[uf-discuss] Microformat Encoding Thoughts

Brian Suda brian.suda at gmail.com
Fri Nov 4 06:07:02 PST 2005


> ...
>  Thus instead we
> found it made much more sense to have the type of phone number be visible information, and simply use class="type" to mark it up.  This method also reduced the size of the class name vocabulary considerably.

While this is true that because of the second-term hypen some data
gets hidden, but i think it could actually reduce the number of
properties. In the XMDP you would specify something like author-* were
the -* is optional instead of the enumerated list of author-XXXXX.
Then if your transforming application knows what to do with the -*
then it acta accordingly, otherwise it fallsback to the author. So if
i have the following example:
<h2>Document Authors</h2>
<ul>
<li class="author">John Doe 1</li>
<li class="author-supervisor">John Doe 2</li>
<li class="author-editor">John Doe 3</li>
</ul>

If the transforming application was converting this to a citation
format that did NOT have and editor property that name might get lost
because there is no mapping. But with the hypen system, it would
simply get converted into another 'author' because the '-editor'
portion is unknown and therefore defaults to the first portion.

This system could make the microformat more flexible in that it
defines the base types (author, title, etc) and then lets you extend
them as needed (author-secondary, title-alternative). Dublin Core does
something similar. Abstract is just a specialized version of
Description.

I'm not saying "this is how citation microformats will work, so learn
it!" i'm just floating the idea to get comments.

Unlike other microformats, the citation microformat is not a 1:1
correspondance with another format, it will be a sort of Bablefish for
all the formats. You take an XHTML document and it can (hopefully) be
translated into several different citation formats. In doing so, there
will be some properties that do NOT translate between citation formats
and therefore get left behind in the shuffle. (the hypen system was an
attempt to prevent this)

The hresume could benefit from the inclusion of citations as well
(these are papers/books/etc i've written...), so keep me in the loop
of what you are discussing so we can combine our efforts.

-brian

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


More information about the microformats-discuss mailing list