Re: ACTION-11 Proposed Short Names

On 2010-2-27 16:30 , Shane McCarron wrote:
> 
> 
> Ivan Herman wrote:
>> I practical comment, though. There has always been a question in the
>> past whether DTD-s or namespace documents or the like should be part of
>> the core distribution as stated above. If a DTD gets stored in /TR space
>> then, according to W3C rules, they cannot be touched any more even if
>> the problem is just a stupid and obvious mistake. If the DTD is stored
>> somewhere else, such changes become easier. The same with RDF
>> vocabularies and the like. I wonder whether we should not treat them
>> separately from the start.
>>   
> 
> Yes - I agree.  In fact, while during development I have these in the
> tree, I don't leave them in the tree at publication time.  We should
> probably identify a location where those could live - or just use the
> location we already have (MarkUp/DTD, MarkUp/SCHEMA).  But I assume that
> the files should be in the source tree and be under source control while
> they are being developed?

Everything is under source control on the W3C web site, wherever you put
it:-) So sure, the editor has a choice to decide where to put thse files
when under development. The only thing we will have to be very careful
about is that when the team contact has to copy the files for final
publications; it should be clear what goes where... I will have to talk
to the team contact about that:-)

B.t.w. I would prefer to have them under the rdfa tree and not MarkUp.
The reason is very simple: RDFa WG people may have CVS read/write access
to that part of the file system, but not to MarkUp.

(Hm. Maybe what this will mean is that for RDFa1.1 XHTML DTD the
location will be somewhere radically different than now? Not ideal...)

> 
>> Also: we may have to have separate namespace documents for RDFa (eg, for
>> small RDF vocabularies). Those may be stored separately these days as
>> http://www.w3.org/ns/XXX.
>>   
> 
> Actually, I would personally object to putting a vocabulary in the NS
> space.  The NS space is for namespaces.  Putting a vocab in there would
> only server to further confuse the line between namespaces and
> vocabularies.
> 

While this is true, the djinn is already out of the bottle: eg, the RDF
vocabulary terms for entailment

http://www.w3.org/ns/entailment

has already been defined there. It may not be ideal, but I would still
prefer to have

http://www.w3.org/ns/rdf

rather than the long official rdf vocabulary/namespace declaration that
we have today and I never ever remember:-(


>>> Finally, in terms of structure on W3C's server in the RDFa area, I
>>> propose:
>>>
>>> 2010/02/rdfa/
>>>   drafts/          - place where drafts for public consumption are
>>> dropped
>>>      Overview.html - file that lists the documents in progress and has
>>>     
>>
>> I'd prefer to use the Wiki for such lists rather than a separate HTML
>> file. Remember the Wiki is, by default, editable to everyone in the
>> group and does not need an additional CVS access (the latter will be set
>> up for all editors but not necessarily for others).
>>
>> Ie, there should be a Wiki page with a table of the history for each
>> document as well a pointer to the latest editor's draft.
>>   
> 
> Yeah, I don't mind that.   Could the drafts folder redirect to the wiki
> page?

Yes. Just tell me what to redirect and the rest is just a matter of
putting a line into .htaccess

> 
>>> links to the latest
>>>      2010/         - folder for each year
>>>        ED-short-name-date/  - folder for each published ED
>>>     
>>
>> I actually wonder whether it is worth storing the previous drafts under
>> the rdfa tree. After all, once a draft is published, we are not supposed
>> to touch it; for historical purposes the W3C publication process does
>> make a copy anyway. So why storing duplicata?
>>
>> I presume the XHTML WG did that; the SPARQL or OWL WG did not. I would
>> like to understand the thinking behind it...
>>   
> 
> I might have confused you.  The only thing I was suggesting would be in
> there would be editor's drafts.  In the XHTML activity and RDFa Task
> Force we would from time to time decide that the changes we had made
> were ready for working group review but not really ready for formal
> review like a working draft.

Ah, I see what you mean! I guess this is very much a matter of personal
work style. If I was a lead editor, I guess I would work on my own local
machine and would cvs commit a new version when I felt that it is good
enough for internal review. For that, a single directory like
/rdfa-core, etc, would be enough. You seem to prefer to have a
semi-private working area on the site, and make separate snapshots time
to time. I presume both approaches are perfectly fine... I wonder
whether we should not leave the choice to the editors on that. I just do
not know... (the two Working Groups that I was active in the last few
years, ie, SPARQL and OWL, did not have such extra snapshot systems.
Actually, the OWL WG edited everything on the Wiki, and there was some
separate Sandro magic to convert the wiki content to real html files...)

Cheers

Ivan


> 

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF   : http://www.ivan-herman.net/foaf.rdf
vCard  : http://www.ivan-herman.net/HermanIvan.vcf