Patrick, (01)
On Thursday February 19 2009, Patrick Cassidy wrote:
> John and PatH
> Your arguments to take a gradual approach, look for small demos
> and eventually try to get broad agreement via the semantic web is
> based solidly on the notion that a very gradual approach is a
> low-cost solution. It's not. (02)
That is only true if for problems that are well understood and which
thus fall squarely into the realm of engineering. If there is more than
a very modest degree of experimentation involved, incrementalism is
indicated. (03)
> The economic losses due to failure to
> implement a base for semantic interoperability are enormous and
> mounting every day. Just because such losses do not show up as a
> line item on someone's budget doesn't mean they are not real. (04)
But likewise, just because a problem exists, does not mean its solution
is at hand. (05)
> The direct method of funding a consortium to develop and test an FO is
> in fact the cheapest method, when the savings due to more rapid
> implementation of semantic interoperability are factored in. (06)
People often toss around the "Manhattan Project" term when they want
badly to achieve some ambitious goal. A similar analogistic argument
used in these contexts take the form "If we could put a man on the
moon, surely we can X." (07)
This sort of argument was used a lot to make the case that whatever
resources required should be devoted to curing AIDS and / or to
creating a vaccine to protect against it. But this mischaracterizes the
problem as one of applying some known science in what is mostly an
engineering endeavor (however large). But that was not the case with
AIDS. Not enough was (or still is) known to cure it or immunize against
it. (08)
The semantic interoperability and the generic / foundation / universal
ontology problems seem to be in the same category: we don't yet know
enough to solve them. Or, perhaps, we don't know what a good solution
would even look like. (09)
Another way of looking at this is that if semantic interoperability is a
problem that is understood and, in principle, solved, then you should
be able to convince a venture capitalist to fund the task of reducing
that understanding to practice. If the economics are of the magnitude
you suggest and the problem's solution is understood, any savvy
investor would jump at the opportunity to do so. Are they? (010)
> Even with a very pessimistic estimate of the likely effectiveness of
> the consortium approach, it is still cost-effective.
>
> We are now living through one of the disasters that the "let the
> market take its course" attitude can produce. (011)
If you're referring to our current economic tribulations, I don't think
the analogy is a good one (and I by no means believe markets are the
magic entities many claim they are). The financial crisis that
triggered our current larger economic problems were about putting no
limits on how one category of businesses (investment banks) used other
people's money. I don't see how that applies to allowing the technology
realm free reign in addressing a problem that, by all appearances, is
not sufficiently understood to be (even roughly) solved. (012)
> Some well-considered pro-active effort can save a lot of money in the
> long run. If you have any qualms about specific parts of the FO
> proposal, suggest modifications. But to imagine that the current kind
> and rate of effort will achieve semantic interoperability without a
> coordinated effort to get agreement on some major part of an FO is
> merely an act of faith with no evidence to support it. (013)
It sounds to me like an assertion that wanting something badly enough is
going to alter the state of progress toward that thing by itself. Or,
less glibly, that the problem is merely that the magnitude of the
effort being put forth is inadequate, that the failure is quantitative,
not qualitative. Again, I don't believe this is the case. (014)
> If we must take some course to a goal based on faith, taking the
> shortest course seems to me to be the best bet. (015)
If we could see what course was shortest, we'd be foolish not to take it
(at least in the sort of endeavor we're talking about here), but I
don't believe we know what that course is. The magnitude of
disagreement in this very forum seems to support this belief. (016)
> Pat (017)
Randall Schulz (018)
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (019)
|