable as it can be.

I'm a ports/doc committer but sometimes I happen to commit to src. But 
when I do it I always do it with review or approval even if it is a PITA 
to find someone, who is interested and has time to review my work. But I 
always do it because I don't want to break anything and actually I don't 
have privileges to do it without approval. This is the policy described 
in committers-guide and in the welcome mail from core/portmgr/doceng 
when one gets a commit bit. And I don't see any reason why it should be 
different with src committers but as I said this isn't the first case 
that an src guy breaks something in the doc/ports trees because he 
thinks he can just figure it out. That's why I got a bit upset but 
please don't take it personally. In the former cases I always kindly 
offered help but seems that nothing has changed so I'm getting tired. I 
think our policies can guarantee the high quality of FreeBSD and I 
really don't want it to be degraded like in the big chaos of Linux.

Gabor Kovesdan
FreeBSD Volunteer

EMAIL: .:|:.
WEB: .:|:.
Received on Sun Dec 06 2009 - 21:24:46 UTC