www-repository mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Carlos Sanchez" <car...@apache.org>
Subject Re: [repo] /www/people.apache.org/repo/m1-ibiblio-rsync-repository/
Date Wed, 13 Dec 2006 09:37:07 GMT
On 12/13/06, Steve Loughran <steve.loughran@gmail.com> wrote:
> On 13/12/06, Davanum Srinivas <davanum@gmail.com> wrote:
> > Steve,
> >
> > -1 if you don't give us tool for us to do the auditing ourselves or if
> > all the rules are not written down. If repository@ folks keep changing
> > their mind, we cannot read everyone's minds.
>
> The rules will be encoded in Prolog, unless I feel an urge to walk RDF
> graphs in JRuby or Jython. I like the idea of declarative
>
>
> Did I mention RDF graphs? Someone lurking on this list has the ability
> to convert the entire metadata graph into N3 notation for other tools
> to use. Once we've synced up the binaries we can even look inside them
> too.
>
> Paolo -how goes the code? Is it public yet?
>
>
> >
> > +1 if it's well documented, no ambiguity and there is an automated
> > tool/checker we can run before our releases. We can't make a release
> > and then get told that we have to re-cut our release with a new
> > version # because one of the folks here had a brainwave/dream last
> > night. There needs to be a documented policy and specific date on the
> > policy they take effect just like all the infra policies.
>
> oh, I am strongly -1 to cutting a new binary release just because the
> metadata is bad. This is what we have to do today, and its wrong. But
> how about fixing up the MD against an existing (and voted on) release?

I don't agree, do you change RPMs or any other package to fix the
metadata? or new versions are released?

>
>
>
> > thanks,
> > -- dims
> >
> > On 12/12/06, Steve Loughran <steve.loughran@gmail.com> wrote:
> > > On 12/12/06, Henri Yandell <flamefew@gmail.com> wrote:
> > >
> > > > d) Project directories are group-owned by the PMC unix group. commons
> > > > by jakarta, sandesha2 by ws etc.
> > >
> > > -0
> > >
> > > We have too much evidence that the individual projects dont publish
> > > artifacts with a rigorous enough process or adequate metadata.
> > >
> > >
> > > I'd propose
> > >
> > > e) artifacts are copied from the project directores to a staging site
> > > where they are audited and validated before being published in the
> > > central repository. Only artifacts that are considered valid will be
> > > published.
> > >
> >
> >
> > --
> > Davanum Srinivas : http://www.wso2.net (Oxygen for Web Service Developers)
> >
>


-- 
I could give you my word as a Spaniard.
No good. I've known too many Spaniards.
                             -- The Princess Bride

Mime
View raw message