[uf-discuss] hCite progress

[Brian Suda]

On 11/13/06, Chris Messina <chris.messina at gmail.com> wrote:
> In terms of "type"... How important is that designation? If you have
> an open-ended list, isn't that similar to a tag or is there real
> consequence to its type-designation?

Well, TYPE seems to be pretty important, it is manditory by (atleast)
RIS and BibTeX so far. Both of those have enumerated lists of possible
values, which there duplicate values (e.g. Book, Thesis, etc.) but
they use different terms (e.g. "mastersthesis" or "THES") so what gets
used for the hCite type? or do we create our own list that maps to the
80% of common types.

We could use tags, but then we are still picking out the tag portion
as the TYPE value. You could do that already now with "Keywords" in
hCite and "Skills" in hResume and "Categories" in hCard. And the value
that gets extracted would need to still have to match to some sort of
logical citation type.

I ate a <a href="/tags/watermellon" rel="tag"
class="type">watermellon</a> yesterday.

I'm not sure how that helps us?

whereas:
This is a <a href="/tags/book" rel="tag" class="type">book</a> i read
yesterday.

That helps on two fronts, #1 we have an established type (book) and #2
we get bonus points for making it a tag so now we can search on all
books! That's great for a blog, but that tag space on Amazon wouldn't
really narrow things down much :)

-brian

-- 
brian suda
http://suda.co.uk