[uf-new] Proposed Microformats: hRebuttal, hEvidence, hSource, hConclusion and hArgument

[Colin Barrett]

On May 25, 2007, at 3:40 AM, Danny Ayers wrote:

> On 24/05/07, Colin Barrett <timber at lava.net> wrote:
>
> Microformats are about paving the
>> cowpaths, and solving real problems.
>
> Do they have to be HTML cowpaths? There's a very deep cowpath here,
> dating back at least to the Socratic Method. Solving real problems is
> definitely in scope - much of the recent work on dialogue mapping is
> concerned with solving "Wicked Problems" [1].

Generally, the microformats that have blossomed have had one of two  
things going for them:

1) They are based on an existing type of data, and were developed  
because a need was identified to mark up information of that type  
(hCard/vCard, hCalendar/vCalendar).
OR
2) They are based upon the type of markup that exists out there. Near  
as I can tell this applies to most other ones, like from hResume to  
XOXO.

The key, though, is to look at existing markup and existing pages and  
identify a problem people are already trying to use markup to solve.  
People have been marking up contact information for ages. Calendars,  
resumes, lists too.

But I really haven't seen people do this to arguments.

>> it seems people are perfectly content to use the semantic information
>> built into English to denote parts of an argument, rather than  
>> marking
>> things up with HTML.
>
> Ok, maybe this is largely the case on the web (though it has been
> done, see [2]), though elsewhere a lot of work has been done on
> getting the machines to help (e.g. see [3]). I'd suggest the relative
> paucity of this kind of material on the web is at least in part due to
> the lack of available tools (such as microformats) for expressing the
> information in a useful machine-processable fashion.

Microformats are about the web. The web is humongous place. If there  
is a real problem that needs to be solved, people will have attacked  
it time and time again, from numerous angles.

I suggest to start, write some plain ol' semantic HTML. Getting the  
markup out there is a good thing. Try to get people you know who run  
other BBSes or are interested to write other types of software that  
marks this stuff up. Get the makrup out on the web, and if it takes  
off, come back and we'll see.

That's my suggestion anyway.

-Colin