[uf-discuss] UID, URL, live microformats (was: Microformat auto-discovery WAS: Plazes & Microformats)

[Benjamin Carlyle]

On Wed, 2006-04-19 at 09:39 -0700, Tantek Çelik wrote:
> On the Web, many (most?) contacts and events also have a canonical or
> authoritative URL.
> In addition to marking up the authoritative/canonical URL for a
> contact/event with class name of URL, why not also use that URL for the UID?
...
> Given that it could make sense to provide updates to any type of microformat
> data (as Brian hints at), I want to strongly consider reusing "uid" from
> hCard and hCalendar as an elemental microformat which can be used with *any*
> microformat in combination with class name "url" to indicate the specific
> instance, as well as where to get updates.

The thing that comes to mind for me is about how this would interact
with hAtom, especially hAtom's use of rel-bookmark for a permalink (and
uid). The concepts appear to be related, and I suspect they are
effectively the same. Atom itself permits the specification of a <id>
element separate to <link rel="self">, however hAtom constructs <id>
from either a <a rel="bookmark"> or from the "id" attribute of the entry
and the page url.

A couple of things:
* Is rel-bookmark still appropriate for hAtom in light of this potential
new format?
* Is rel-bookmark even valid?
(yes, I know it's part of the html standard... but I still think it's a
somewhat resonable question)

The validity question comes to mind with Ryan's discussion a few posts
back about the possible rel-vcard concept. He said that A is a vcard of
B doesn't make sense. A is a vcard regardless of the existence of B. The
use of the url class in hcard and hcalendar is also a data point.

It doesn't really seem appropriate to say that A is a bookmark of B...
only that the url of A is should be used in place of the url of B when
bookmarking. If you use the hcard url reasoning, we would say it should
use the class of the <a> element rather than the rel of the <a> element.
After all, it isn't the remote page we are talking about.. it is the url
to the remote page.

Whichever way you fall, either on the rel-bookmark or the "url uid" side
of the fence... it seems reasonable that the hAtom and other formats are
consistent in this where possible. Deployed base is always a
consideration in such questions, but it is probably worth exploring the
thought experiment: "What would be our ideal outcome?". It is then
appropriate to bring things back to earth with "how close can we get to
the ideal in practice"?

Benjamin.