[uf-discuss] Microformats in Google Maps

Andy Mabbett andy at pigsonthewing.org.uk
Thu Aug 2 01:14:50 PDT 2007


In message <002401c7d4a6$57078a00$116bacca at COMCEN>, Michael MD
<mdagn at spraci.com> writes

>> So it appears that none of the address sub-elements are being classified
>> at all, just simply poured into the adr td and broken up with br's.
>>
>> To be parsed they should be wrapped in classified elements like the
>
>
>I'm not surprised at all to see this kind of thing out there.
>
>There are a lot of cases where people may have freeform text addresses
>and wish to mark them up somehow...  - especially where there is a lot
>of user-entered or legacy data (Google would have plenty of both!)
>
>There are no reliable automated means of splitting such freeform text
>into the seperate elements required by hcard and to expect everyone to
>manually re-enter everything is definately asking too much!
>
>Freeform text addresses might not be good for conversion to hcard/vcard
>but can still be very useful to humans!
>- it seems obvious to me that people out there (regardless of what
>anyone says) will try to stick them in adr somehow!

I addressed this in:

 <http://microformats.org/wiki/hcard-brainstorming#ADR_with_no_children>

but people (Brian Suda, chiefly) insisted that my suggestion should not
be followed, and that such addresses should instead be wrapped with
class="label" - which is supposedly for:

        "formatted text corresponding to delivery address"

Note the use of "formatted".

More recently, Tantek added:

  <http://microformats.org/wiki/hcard-brainstorming#implied_adr_subproperties>

which strikes me as unworkable, being overly complex and not suitable
for internationalisation (not just in non-English speaking countries,
but outside the USA)

-- 
Andy Mabbett


More information about the microformats-discuss mailing list