Re: HM.applications-Translations

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

Ah, ok. I slightly missunderstood your RDF there for a moment. But this
is excellent Sean, in many ways. I assume that the syntax could look
something like...

1)
  [ :authorOf this; 
    :hu-author-profile "http://ismobile.com/niclas.olofsson-v-1_1.xml" ] 

And the user could provide the following profile for the search.

2)
    http://ismobile.com/niclas.olofsson-public-v-1_0.xml

or

3)
    http://humanmarkup.org/swedish-blond-public-v-1_0.xml

or the combination of the two (or more).

Where the difference and purpose would somewhat similar to having a
system and a public schema or interface. It would mean that the public
profile I use is probably a filtered version of my private/system
version, mostly for integrity reasons. The profile of the data would be
used to make judgment calls on the data, and the profile of the search
would be used to make judgment calls on the search criteria.

The reward is in enhanced quallity of the interaction between the author
and the user, yes. This is pretty much what my best guess HumanML is
about. I was looking for reward. Well, perhaps this was slightly wrong
then. I think the wording should have been .. ease of use, simplicity, 
cost effecient. If you provide the profile in a way like 1) above, the
cost for the author is kept at a minimum. The least he can do is to
provide NIL and he probably will be classified as conformant to the
common universal profile (the one that describes this human as being
part of the universe). If the author happens to have a personal, private
profile (custom made for him only) he can choose to use this, if not, he
can still choose from other common humanmarkup profiles availble. Since
this search engine is HumanML enabled, it knows how to apply to profile
on the data. No need for human interaction there. Almost no cost
involved. The example works a little bit like a marketing department. If
and only if it is in the interest of the author to enhance the
communication, it is a cost involved. Thus, the catch is pretty much the
same as in VRML. The cost is motivated by the spread of the medium, the
size of the userbase. For this application to work, HumanML must spread
like a virus. Correct? Is this true for all HumanML applications? Does
this mean that the HumanML business model sucks big time?

Cheers,
/Niclas

"Sean B. Palmer" wrote:
> 
> [...]
> > If I'm the author of the data, what's in it for me?
> 
> What's in it for you is that when you write:-
> 
>    "I love Norweigians."
> 
> And then sign your document:-
> 
>    [ :authorOf this; :nationality "Swedish" ] .
> 
> People can tell that you're being sarcastic to the extreme.
> 
> My contrived example aside, I hope the principle is clear: the benefit to
> the author is that their information has less of a chance of being
> misinterpreted once the document it read in a wider context than that which
> the author intended, or was aware of.
> 
> --
> Kindest Regards,
> Sean B. Palmer
> @prefix : <http://webns.net/roughterms/> .
> :Sean :hasHomepage <http://purl.org/net/sbp/> .

-- 
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