Re: Editors: Going with hg or sticking with CVS?

On 10/22/2012 08:27 PM, Arun Ranganathan wrote:
> Consider this a resounding OK from Arun
>
> A few things to bolster my earlier point about working more in public:
>
> 1. Repo + commit histories should be public.  This helps authors (e.g. 
> it helped author of book on subject; also helps others see changes to 
> issues and who fixed them, including commit comments of course).
>
OK, we'll have the HG set-up by TPAC then!
> 2. Work flow tools should be public.  WebApps WG uses the W3C bugzilla 
> installation.  This allows non-WG members (and those that don't attend 
> our Conf. Calls) to file bugs on the spec.  This has been invaluable 
> (e.g. Ms2Ger has already been cited as an example .)  Not 100% sure 
> about Issue Tracker, but it seems that while the output is public, the 
> participation is WG-only.

I'm not sure re Bugzilla vs. Tracker. We seem to have a decent flow 
using Tracker, and right now we have both set-up. Earlier, it seemed 
Tracker would be used for WG management while Bugzilla would be used 
more for public comments etc. Is that OK?

>
> ----- Original Message -----
> > As it seems all non-WG members who commited to DomCrypt (i.e. man of
> > mystery ms2ger) are OK with HG, so as soon as I get an OK from Arun
> > I'll
> > hit the switch button and update the WG homepage. Hopefully by next
> > meeting!
> >
> > On 10/19/2012 07:27 PM, Ryan Sleevi wrote:
> > > On Fri, Oct 19, 2012 at 10:01 AM, Harry Halpin <hhalpin@w3.org 
> <mailto:hhalpin@w3.org>>
> > > wrote:
> > >> So I checked in with Systems Team over
> > >>
> > >> cvs.w3.org:/w3ccvs
> > >>
> > >> being available or mirrored magically with:
> > >>
> > >> dev.w3.org:/sources/public
> > >>
> > >> as currently, looking at dev.w3.org <http://dev.w3.org>, there's 
> no "2012" branch
> > >>
> > >> They responded that dev.w3.org <http://dev.w3.org> and cvs.w3.org 
> <http://cvs.w3.org> are deliberately
> > >> separate
> > >> repositories with different access rights and different services
> > >> so should
> > >> not be mixed as only cvs.w3.org:/w3ccvs reflects on www.w3.org 
> <http://www.w3.org>
> > >> mirrors while
> > >> dev.w3.org:/sources/public has publicly visible cvsweb and
> > >> anonymous public
> > >> cvs pserver.
> > >>
> > >> Another option is we have a HG (mercurial) repo
> > >>
> > >> https://dvcs.w3.org/hg/domcrypt
> > >>
> > >> That we could reset to make "webcrypto" and then move over Editors
> > >> Drafts of
> > >> both the existing API and a "new" use-case document over there.
> > >>
> > >> Editors - any opinion?
> > >>
> > >> Also, a plus of dev.w3.org/dvcs.w3.org 
> <http://dev.w3.org/dvcs.w3.org> is the public nature of the
> > >> repos.
> > >> Arun thought that the more public, the better, and I tend to
> > >> agree.
> > >>
> > >>     cheers,
> > >>         harry
> > >>
> > >>
> > > HG. All the way. I have previously expressed support for this on
> > > calls.
> > >
> > > If anything, simply being able to diff between revisions is worth
> > > whatever reposistory-switch overhead. I'm a big fan of small
> > > commits
> > > with easily referenced URLs, which the current CVS system does not
> > > encourage.
> >
> >
> >