[uf-discuss] Citation: next steps?

Michael McCracken michael.mccracken at gmail.com
Thu Sep 21 18:09:58 PDT 2006

Back to the citation fray:

I think the ROLE attribute of vcard means something different than
what you guys were describing with a role.

You were saying role = "what this person's relationship to the cited work is"

I dug around for the vcard rfc, which says that the ROLE vcard
attribute is based on job description, so for the hcard, role = "what
this person does for a living".
It is also supposedly intended to draw from a list of roles described
as business category or occupation, in (ANSI?) X.509, which I can't
find online.

I can easily imagine wanting the vcard role in addition to the
citation role - for instance, on my CV page I want to have a single
hcard for myself with the role "graduate student", while my role on
each paper I cite in my references list may be different.

Unfortunately I don't have an answer for how I think we should mark up roles.
I agree that having a creator,role,value structure is nice, but I
can't think of a good way to mark it up.
There's nothing I know of that's marking things up like that out there
on the web, and unless it's a coincidence, all discussion on this
topic stopped once we got onto the role idea. I stopped because I was
waiting for an idea to come to me about how to write that in XHTML.

The only ideas I have come up with involve hiding the span containing
the role with CSS. If we just have a single creator class (and a small
number of other role classes), we can do this:
<span class="vcard creator"> /*my vcard*/</span>

with roles, we could do this:
<span class="vcard"><span class="role">author</span>/*my vcard info*/</span>
but that clashes with the vcard 'role' attribute, which may be OK but may not.
If we are parsing this, do we have to treat hcards that are in a
citation differently from hcards elsewhere? (In order to avoid losing
the 'role' element to the hcard, or to first look for a 'citerole'
element or something before passing it off to an hcard parser?)
I'd like to avoid that extra complexity.

Does anyone have a good suggestion for marking this up?

I'm not sure if I'm missing an obvious good solution here.


On 9/1/06, Bruce D'Arcus <bdarcus.lists at gmail.com> wrote:
> On 8/31/06, Timothy Gambell <timothy.gambell at aya.yale.edu> wrote:
> [...]
> > > hCard has a role term, though I don't know if it is consistent with
> > > this?
> >
> > Certainly an appealing possibility. Unless the proprietors of hCard
> > object, I think we should use it. Do you agree?
> Well, the problem with role to me is the semantics are unclear (a role
> isn't really a property of a person, but a relation between a person
> and some other thing). But I really have no strong opinion.
> > > It is; really more a "producer". The DC group considers it a
> > > contributor, and has wanted to get rid of dc:publisher and use that
> > > instead.
> >
> > Dropping publisher and marking it up as a contributor with a role of
> > publisher sounds like a good proposal to me.
> I'm not saying to drop it really; just giving an example of how to
> think about it.
> Bruce
> _______________________________________________
> microformats-discuss mailing list
> microformats-discuss at microformats.org
> http://microformats.org/mailman/listinfo/microformats-discuss

Michael McCracken
UCSD CSE PhD Candidate
research: http://www.cse.ucsd.edu/~mmccrack/
misc: http://michael-mccracken.net/wp/

More information about the microformats-discuss mailing list