[uf-discuss] citation microformat?
tjameswhite at yahoo.com
Wed Jan 18 19:00:52 PST 2006
I've been (slowly) working on the bibliographic citation format for a
while now. Brian Suda and I had a lengthy discussion a while back, and
Edward V. and I have had some preliminary discussions.
--- Tantek ï¿½elik <tantek at cs.stanford.edu> wrote:
> In addition, as way of "moving forward", it would be great to see
> analysis of the implied schemas illustrated by the examples in the
> cite-examples page:
After I added those examples to the wiki, I did a quick analysis --
in a nutshell patterns for titles include:
* id="title" class="producttitle"
* id="lblTitle" class="book_headline block"
* table cell, no mark up
I also posted my thoughts on starting NOT with a full bibliographic
microformat, since that tends to get overly complicated quickly, but
rather to start with a title design pattern. My thoughts were lengthy,
so instead of sending them to the list, I linked to them on my
blog. Perhaps that resulted in less discussion.
At any rate, my concern now is with the title issue. I see that under
hAtom it was determined that class="title" couldn't be reused because
it was already in use by hCard.
While I understand the idea behind not reusing names, this practice
seems to have 2 problems: 1) long-term complexity, 2) runs counter to
Keeping things simple by not reusing names has its merit. However, as
we build more microformats I think we run the risk of needing to invent
non-intuitive class names because the most appropriate one is already
taken. It looks like class="title" for hAtom was solvable by using an
alternate name, but now that is taken away from the next format.
As for the principles, we are suppose to "pave the cowpaths". The few
examples I've placed in the wiki break out to about 60/40 that use
"title" (or some derivative) to refer to a book title. Furthermore,
look at virtually any card catalog, amazon.com or other book service
and you will find search criteria for "title". Also, look at the work
Edward has done for the Ann Arbor District Library XML feed.
I would say that this path is pretty well worn. (And yes, I admit this
blinds/biases me a bit to seeing alternatives, but I'm trying to be
open to reasonable suggestions.)
As for any confusion re-used names may cause parsers/application, I
think that is secondary to easy of use by people. At least when I think
about People First, Machines Second, I think in terms not only of human
readable, but also easy-of-use by people. Computers are here to do the
complicated bits for us and should be able to figure out if they are
reading an hCard or a bibliographic citation. (Easy for me to say since
I don't write apps.)
I hope all this makes sense -- it's getting late and I feel like I'm
rambling. At any rate, I hope we can have some good, productive
discussions about a bibliographic format and get this thing moving.
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
More information about the microformats-discuss