[uf-discuss] CFP microformat?

[Ryan King]

On May 30, 2006, at 5:49 AM, Angus McIntyre wrote:
> At 21:45 +0200 20.05.2006, Ryan King wrote:
>> On May 20, 2006, at 9:13 PM, Tantek Çelik wrote:
>>>  A good first step might be to see if you can get these  
>>> conferences just
>>>  using hCalendar to start with ...
>> I want to reemphasize this....
>>
>> The use case you described sounds like a specialized case of  
>> "events + todos"
>> which sounds like exactly hCalendar. If hCalendar isn't  
>> sufficient, the
>> only way to know reliably is try it out first.
>
> There can be calls for papers which don't have such an obviously  
> 'eventy' nature, i.e. calls for journal articles. There's certainly  
> a due date in almost all cases, but other attributes may be very  
> specific to a CFP rather than an hCalendar item. Keywords - tags -  
> is one obvious one, and paper length is another. I offer as an  
> example the call for journal articles at:
>
>     http://www.researchforsexwork.org/target/calls/r4sw09.html
>
> whose editor has been complaining to me that her contributors are  
> apparently incapable of reading the part that says "maximum number  
> of words is ..."

Doesn't surprise me. I'm not sure if we can solve this problem with  
microformats, thought. :D

> This example includes:
>
> 	- journal title ("Research for Sex Work")
> 	- journal instance title ("Sex Work and Money")
> 	- due date (15 Dec 2005)
> 	- paper length (1200 words)
> 	- acceptable languages for submissions (English, French, Chinese ...)
> 	- contact address (an obvious hCard candidate)
> 	- suggested topics (which are more than just tags)
>
> Many CFPs will have multiple due dates - the due date for  
> submission of an abstract, and the due date for submission of the  
> final article. In some cases there may even be a due date for  
> submission of the camera-ready copy of accepted articles.

These can be encoded as multiple todo's (they are separate tasks, no?).

> hCFP starts to look like a candidate for a complex microformat that  
> contains an hCard, plus hCalendar entries for due dates, plus  
> perhaps a microformat representation of a conference, book or  
> journal (which may have hCalendar and hCard entries themselves),  
> plus some CFP-specific information like paper length and submission  
> languages.
>
> This might be in 20% territory, but in other ways it's quite a  
> natural application of microformats and the payoff - automated  
> identification of CFPs - is worthwhile.

I'm not going to say that a microformats for CFP's is unnecessary.  
Nor am I going to predict that it will never exist. Nor will I say  
that it wouldn't be useful.

However, it does smell 20%-ish. It also seems that most of the  
building blocks are there already. And, because of these two items, I  
think that those interested in applying microformats to this problem  
should experiment as much as they can with existing microformats. I  
know this won't cover every bit of interesting data, but I think it  
can get folks quite a ways towards having some useful tools  
(remember, 100% is never a goal around here :D).

I know that Michael McCraken has experimented a bit with this, I'll  
get to that later (I'm in catch-up mode after spending 2.5 weeks on  
the road).

thanks,
ryan