Paul, Doug, Pat, Kingsley, Leo, and David, (01)
As I'm writing this note, there is a talk show on the radio in the
background. The host is in New York, and somebody called on their
iPhone while being stuck in traffic in Abu Dhabi. That is an example
of what successful technology can do. (02)
As for ontology, I'd like to ask what problems we might solve that
(1) could have a comparable impact on society, (2) take advantage
of the existing IT infrastructure, and (3) be built in a time frame
comparable to the development of the iPhone. (03)
PT
> I assume your comments in this thread are in the
> mode of a professor heckling his seminarians to defend their theses, and
> are not charged with any personal animosity towards those who favor this
> or that technology. (04)
I have no animosity whatever about any technical disagreements. (05)
But I am *not* trying to heckle academics. I love academic studies,
but there's a huge gap between a PhD thesis and a product -- or between
a committee of academics and something that works. (06)
PT
> My operating formula is simple: I adopt any technology, methodology,
> principle, or practice that helps this aim; I avoid what doesn't; I
> assail what degrades it; and I assay what shows promise. (07)
I agree with the principle. But instead of asking for "promise", I
suggest that we ask how and whether we could achieve the success
of the iPhone, Google, Facebook, etc. (08)
DF
> The [W3C] have been too concerned with simplicity (imho), restricting
> statements to sets of triples and limiting the expressiveness for the
> purpose of decidability.
>
> Mainstream IT does not restrict its languages to ones that guarantee
> that any program written in them will complete in polynomial time.
> Mainstream IT looks at the simplified tools, concepts, and standards
> of the Semantic Web and finds that they do not meet their needs. (09)
Mainstream IT needs something that works. They have engineers who solve
problems within the constraints of budgets, deadlines, and resources. (010)
The SW gave them tools that are completely disconnected with their
problems. Those few people who tried to use them failed to solve
their problems within the required constraints. (011)
DF
> If we want uptake, we have to move beyond the simplified syntax of
> triples and the simplified DL languages, imho. (012)
The W3C made the mistake of starting with syntax. That mistake can't
be corrected by just moving to a different syntax. (013)
I would focus on the huge backlog of legacy systems and people with
legacy skills. If you consider their skills obsolete, you'll fail.
But if you consider them a valuable resource, you have a chance to
succeed beyond your wildest dreams. (014)
PC
> I believe that most or all of your concerns can be addressed by using a
> common primitives-based foundation ontology (PIFO)... (015)
As I've said for years, I am not against the idea of using primitives,
but that is just one among many academic proposals. The major question
is how do you enable people with their current skills (possibly with
a short course) to use the tools successfully to solve their problems. (016)
KI
> RDF as a moniker is an unfortunate conflation of:
>
> 1. Data Model - EAV enhanced with URIs plus explicit semantics for
> typed literals and language tags .
>
> 2. A number of data representation syntaxes and across-the-wire
> serialization formats.
>
> With the above in mind, there is not broken genealogy re. LISP and
> other critical pieces of this innovation continuum re. structured data
> representation. Thus, we have to try to speak in clearer terms about RDF. (017)
Those are useful comments about the underlying technology. But my main
complaint about the SW is not about Tim B-L's vision in 1994. I liked
the vision from the beginning. My major complaint is that the vision
was never connected with any problems that anybody needed to solve. (018)
Tim's major success was the WWW -- which is in the same category as
the iPhone, Google, and Facebook as successful from Day 1. He was
working as an engineer who had to solve the customer's problem
(enabling physicists to share research papers) within the constraints
of budget, deadline, and resources. He finished it in one year with
a few assistants. And it worked. (019)
For the SW, the W3C had no clue about what problem they were solving,
how they would solve it, or what kind of budget, deadline, or resources
they would need. But they wanted to do something, so they started
at the *worst* possible level: syntax. (020)
KI
> a significant number of folks will still interpret [RDF] as a statement
> about RDF/XML syntax rather than an exploitation of the RDF Data Model. (021)
Both the syntax and the data model were premature optimizations that
should never have been considered until *after* the W3C had some idea
of what problem they were trying to solve. (022)
Leo
> It [naming scheme] really depends on both the tools and the ontology
> developers’ knowledge of English. And of course it helps if you eventually
> have vocabularies you can map to that. (023)
I don't disagree. But that is still syntax. (024)
And the word 'eventually' implies "no deadline". No serious engineering
project has a deadline of eventually. (025)
DE
> So what's the purpose/value/utility of the ontology view?
>
> If grunts are working away with their local vocabularies/jargon/local
> opaque language (& VERY unlikely to be aware of the ontology in the
>background),
> what is the value add of the ontology? (026)
That is a good question. But you can only answer a question about value
in terms of problems that are solved better, faster, or cheaper. (027)
Recommendation: Instead of proposing yet another syntax, data model,
semantic theory, or super algorithm, I suggest we focus on *problems*
that people actually need to solve -- and *how* ontology could help
solve them within a reasonable time frame. (028)
To be precise, I define "reasonable" as the time to develop the iPhone
and make it work successfully with the current infrastructure. (029)
John (030)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (031)
|