Re: Intolerable CVS behaviour

[Tomasz Kłoczko <kloczek rudy mif pg gda pl>]

On 24 Feb 2001, Almer S. Tigelaar wrote:

> Dear Miguel
> 
> (sorry if you get this twice)
> 
> > We could develop a web-based system, or a hacked gtranslator that
> > would pull the `po' files and allow contributors to commit them,
> > without risking any problems.
> 
> 
> 
> Even though this may be true, I don't understand how/why we get from
> someone who commits wrong 'code' to CVS to this whole translation issue again.
> 
> Web translation interfaces are very counter-productive, I love my emacs
> PO mode. Restricting access
> to /po directories under a module might be acceptable to some.

I think restrictions isn't good.
On gnome development I'm start as translator (especialy for mc) but most
of my work (not so many) was fixing same small automake bugs and few 
times some small bugs/tpos in C source code level. Also I think
people who are real C programers also sometimes prepares translations.
By separate this two tasks and prepare restricting access to some
resources you can simple loos many small or very small changes (sometimes
typos).

More than two years I manage cvs.pld.org.pl where all people have full
write acces to everythink (now write acces have ~300 developers). PLD team
uses this for own task but also in the same repository are stored many
other projects (few widely known: minicon, ispell, shadow, pinfo and few
othe not so widely known as BSD tcpd, emma, mudledftpd or solid-po3d) and
also LDP uses oure cvs repo for store all HOWTO and other resources. In
past I observe many times cases where someone who want on begining have
writte access to only one module uses this for fixing sometimes very small
but also sometimes importand bugs. Yes .. sometimes this is cause some
misunderstood or someting but at all gain full rw acces brings more good
thins than bad. I think situation in GNOME repo is very similar. All we
are people and sometims we make mistakes (like me in libxml case) but
after each mistake or missunderstood all must have chanse to fix own
incorrect commit. In this cases more importand is maintainer which must on
first observe all and sometimes send some comment or simple "why you
prepare this in this way ?" before stop cvs write access (I remember one
case where I stop rw acces to PLD repo which I manage, three or four
months after we start).

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie mają problemów, tylko sobie sami je stwarzają*
-----------------------------------------------------------
Tomasz Kłoczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek rudy mif pg gda pl*