Re: HM.applications-Translations

[Niclas Olofsson <gurun@acc.umu.se>]

A serious article for a change.

I like your example Sean. It's very basic and easily understood. More of
these from all of us please. However, I'm not going to provide one in
this mail.

"Sean B. Palmer" wrote:
> You'd need a rulebase 

Ok. You have shown the basics of an application of HumanML. I like
systems that rewards the author. Unfortunently I think this is not the
case in many meta-data or annotation standards/techs. In the application
you presented the reward goes to the the user of the data, not to it's
author (with a little "but" attached to this observation).

I seldom do schemas when I'm the sender of data. It is not in my
interest, because there is no reward. If I reviece XML I often do it,
because the reward is instant. 

A HumanML enabled application instantly rewards it's users, the authors
of the profiles. This is a good thing. HumanML enabled applications are
user friendly by nature. To the authors of the data, it is un-rewarding,
thus I have the feeling that it HumanML applications need better
mechanisms than rulebases, maps, trees, ontologies, taxonomies, etc. 

In this particular application (the search engine), how do one model the
interaction so that both the author and the user gets the reward? Is it
possible? If I'm the author of the data, what's in it for me?

Cheers,
/Niclas
-- 
Niclas Olofsson - http://www.ismobile.com
Product Development, isMobile, Aurorum 2, S-977 75 Luleå, Sweden
Phone: +46(0)920-75550
Mobile: +46(0)70-3726404