[microformats-discuss] Definition of "microformat"?

[Tantek Ç elik]

Hi Bob,

Welcome to the microformats-discuss list!

On 9/15/05 3:21 PM, "DuCharme, Bob (LNG-CHO)" <bob.ducharme at lexisnexis.com>
wrote:

>> It seems that the distinguishing feature of microformats is the way they're
>> created, not what they look like when its all done.
> 
> This implies that, given two small, well-formed XML documents, you couldn't
> look at them and say "this one is a well-done microformat, and the other one
> isn't" without knowing how their authors went about creating them. I'm sure
> that this isn't what you meant.

There are a few distinctions to be made here.  For starters:

1. data vs. spec/format.  You're asking about comparing two XML *documents*
and then asking if one of them is a "well-done microformat".  That's not
quite an appropos comparison.  Comparing two format specifications and
asking if one of them is a "well-done microformat" would make more sense.

2. creating content vs. creating specifications.  These two are very
different.  Authors are creating content that uses microformats in all sorts
of ways, hand coding, tools, automatically etc.  OTOH, authors of
microformat *specifications* are creating them quite diligently, by
proposing problems spaces, doing background research on existing publishing
practices, and existing interoperably implemented specifications,
brainstorming, and finally making proposals and authoring a microformat
specification.


> I think that, while looking at two such documents, a review of the
> microformats principles would give some clue as whether one works as a
> microformat better than the other.

See above.

Often you may not even realize that a document is a using a microformat
unless you look very carefully at it.  This is one of the "micro" aspects of
microformats, they often involve very little change to existing content.
rel-tag is a good example, where literally simply adding rel="tag" to a
hyperlink to a tag page turns that link into a "tag" for that piece of
content.

> Like you said, if there's no good
> definition of what a microformat is, then at least the design principles
> give some guidance about set membership.

Ryan's right that we are still figuring out some of the details, and thus
there is no precise mathematical equation for deriving a microformat.

And yes, your observations are right on about using the design principles to
determine set membership, or how much a particular format is a microformat.
 
In particular, somewhat as an exploratory exercise, I applied the principles
as an analysis tool to "RSS" as a format and came up with some interesting
conclusions.

<http://microformats.org/discuss/mail/microformats-discuss/2005-August/00059
8.html>

I hope that helps.

Thanks,

Tantek