[microformats-discuss] Re: Educationg Others

[Dr. Ernie Prabhakar]

Hi Scott,

On Oct 3, 2005, at 3:15 PM, Scott Anderson wrote:
> Basically, I don't want to have to support two different XML formats
> that describe the same data. I am being told that the XHTML used in
> microformats is indeed portable. Has this been demonstrated to be
> true? I have a lot of doubts that the context is portable or the
> support requirements are trivial.
>
> If I do need to support a secondary format that gets used to generate
> microformats in my XHTML does anyone know of a good candidate for this
> besides RDF?

I'm not sure we're even talking about the same problem.  By "on the  
web" we here (I think) all mean 'the human-readable web'.  If your  
web is the "semi-private, machine-only web of automated web  
services", then we're pretty much living in different universes.

For example, If you are looking at XML as a way to, say:
	- replace your existing SQL database as an internal datastore; or
	- define a controlled vocabulary for machine-readable apps in a  
vertical industry with good governance
then by all means, use a custom XML schema. That's not the problem  
we're trying to solve here.  We don't want a zillion microformats,  
one for every vertical industry imaginable.

However, if you are planning to do anything involving the *mass  
market* web, where HTML is *already* part of your workflow, then I  
would encourage you to look at "piggybacking" microformat semantics  
on top of XHTML rather than publishing a *second*, distinct, XML  
output. That way, it allows anonymous consumers the chance to extract  
semantic data directly from your existing web pages, rather than  
having to negotiate a specific XML format.

That's all we're saying. I think. XHTML is a good format for:
	a) externally-publishing data you (also) want humans to read
	b) storing data that is primarily document-centric and thus maps  
nicely onto HTML's native semantics

You'd be surprised how many people try to use XML for *that*, which I  
suspect  why Tantek and others react so strongly to what they call  
"XML zealots."  But, IMHO, if you are truly doing something for which  
XHTML itself has zero value -- and a custom XML solves both your  
technical and social problems -- then please, go for it!

Best,
-- Ernie P.