[uf-discuss] Canonical hCards (was: Search on CSS element)
chris at placenamehere.com
Wed Jan 24 14:26:38 PST 2007
On Jan 23, 2007, at 1:22 PM, David Janes wrote:
> Continuing the tradition of riffing off other threads to talk about
> what's on my mind...
> On 1/23/07, Ryan King <ryan at technorati.com> wrote:
>> It's not a full CSS selector based search engine, but http://
> Wouldn't it be great if there was a well defined way of getting from
> _an_ instance of a hCard to the _best_ (or canonical) hCard for that
> Just saying; I know that some work has been done on this.
> Regards, etc...
Not to double back on a few days worth of replies, but while I
understand the desire to eliminate noise from search results [or
elevate more complete results] I'm not sure how I feel about two things:
(a) that this isn't an issue simply to be solved by consuming
applications. E.g. "search for detailed contact data" "return only
results with email address" "match on Name 'Chris' and url contains
'placenamehere.com', or some non-hcard solutions be them XFN,
claimID, other avenues for tying these bits together.
(b) I think with any solution we need to be careful on the definition
and reliance on these flags, and more generally the notion that
there /is/ a best out there for any given individual. Just one case
would be a person who might have their contact info on a business
context, a context related to their position in a different
organization, and then yet another in a more personal context [blog,
resume on monster.com, etc]. While all are probably more informative
then a simple fn-only card from a reply on a random blog post on the
net, what is 'best' is based on context and the data will most likely
not be consolidated into one place.
P.S. the technorati search is currently giving me no results for any
contact searches I try, I'll follow up with the proper parties offlist.
[ Chris Casciano ]
[ chris at placenamehere.com ] [ http://placenamehere.com ]
More information about the microformats-discuss