labs-labs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Santiago Gala <santiago.g...@gmail.com>
Subject Re: Subject: Re: Lab for distributed SCM?
Date Sat, 23 Feb 2008 12:05:37 GMT

El vie, 22-02-2008 a las 17:46 -0800, Joe Schaefer escribió:
> If people want to experiment with using git-svn or
> svk or anything else, and there's something about
> our infrastructure which prevents that, you need
> only talk to infrastructure about it to see if it
> can be resolved.  In most cases it can, rather easily.


I think having a closed list is a big part of the problem. In "normal"
projects the only list that is not public is purposely called private.
The only exceptions are a few sensitive lists (one or two per
committee/project) ... and *all* the infrastructure ones.

I don't think this makes sense, specially as there is already one
specifically named "private".



El sáb, 23-02-2008 a las 10:37 +0100, Mads Toftum escribió:
> On Fri, Feb 22, 2008 at 09:39:17PM -0800, Matthieu Riou wrote:
> > OpenSolaris is failing with Git but I'm not sure it's fair to put the fault
> > on gSCMs, it's Sun's management that's at fault. OpenOffice has similar
> > issues with CVS. For that matter there are quite a few scripts on SVN to
> > maintain vendor branches (aka forks).
> > 
> Opensolaris uses Mercurial, not git. They spent very long time testing
> git, mercurial and many other options. They did some very thorough
> testing and comparison to reach that result.
> I think Mercurial works perfectly well for the kind of development they
> want to do (any "failings" are due to  politics/legal, not the tool).
> Just as well as I'm sure git does what it's supposed to for the linux
> "benevolent dictatorship" model. But neither of those models are what
> the ASF stands for and I'm very strongly opposed to providing tools and
> support to those who wish to deviate from our model of community based
> development.

There is a lot of tagging in the previous paragraph (mercurial -> good
for opensolaris; git -> linux; none -> ASF). More tagging: dSCM in the
ASF -> "those who wish to deviate".

"deviate"? I hope it is in the sense of "aberrant: markedly different
from an accepted norm" and not in the "pervert: a person whose behavior
deviates from what is acceptable especially in sexual behavior". I guess
I can take the first meaning. But it implies authoritarian thinking, as
there needs to be a "norm" to judge against.

As I was saying, a lot of tagging and not a single line of logical
reasoning. The discourse ends up like: "you are not one of us: go away".
This is all the meaning I can extract. I *am* member of the ASF. I won't
go away or shut up so easily. Specially when insulted. I tend to keep
the proportion, and return approximately the same level that I get.
Please tell me if I haven't this time.


> I completely agree with Roy that distributed SCM isn't for us and I
> agree with Paul that labs isn't the place. On a personal note as part of
> the infrastructure team, I'm also rather disappointed at the actions of
> some of the people who are trying to force their world view in through
> the backdoor and yet there's been not one single mail on the topic going
> to any of the infrastructure lists.
> 

You can have the opinion that suits you more. At least Roy has treated
to convince other people, you are just trying to scare us.

Again: "force their world view in"... this is authoritarian discourse.
The "official" reaction to people asking for different tools has been
exactly this: forcing the established world view using tools. I seem to
recall that the ASF was a "loosely connected group of projects with a
legal umbrella". This world view used to be kosher. What has happened to
it?

Re: the lists, I already said this to Joe: I don't think we need more
private lists. I don't think research like this one needs, benefits or
even wants a closed environment.

If the lab is started, it won't be oriented to "force their world view
in through the backdoor", no need for it. The market is speaking about
this: anybody exposed to different open source project is already
exposed to git, bazaar, darcs and mercurial. I needed to install those
in the last 2 years due to different projects, and in this order. I
found them better tools than subversion. Should I refrain from being a
deviate (i.e., be a conformist) and stay silent?

Regards
Santiago

> vh
> 
> Mads Toftum
-- 
Santiago Gala
http://memojo.com/~sgala/blog/


---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org


Mime
View raw message