[uf-discuss] Microformats, WebApps 1.0 and UI widgets in browsers

[Karl Dubost]

Just for the record,
http://ejohn.org/blog/microformats-in-firefox-3/

Open to comments and ideas.
    3.
       Karl Dubost, W3C Said,

       February 1, 2007 @ 2:21 am

       Mike,

       is there a place where you collect the comments, feedback, etc?
       First times are very important and lead to cool things, great  
improvements and painful ties sometimes. So indeed when a browser is  
proposing new UI widgets directly related to the *semantics* of  
content, we have to be very careful.

       At first, I say “cool, very cool!”. Then, taking a step back,  
I think what about the documents which have been created for the last  
15 years before microformats effort existed. These documents contain  
class names which are probably and most certainly very similar to  
some values defined by microformats community. So there will be  
documents where a UI widget will be activated but not with the  
intended meaning. Basically it is changing the contract between the  
author and the reader by hijacking the intended semantics.

       There /was/ a solution for this profile attribute URIs with  
the URI of the used profile. Problem ahead it seems that some  
developers want to suppress this attribute in HTML document.

       I think there is a possible win-win here. The Mozilla UI  
widget could be activated only when the right URI (profile attribute)  
is really here. (a bit like the doctype switching). It will encourage  
people to use the right URIs, because the effect would be immediate,  
it will not hijack documents previously written. Everyone win.


PS: Another switching mechanism could be used as well if appropriate.





-- 
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager, QA Activity Lead
   QA Weblog - http://www.w3.org/QA/
      *** Be Strict To Be Cool ***