[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Fwd: failure notice (I'm pissed now)

From: Erik Huelsmann <e.huelsmann_at_gmx.net>
Date: 2004-10-31 08:33:02 CET

Hi Karl,

> "Erik Huelsmann" <e.huelsmann@gmx.net> writes:
> > Ok, now you really pissed me off:
> >

[ snip ]

> > *All* of the above is in violation with project procedure or ethics.
> >
> > One more thing -if that's at all possible- and I'll have to propose
> commit
> > access revocation.
>
> Oooh, ouch :-).
>
> This might be a *bit* harsh, Erik, although I agree with the some of
> content of what you are saying. But not all. Note that:
>
> 1. Quite a few of our contributed scripts have the author's name and
> email addresses in them. We've been okay with that, because they
> really are developed in a different way than the rest of
> Subversion. (The "territoriality cost" is lower, while the
> "contactability" benefit is arguably higher.)

Well I don't agree with the 'territoriality cost' being lower since we want
*all* of Subversion to be as good as it can be, not only the core part. If
someone has the skills to co-maintain a something in contrib/, then (s)he
should feel invited to do so. On the other hand, I do agree with the
contactability benefit, so I guess we'll leave it at that.

> 2. Yes, we want the tigris account name to be the same as the SVN
> committer name. But it would be reasonable to blame me, or
> whoever, for not having noticed the discrepancy, since we were
> the ones who set up the SVN commit account. (The requirement was
> not even documented until fairly recently, as I recall.)

This is why I wrote 'procedure or ethics': The project is filled with
examples of how to do things right. Our standards are high, meaning that
there are only a little number of places where things are not as they should
be.

> 3. We haven't been perfectly consistent about including the email
> addresses as well as the names of contributors whose patches we
> apply. If Stefan was simply using the logs as an example, he
> could have been guided either way. Sure, *most* contribution
> commits include an email address, but check out r10889, r10791,
> r10773, r10690, r10389, r10217, r9894... among many others.
>
> We simply cannot get mad at Stefan for failing to follow an
> undocumented guideline that was always inconsistently applied.

That's the problem with ethics, not everybody stays in line. But I would
say there are enough examples between HEAD and 9894 to outweigh the few you
pointed out. Anybody looking at the last 100 trunk commit logs would have
to conclude that usually e-mail addresses *are* part of the attribution.
Even amongst committers e-mail addresses are added: that sure should be a
strong hint.

> The core of your complaint is about log formatting, and I agree Stefan
> should start formatting the log messages in the usual style (and fix
> the old ones). However, this hardly rises to the level of commit
> access revocation -- and the fact that his mail bounced doesn't
> amplify the offense, even if it did make your life a bit inconvenient.

Alright. It being morning now, I'm somewhat milder. You're right that the
mail part does not amplify the offense of log formatting. It was the
cumulation which made me bring that part in. But I can't remember one log
message being correctly formatted at commit time since I joined the project.
 Ok, so his commit rate is low. That does not make it alright to just commit
and see if a committer complains.

I guess my point is: how many times do you get to do it incorrect before
it's decided you just can't do it?

> Stefan, could you please fix your email address in COMMITTERS, fix up
> this log message, and fix up previous log messages?

Heh. That last part is even *stronger* than what I said :-)

bye,

Erik.

-- 
Geschenkt: 3 Monate GMX ProMail + 3 Top-Spielfilme auf DVD
++ Jetzt kostenlos testen http://www.gmx.net/de/go/mail ++
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun Oct 31 08:33:25 2004

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.