[iDC] The three unsolved problems in RDF/OWL ontology

[Paul Prueitt]

The four unsolved problems in RDF/OWL ontology

ontology merge
orchestration of different viewpoints
a process model that (sufficiently) accounts for changes in context
discovery of new interpretations of data


all of these are often managed well by humans using cognitive  
abilities and natural language.

Topic maps is a pre-cursor to a semiotic system (a system of emerging  
signs used to convey information), it also has difficulties with  
these four unsolved problems.  However, the topic map standard was  
originally developed to not fully constrain meaning, but rather to  
allow an interpretative process.

Topic maps was not designed to do what the RDF/OWL folks are trying  
to do, and RDF/OWL over constrains "formal semantics" so that there  
is a limit to the usefulness where usefulness is measured by average  
people working on everyday problems.


The difference between RDF/OWL and topic maps is very close to the  
difference between early Wittgenstein and later Wittgenstein.  The  
difference is seen in many religious arguments, and uncovers one's  
predisposition to scientific reductionism or to a post reductionist  
philosophy of natural science.

What the bleep is a GRDDL?



On Mar 17, 2008, at 9:40 AM, Danny Ayers wrote:

> On 17/03/2008, Paul Prueitt <psp at ontologystream.com> wrote:
>> What stops the exploration of topic maps in the US markets is the
>>  stranglehold that RDF has.   One may choose to ignore the mind set
>>  that AI and RDF have created over funding.  (www.secondschool.net)
>
> I've been following RDF since fairly early on, and historically it's
> been notable how much of the development has been bottom-up, and
> driven by independent enthusiasts, with little more than lip service
> at best from the big players.
>
> While RDF might now attract funding that might otherwise have gone
> into topic maps, I'd suggest the fact that knowledge representation on
> the web is actually getting funding is a positive thing - there are
> 1001 different areas the money could have gone. This isn't necessarily
> a bad thing for topic maps, which are under the same general umbrella.
> Build a compelling app with topic maps, use the buzzword 'semantic'
> when you talk to the VCs, laugh all the way to the bank...
>
>>  Yes there is some value in RDF, I celebrate this also.   The  
>> point is
>>  that there is no free market here, there is gaming and fraud.  So  
>> RDF
>>  and topic maps do not compete on merits - unless the merits are
>>  defined as profits to the incumbent IT powers.
>
> This seems a very glass-half-empty, and not a little paranoid
> perspective. If you want to encourage wider adoption of topic maps
> then your time might be considerably better spent writing a GRDDL
> transform for XTM, rather than tilting at windmills.
>
>>  Eventually, like alternating current overcoming the demonization and
>>  distortions by Edison and friends, topic maps will be seen to be
>>  interpretable in real time by humans and yet process-able by
>>  computing environments.  RDF is not interpretable by normal folks in
>>  everyday situations.  The use of RDF requires a clergy.  (Oh my, how
>>  nice!  Profits and control.  )
>
> Sorry, you are misinformed. The model is simple, and there's no reason
> for the syntaxes to be difficult. Check Turtle syntax, RDFa, even
> microformats like XFN which can be directly interpreted as RDF. An
> awful lot of *existing* applications can use  RDF as a medium, no
> change to end-user requirements at all.
>
> Even when you do deal with the syntax directly, it's not exactly
> rocket science.
> Try:
>
> @prefix : <http://xmlns.com/foaf/0.1/>
>
> [ a :Person ; :name "Danny Ayers" ]
> :knows
> [ a :Person ; :name "Paul Prueitt" ] .
>
> Ok, I don't expect you to know what the @prefix line is for (it makes
> the terms prefixed with : into URIs), but the rest is a lot more
> legible than, say, XTM.
>
>>  The cute arguments by the RDF crowd about the
>>
>>> all intents and purposes topic maps can
>>> be expressed in RDF
>>
>> was a polemic strong enough to inhibit topic map as an alternative.
>>  Yea for the incumbents!
>
> There's no need to view Topic Maps and RDF as rival technologies when
> Topic Maps can be expressed in RDF.
>
>>  Natural language use does not force a logic on us, in the same  
>> way as
>>  OWL does.
>
> No-one's forcing OWL on us.
>
> Topic maps are entailed when interpreted by humans, much
>>  like a book.  Logic is a religion.  It does have usefulness, but if
>>  there is no soul to the religion; it is only one more form of  
>> control
>>  over the masses.
>
> That isn't any more useful than saying "numbers are a religion" -
> logic is only one more branch of mathematics (that also happens to
> play well with computers).
>
>>  You say:
>>
>>
>>> While the Semantic Web
>>> languages are crude in expressive ability compared to human  
>>> language,
>>> they are considerably more tractable.
>>
>>
>> and I say " in what way "tractable".
>
> Much easier to use in software.
>
> The end goal here is not to
>>  turn over the world to a logic machine, but to empower people to
>>  communicate in new ways in real time about things that are  
>> happening.
>
> I don't disagree, check :
> http://sioc-project.org/
> http://www.foaf-project.org/
>
> While I agree communication is the biggy, computers can also help us
> with a lot more, for instance: computation.
>
>>  Unless there is some other kind of comment, I see no reason to
>>  continue this well know exposition... we have all seen this before.
>
> Fair enough.
>
> Cheers,
> Danny.
>
> -- 
> http://dannyayers.com
> ~
> http://blogs.talis.com/nodalities/this_weeks_semantic_web/