Martin McEvoy wrote:
On Sat, 2008-02-02 at 14:20 -0500, Manu Sporny wrote:
This is the start of an argument for namespaces:
http://en.wikipedia.org/wiki/Namespace
I think microformats already DO support limited namespaces, not fully
loaded namespaces like RDFa does, but kind of micro-namespaces, its
difficult for me to explain what I mean but here goes...
I agree that they seem to in some instances. For instance in:
http://www.mail-archive.com/[EMAIL PROTECTED]/msg09243.html
in a reply to my question why fn isn't enough, why we always need to
wrap it up with vcard, Andy argued:
"the classes "fn" and/or "n" might already be used, with different
(or no) semantic meaning" [than the "fn" used under a vcard]
I replied in:
http://www.mail-archive.com/[EMAIL PROTECTED]/msg09248.html
"what I'm reading here is that classname "fn" may have different meaning
if used outside of an element of class "vcard". Saying this is to me
equivalent to saying the "vcard" classname syntax is syntactic sugar for
the concept of a namespace (as is "vcard-fn" or "vcard:fn"). My
understanding was that the concept of namespace, not just its xml
syntax, was an antipattern in microformats. Am I mistaken?"
The discussion ended here unfortunately.
AFAIC, I think that class names should not have context-sensitive
meaning (i.e. different meaning depending on why class wrap them),
except if the parent meaning precise its meaning. In other words, "fn"
should always be used for "formatted names", but if "fn" is used for a
song's name as it is today, then it may be inferred, that fn is not just
a name, but the title of the song, since "title" is the word of choice
for a work of art's name (see http://www.answers.com/title&r=67).
Namespaces allows the same word to mean different things under different
NS. I think this is bad and my understanding is that is what
microformats are trying to avoid, not just the XML ns syntax.
On the other hand, allowing more precise meaning to be inferred from a
class name and the context where it appears makes sense to me.
Guillaume
_______________________________________________
microformats-new mailing list
[email protected]
http://microformats.org/mailman/listinfo/microformats-new
