RE: HM.applications-Translations

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

One way to think of an ontology is as a set of scoped candidates 
for the WHERE clause of a SELECT statement.   Again, the architecture 
of the Semantic Web is fairly unremarkable.  All it really is 
is an attempt like more Berners-Lee projects to limit representation 
to a single system.  There are pluses and minuses to that just 
as the myths of naming, location and addressing he promoted have 
advantages and disadvantages.  From an engineering perspective 
of single system unity, there are advantages; from a perspective 
of information lifecycle, there are disadvantages which is why 
OASIS worked hard to get the XML Catalog systems up and running.

A cultural description can be just another set of search terms. 
Think of a schema, a DTD, etc (apart from the data typing) as 
a way to impose a view.  Where you put a view in the pipeline 
is up to the application.

I think we will have to get above implementation and discuss 
the overall metaphor (in the XP sense) that guides the 
project.  For me, that is obviously human awareness and 
so the challenge becomes to express this in an implementable 
representation.   I've asked previously, what is unique about 
human communication, and the problem is programmers are not 
humanists or human communications experts.  They want to 
go directly to the representation implementation and skip 
the meat of the content.   


Len Bullard
Intergraph Public Safety
clbullar@ingr.com
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Mark Brownell [mailto:gizmotron@earthlink.net]
Sent: Wednesday, September 05, 2001 6:11 PM
To: gurun@acc.umu.se; humanmarkup-comment@lists.oasis-open.org
Subject: Re: HM.applications-Translations


> Mark Brownell wrote:
> > How about a selection option by the user to do a culturally equivalent
> > procedure in the search instructions. 
> 
> Am I oversimplifying if I would suggest +".se" to be such an
> instruction? .se itself is of course not a cultural instruction, but
> rather a filter that I know will make the search more interesting to me.
> .se for me just mean that the probability that it will be something from
> my culture is much greater than if I put .com in there. It's just a bad
> example, and as such I can put whatever I want into it. Seriously, do I
> understand this discussion correct if I understand it as sending a
> cultural profile to the search engine it would give me the results based
> on this? It would mean that I can search on turkish things using a
> turkish cultural profile.
> 
> Cheers,
> /Niclas

I'm always tinkering with the machine or at least trying to think of ways to
do things. 
So I would first think of a simplification mechanism for the user of that
search engine 
as a first solution. So I see options that the user can use to trigger SQL
instruction 
that are then sent to the online database. The user never sees the
complexity of the 
SQL yet through a more human readable interface adds cultural conditionality
to the 
search engine's function. Now I'm thinking of data located in databases on a
server 
somewhere, that when queried, has the ability to return information in the
form that 
the machine request it in.  So this search engine can have several input
capabilities 
that predefine the results that it will return. I hoping for a stack of MTML
formatted 
RDF Dublin Core styled portal objects that I can run in the sub-browser and
run 
relational text gathering queries for final results. I'm trying to build a
system that 
works partly on the server and partly on the computer that the user is
using. My goal 
here is to achieve speed and archiving at the same time. 

Mark


----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>