Hi all,
As last Friday's mails have demonstrated, there is disagreement between
us on what the Fresnel styling part should do and how these goals should
be achieved. The discussion is scattered between several threads and
sometimes addresses problems that are no longer relevant. I'm trying
here to give a summary of all this, with the (conflicting) opinions that
have been voiced, so that we can start on new grounds and go forward.
I'm splitting this summary into two emails, one dedicated to Lenses and
the other one to Styles.
Jacco, Lloyd, I am answering some of your questions and comments in
those two emails.
-----------------------
I. Lenses
-----------------------
That is the part we all pretty much agree on, as presented in the
Fresnel manual [1]. Some issues, though:
1. Jacco's wondering about the possibility to define lenses as
refinments of other lenses. We do have such a mechanism. See section 5.2
Lens Inheritance [2]. Does this meet your needs?
2. How do we resolve conflicts between different lenses that apply to
the same instance? This is indeed still underspecified, and we need to
work on this. We can define a specificty computation method for FSL (as
there is one for XPath in XSLT and one for CSS selectors). We can define
on for SPARQL queries. The main problem comes from the fact that we have
no method for comparing an FSL expression with a SPARQL query right now.
Note that we do *not* have the same problem when it comes to comparing
basic selectors (as defined in section 1.3 - Simple RDF naming) and FSL
expressions, as basic selectors are actually all valid FSL expressions
with the same exact syntax and semantics.
3. Is it right to determine the order of appearance of properties in the
lens, rather than in the style? Nobody seems to have strong feelings
about this. I don't. If I had to choose, though, I might leave this to
the styling part. This does not really impact the current Lens
vocabulary. It just means that the set that is the object of
fresnel:showProperties properties should not necessarily be intepreted
as a sequence, though it could if ones wants to).
4. Selecting class-less (untyped) RDF resources: as Stefano said, this
is possible. Most of the time you will need FSL or SPARQL to do that,
but it is definitely possible.
[1]
http://simile.mit.edu/repository/fresnel/trunk/docs/manual/FresnelVocabulary.htm
[2]
http://simile.mit.edu/repository/fresnel/trunk/docs/manual/FresnelVocabulary.htm#inheritance
--
Emmanuel Pietriga
INRIA Futurs - Projet In Situ tel : +33 1 69 15 34 66
Bat 490, Université Paris-Sud fax : +33 1 69 15 65 86
91405 ORSAY Cedex FRANCE http://www.lri.fr/~pietriga
Received on Sun Apr 24 2005 - 10:18:01 EDT