[uf-discuss] Comments from IBM/Lotus rep about Microformats

[Mike Schinkel]

> Forgive me if I have missed something, but could a parser not 
> understand multiple formats if the HTML used was also 
> meaningful?  For example a blocklevel element (say a <p>) 
> could contain some content that was marked up with one 
> microformat, and another blocklevel element could contain 
> content marked up with another entirely different 
> microformat.  The fact that they shared the same page isn't a 
> problem.  A parser could easily identify child relationships 
> within the HTML and extrapolate.

A.) Possible, but that would require putting lots of intelligence into
parsers, which is generally considered A Bad Thing(tm) w.r.t. web
architecture goals.

B.) But if the two microformats markup similar content then probably not
possible.

> Granted this wouldn't be so easy if two microformats were 
> muddled together on the same page.  And if they were then 
> maybe there are two questions to ask.  1/ Is the microformat 
> in need of some additional elements?, and 2/ Is the author of 
> the page trying to do too much.
> could it be laid out differently?

The biggest problem is when two (lowercase) microformats get developed by
authors having no knowledge of the other, and then each microformats gets
adopted by different constituents. Finally, the content of a website is a
crossover for both constiuents creating a conflict between the microformats.

For example, lets say that two microformats one for business news and the
other for the car industry were independently created at roughly the same
time. Both includie markup for identifying a company; the subject of a news
story and the manufacturer of automobiles, respectively.  Since they are
both trying to mark up the same data, there's a good chance they will use
some of the same terminology but they may use it in conflicting ways.  

Now let's assume a site that provides auto news comes along, and ideally
needs to use both the news and the car industry microformats. Unfortuantely
they will have to choose only one because the two microformats conflict.  I
view creating an architecture that is likely to creat conflicts like this as
A Very Bad Thing(tm), especially when it would be so easy to resolve at this
point in Microformats' evolution.

-Mike Schinkel
http://www.mikeschinkel.com/blogs/
http://www.welldesignedurls.org/