On 12/18/06, Stefano Mazzocchi <[EMAIL PROTECTED]> wrote:
robert burrell donkin wrote:
> RAT (release audit tool) is a micro-application which automates basic
> audit checks made on releases. it was developed to reduce the time
> taken to review incubator releases. i've been asked on a few occasions
> why RAT isn't at Apache. here's a more considered public response...
>
> the short answer is that RAT started before labs
>
> the long answer is that IMHO RAT is not really suitable for labs
>
> pros
>
> * labs are cool :-)
> * same karma policy (if you're Apache, karma's fine by me)
> * very low barrier to Apache folks contributing ideas
> * Lab RAT is cheerful word play
>
> cons
>
> * RAT is (sadly) no longer really experimental
> * an Ant task is likely to happen soon. this means binary releases :-/
> * Apache is unlikely to be a suitable long term home for RAT. too
> small for democractic self-organisation but no suitable home as a
> subproject (unless i've missed one...)
>
> but am i right? opinions, if you wish...
>
> i try to keep RAT source only with the minimal dependencies imported
> using svn:externals against tags. labs could be a good home for some
> of the library components that are probably going to be needed. the
> code is likely to be released sooner or later downstream as part of
> RAT. does this sound ok?
I'm *very* much interested in collecting as much RDF metadata about the
operations of the foundation as possible.
Things like projects.apache.org, the new people.apache.org front page,
labs's DOAPizer and RAT all fall in the same category: they plan on
producing and consuming metadata to provide useful services to the
foundation.
+1
This is why I think it would be beneficial if all this activity happened
so that all the people involved watched over one another's shoulder,
rather than in independent silos.
+1
For example, by reading your blog, a lot of what you plan in doing for
RAT is covered by DOAP and recent problems with the people.apache.org
FOAF formats show that RDF is a little more tricky to deal with than XML
(and for very little benefit at this point, I must admit).
I have been working with RDF for almost 4 years now and I have come
across most of the problems that we'll come across here and this is the
reason why I feel labs should try to consolidate the various works on
RDF data creation while leaving others to consume this data they way
they please.
+1
(i was already been thinking about starting the RDF work here :-)
On the other hand, I understand that labs' restrictions might negatively
impact a transition from an external and free location, but that is
something that you have to judge.
i see relocating RAT as orthogonal to finding the best place to
develop meta-data producers and consumers
Either decision won't stop us from using RAT if/when we have to, but
making it a lab it feels it would certainly ease communication (for
example, I don't see myself subscribing to the RAT mail list if it
wasn't part of this... why? well, I feel that even if I do, I would be
talking to you and not many others, while here there is already an
incredible pool of talents lined up and watching, talents that it's
easier to inspire).
So, the way I see it is to choose between freedom of operation vs. being
closer to the metadata creation social activity.
i'd prefer to see an independent effort here in the labs which RAT (or
any other application) can use.
the same goes for the proposals about similar efforts over in
mavenland: i'm not sure whether maven will end up using RAT or
developing their own but i'd like to think that we can collaborate on
many of the basic components required. labs would probably be a good
place to do it.
- robert
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
