[uf-new] Namespace anti-pattern and hAudio TITLE (was: hAudio FN or Title)

Mon Feb 4 15:05:30 PST 2008

On 2/4/08 1:25 PM, "Martin McEvoy" <info at weborganics.co.uk> wrote:

> Hello Manu
> 
> On Mon, 2008-02-04 at 15:17 -0500, Manu Sporny wrote:
>> Then, I assert that your definition of what is and isn't a namespace
>> is out of step with the common definition of a namespace
> 
> If "namespaces" did exist in microformats? it would make it impossible
> to re-use other objects such as "title" in other microformats because
> names in namespaces can only be declared once and only have one
> contextual meaning?
> 
> No one actually will admit to the existence of a namespace in
> microformats but there is lots of evidence suggesting otherwise.
>
> either intentional or not, Microformats MAY have created their own
> "namespace" of a kind I think?

The problem is with this loose use of term "namespace" or "namespace of a
kind", not with microformats usage thereof which will result in endless
semantic arguments which are not useful.

Frankly, this entire thread is one of the reasons we have this mailing list
guideline:

http://microformats.org/wiki/mailing-lists#Bad_topics_for_discussion

> Microformats declare a formal profile[1] in their proposals[2]
> 
> [1] http://gmpg.org/xmdp/
> [2] http://microformats.org/wiki/haudio#XMDP_Profile
> 
> and then do this...
> 
> <head profile="http://gmpg.org/xfn/11">
> 
> very GRDDL Like
> 
> <head profile="http://www.w3.org/2003/g/data-view">

The uses are quite different actually.  XMDP uses the profile attribute for
defining terms, and GRDDL uses the profile attribute for defining how to
transform.

> Maybe someone can explain to me why declare a profile if there isn't
> some kind of metadata description at the end?
> 
> http://gmpg.org/xfn/11

XMDP provides a simple description mostly just to define the terms.  So far
we've resisted additional metadata similar to DTDs or XML Schema docs.

Another big difference is the way XMDP overrides work, linked from [1] cited
previously above:

http://gmpg.org/xmdp/description

Note that if two different profiles define the same term, only the first
definition is used.  Thus they are not "namespaces" in any data/programming
functional way (like XML namespaces), as such namespaces make such multiple
references exist simultaneously and distinctly by keeping them in their own
silos.

I have added both of these to the XMDP FAQ:
http://microformats.org/wiki/xmdp-faq

Tantek