Quoting Dan Frankowski <dfrankow cs.umn.edu>:
> FYI, here is an academic paper "Semantic
Wikipedia" with ideas I think
> are related, presented at WWW '06:
>
> http://www.aifb.uni-karlsruhe.de/WBS/hha/pa
pers/SemanticWikipedia.pdf
>
> Dan
Thanks Dan.
Actually, most of the "Semantic Web" style wikis
are not similar
to what I'm doing. The paper you cite is certainly one of
the
better ones, but most of the SW wiki projects are far too
geek-
friendly and human-unfriendly, where people are commonly
expected
to edit RDF almost directly and neither require those
performing
the edits to actually understand either the modeling
langauge nor
the ontological commitments they're making (indeed, most
people
will have no idea of the underlying modeling language even
though
that language has some very stringent commitments such as
monotonicity and no facility for context or subject
identity).
The wiki aspects of the Ceryle project lean much more
towards
informal modeling, faceted classification, and managed
metadata,
but the typed links aspects are somewhat similar. I
constantly
have to ask myself if I actually think "normal"
people would
ever bother to create the kinds of wiki markup that are
often
advocated in many of these projects, including my own.
One of the bigger faults with most of the "Semantic
Web" projects
is the use of formal logics (like Description Logics, the
basis
behind OWL and the W3C work) to support informal systems
like
found at Wikipedia -- it's simply the wrong tool for the
job. You
can't use formal logic like FOL to model informal discourse
relationships; there's no logical basis to make the enormous
leap
between the logic and the terms used in the
"ontology", therefore
no logical reasoning inferencing can be done -- it's simply
not
reasonable (in the formal sense of that term).
This is not to say people won't actually do this -- it's
just
that the ontological commitments they make have no
epistemological
basis whatsoever, so their results are dubious at best. I
don't
know if I've ever seen a project that even bothers with
things
like synonymy or polysemy, or the idea that multiple views
are
being represented within one context (!). Cyc's
microtheories
might seem to handle this but are not permitted to be in
conflict
with the Base knowledge base (to my knowledge -- there seems
to
be some question that if any predicates cannot be
axiomatized
they might be more properly considered primitives in the
BaseKB),
whereas in reality we probably need to be able to model
context,
disagreement, ambiguity, and even deliberate chaos and/or
obfuscation (such as metaphor, simile, irony, sarcasm,
humour,
and even absurdity). If someone writes "George Bush is
a pig"
we don't want the system to reason that his species is
scrofa
instead of sapiens. As you might imagine, we're a very long
way away from that kind of modeling, though some might get
the
impression given the sci-fi way these systems are often
touted.
My project makes almost no ontological commitments
whatsoever
(nor could it, if one considers that a wiki is
group-edited).
But as you may correctly surmise, these kinds of issues are
well beyond the scope of most "Semantic Web"
projects -- they
just shouldn't be.
Murray
............................................................
...............
Murray Altheim <murray06altheim.com>
=== = =
http://www.altheim.com
/murray/ = = ===
SGML Grease Monkey, Banjo Player, Wantanabe Zen Monk
= = = =
In the evening
The rice leaves in the garden
Rustle in the autumn wind
That blows through my reed hut. -- Minamoto no
Tsunenobu
_______________________________________________
Jspwiki-users mailing list
Jspwiki-usersecyrd.com
http://ecyrd.com/cgi-bin/mailman/listinfo/jspwiki-users
|