[uf-new] Re: hAudio FN or Title

Martin McEvoy info at weborganics.co.uk
Sat Feb 2 03:11:41 PST 2008


On Fri, 2008-02-01 at 17:33 -0800, Guillaume Lebleu wrote: 
> Martin McEvoy wrote:
> > On Fri, 2008-02-01 at 22:09 +0000, Andy Mabbett wrote:
> >   
> >> The more I consider this, the more I am convinced that class names 
> >> should not be shared between microformats. 
> >>     
> >
> > For what its worth I think you may be right for example "fn" was only
> > used in hcard this meant just the name of a person or organization,
> > great stuff powerful and clearly defined. hAudio and hreview re-use "fn"
> > to mean other things too, haudio it means, a title of a playlist, album,
> > or track and the name of a contributor or organization (makes your head
> > spin trying to explain that to people), in hreview it is the name of the
> > thing that is under review and the name of the reviewer, kind of
> > devalues the semantics of "fn" don't you think? instead of being
> > something specific its turned into something more general. So yes in
> > light of the above maybe we shouldn't re-use class names so willingly,
> > most of the community that spend a lot of time working these things out,
> > the only time we should is if what we are trying to describe has the
> > exact same meaning, only then would it make sense to re-use.
> >   
> 
> Wouldn't you say that this is equivalent to saying that what is really 
> important is that the meaning of class names be context-insensitive, 
> i.e. a class name like fn's meaning should not change when used within 
> an haudio or an hcard ?

correct I think :)  

> 
> By the way, about "fn", as you know, it stands for "formatted name", 
> i.e. how the name is displayed, which is really a useful distinction to 
> make when:
> 
>     * on one hand the name exists in other representations not intended
>       for display, i.e. "structured" "tokenized"representations useful
>       for indexing/searching in particular, where honorific title, given
>       name, family name, etc. are distinguished,
>     * and on the other hand, the actual representation used for display
>       (meaning really for serial representation, i.e. printed on a
>       letter) must be kept handy b/c the formatting rules vary widely
>       from locale to locale, and finding them out, representing in code,
>       and running them each time you want to print does not make sense.
> 
> Assuming this more precise meaning, I don't think it applies to a song 
> name, which does not have to my knowledge a tokenized form.

No it doesn't, not really, suppose we take this:

<span class="fn">Start Wearing Purple</span>

If we were to assume the precise meaning, in context or hcard the above
example would be a first name "given-name", a secondary name
"additional-name" and a last name "family-name" applications that may
index the above example may take it to mean precisely that! which is of
course incorrect. A thing I may add is, is the above example the
formatted name of and album, a track, a person or an organization? and
is "fn" appearing in the correct context? hcard has the ability to
expand the meaning of "fn" haudio does not. Do you not think that
context is an important factor in microformats?
 
you could use a bit of posh instead an example may be:

<cite>Start Wearing Purple</cite>

and not even trouble ourselves with the meaning of "fn", of course our
haudio title does mean a little more than just a cited source or
reference but the semantics are correct <cite></cite> is not appearing
out of context as "fn" is currently in haudio.

>  (BTW, 
> assuming this meaning, "fn" may not even be always appropriate even for 
> all displayable by nature HTML-based representations of a person's name: 
> take the particular case of an HTML table where a person's name is 
> typically in a tokenized representation, which may not correspond to the 
> localized serial representation a.k.a. "formatted name")
> 
> Guillaume

Thanks

Martin McEvoy
> _______________________________________________
> microformats-new mailing list
> microformats-new at microformats.org
> http://microformats.org/mailman/listinfo/microformats-new



More information about the microformats-new mailing list