Re: Feeling stupid about references

I was not aware that we did not have access to the extraction code.  That's
unfortunate.

My requirements?  I don't have requirements per se.  Just a desire for
perfection.  As long as the salient information in the reference is correct
(the URI) we are fine.  But inconsistencies between standards are the thing
that make standards lawyers crazy.  So if this one says "xyz - W3C
Recommendation dated whatever" and that one says "xyz, W3C Proposed
Recommendation dated *the same whatever*" someone is going to complain.
 And speaking as the person who single-handedly set XHTML Modularization
back a year and opened the door for the whole HTML5 / XHTML schism because
of just such a clerical error, I am perhaps unreasonably sensitive to it.

Spec inter-dependencies are only going to increase as we go forward.
 Automated extraction of the information in RDF (via RDFa generated from
ReSpec) will make it easier to ensure that the database of references stays
up to date.

Is there a ReSpec reference syntax that can get me a specific version of a
spec?  I know that there used to be just lots of entries (for XML, for
example).  Is there a way to say [XML:specific-revision] and have the
reference in the text say XML but the data for the reference be some
revision?  Maybe that is a way to allow me to add a labelled version to the
repository by hand that I can reference and share with others who need to
reference it before it is actually published.  The alternative is that we
be aware of every spec that is going to be published around the same time
and ensure all the  editors of those specs know that some reference they
are making is going to change and they need to do something local to change
it...  That just feels like it is not going to scale.


On Fri, Jun 21, 2013 at 1:18 PM, Tobie Langel <tobie@w3.org> wrote:

> On Friday, June 21, 2013 at 5:45 PM, Shane McCarron wrote:
> > So... here is my suggestion. Obviously continue to permit local entries.
> Careful, smart people will be able to get those right. But also enable a
> way to tell the automated extraction system that the source of a new
> version is temporariliy *over here* - not yet in TR space.
> That would imply having access to the code that extracts meta data out of
> the specs based on pubrules. Unfortunately, we don't. So whatever algorithm
> we come up with to do the same risks creating the dame same subtle
> difference you were worried about.
>
> I'm not sure I understand precisely what your requirements are. It seems
> you want a discrepancy-free environment (we all do), but I was wondering if
> there were areas where you'd be more willing to be a tad more lenient than
> others (e.g. it's OK if the ref changes slightly after a spec is published).
>
> Thanks,
>
> --tobie
>
>
>
>


-- 
Shane P. McCarron
Managing Director, Applied Testing and Technology, Inc.