[uf-discuss] hAtom - ready for 0.1?

[Chris Casciano]

On Feb 28, 2006, at 5:53 AM, David Janes -- BlogMatrix wrote:

> Chris Casciano wrote:
>> My own expectation is not that it contains copyright statements, 
>> licensing information and links to feed information in addition to 
>> contact information wrt. page authorship. But then I look at Tantek's 
>> implementation on his blog and it works as a very tight address 
>> inside a list of other page related information. Works there, but 
>> where else? What if the presentable information is more then just one 
>> data item? I'm using a really silly and informal line on 
>> ChunkySoup.net that is a case of it not translating well to this 
>> usage:
>
> The uF community is taking ADDRESS to mean something semantically "in 
> the spirit of the HTML 4.01 spec [1]:
>
> | The ADDRESS element may be used by authors to supply contact
> | information for a document or a major part of a document such
> | as a form.
>
> Regards, etc...
> David
>
> [1] http://www.w3.org/TR/REC-html40/struct/global.html#edef-ADDRESS
>

Right. That's the same definition that is used in 401 which I had 
linked to in one of my replies. But given it means something equivalent 
to contact information and not "licensing information and contact 
information" or "a bunch of random stuff in a footer" it just doesn't 
map well to the real web and thus is rarely used.

So from where I sit its either rewrite stuff, redesign things to 
include information that isn't there (or is currently linked, abuse 
ADDRESS for sake of conforming to hatom over html or just decide a page 
content just doesn't map well to Atom or that I otherwise can't be 
bothered to conform to hatom in a way I'm happy with.

Again, the rest of the spec seems to map real well to real world usage 
-- outside of a few occcations where there aren't dates that map well 
to individual entries (either no date, or one date heading for multiple 
posts). But authorship is something that is VERY regularly defined 
outside of the individual entry both on the 1 man personal context and 
the anonymous corporation side* and the solution provided to defining 
the required data just doesn't seem very that practical to me.

Not much more to really say on it so I'll end my participation in this 
discussion there and hope that with time enough other people will run 
into this wall with hatom (or maybe hreview or some other µf that needs 
to pair page data with entry data).

-- 
[ Chris Casciano ]
[ chris at placenamehere.com ] [ http://placenamehere.com ]