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

[Scott Reynen]

On Dec 5, 2006, at 10:48 AM, S. Sriram wrote:

> From: "Mike Schinkel" <mikeschinkel at gmail.com>
>
>> http://listserver.dreamhost.com/pipermail/whatwg-whatwg.org/2006- 
>> December/00
>> 8462.html
>>
>> I wonder if his issues can be addressed?

Ian said:

> "class", "rel", and "profile" are the extension mechanism for HTML

And Elias said:

> We have tried using microformats as an extension mechanism for HTML

That response confuses the issues here.  Ian pointed out the only  
allowed methods of extending HTML and said that microformats use  
these methods.  That Elias is unsatisfied with these methods is a  
problem with HTML, not with microformats.  We're not in charge of  
HTML here.

> How about a distributed parser-discovery service
>
> More specifically a YADIS discovered JSON returning uf-specific  
> parser:
>
> 1. Place an entry on the uf-authored page detailing the ufs-used
> <meta name="ufs-used" content="hreveiew hatom hwidget" />

As Ian mentioned in the message Elias responded to, HTML already has  
this functionality with profile headers:

http://microformats.org/wiki/profile-uris

> 2. Place a yadiservices discovery pointer to where parser(s)
> maybe found,  (on the same uf-authored page)
> <meta name="ufs-used" content="hreveiew hatom hwidget" />
> <meta http-equiv="X-YADIS-Location" content="http:// 
> www.blahblah.com/path/to/yadis-file" />
>
> 3. add parser service data to the (existing) yadis file pointed to  
> within
> the uf-authored page.
> <?xml version="1.0" encoding="UTF-8"?>
> <xrds:XRDS xmlns:xrds="xri://$xrds" xmlns="xri://$xrd*($v*2.0)"><XRD>
>    <Service>
>        <Type>http://openid.net/signon/1.0</Type>
>        <URI>http://www.livejournal.com/openid/server.bml</URI>
>    </Service>
>    <Service>
>        <Type>http://microformats.org/hreview/1.0</Type>
>        <URI>http://www.blah.com/path/to/uf2json-parser</URI>
>    </Service>
>    <Service>
>        <Type>http://mysite.com/hwidget/1.0</Type>
>        <URI>http://www.mysite.com/path/to/uf2json-parser</URI>
>    </Service>
> </XRD></xrds:XRDS>
>
> 4. ..domain../path/to/uf2json-parser is a REST-call that is passed
> a 'uf-snippet' and returns a JSON object.
> Browsers that are uf-aware would call the parser with the uf-snippet,
> and than hand of the JSON to the storing module.
>
> CONS: The parser needs to be 'hosted', incurring bandwidth costs.
> PROS: Roll your own microformat and parser - or - *leave your html
> as is and just build a parser for it and point tothe parser from  
> within the page.*

What you've described above is a process for converting all  
microformats to JSON, but that doesn't really solve the problem Elias  
described.  It just changes the format.  Each individual parser still  
needs to figure out what the JSON means, where before they had to  
figure out what the HTML means.

In HTML or JSON, new formats need new parsers, which must be written  
by someone.  Elias is coming from an RDF background, and microformats  
simply aren't RDF, and they never will be.  And that's a good thing.   
If what you want is RDF, just use RDF.

Peace,
Scott