[uf-new] Revisiting grouping problem solution proposal: hset

Manu Sporny msporny at digitalbazaar.com
Tue May 22 09:39:21 PDT 2007


Just to clarify a few of Martin's points...

Martin McEvoy wrote:
> The current hset proposal would say
> 
> class="hset with-client-event.FOO_Sales_Pitch"

This should be (note the replacement of a ' ' with a '.'):

class="hset.with-client-event.FOO_Sales_Pitch"

> the whole class declares that its a grouping, a client event and its
> sales pitch about foo, 
> * it doesn't need to be hierarchically grouped or even on the same page

In an attempt to not create confusion - hset identifiers are only
relevant on the same page. If you had one page that contained an hset
named 'hset.foo.bar' and another page that contained an hset named
'hset.foo.baz' - per hSet there is no relationship that you could infer
from the similar names.

However, if both of those identifiers were retrieved from the same page
- you could infer a relationship between the two elements.

> * It can be sparsely grouped 

Martin, unless I interpreted Chris' problem solution proposal
incorrectly - his proposal does support sparse grouping.

> It relies on the child to be grouped together with the parent for it to
> have any meaning.

So does hSet... the only difference is that in the case of hset may be
done out-of-order. Chris' proposal must be done in-order. For example,
this is valid in hset:

hset.foo.bar.fnurt.boing
hset.foo.bar.fnurt
hset.foo

With Chris' proposal, the parser wouldn't be able to infer any
relationships until the entire page had been parsed. Although, that's
really not that big of a deal...

> If you had Many ID's on a single page all attaching their special
> meaning, do you think this may cause some problems for browsers or
> phraser? 

It really shouldn't - browsers and parsers should be able to handle
thousands of IDs per page.

Chris' proposal accomplishes the same things that option #3 accomplishes
 - it solves the grouping problem. Really, I think this has turned into
a discussion on whether we go with Option #7 (Chris' proposal) or Option
#3. Arguments for going with Option #3 have been provided... input from
the community on which one to choose would be great.

-- manu



More information about the microformats-new mailing list