David, (01)
That point is critical: (02)
DE> If one must be an ontologist to use ontologies, the effort will
> never get to the stadium parking lot, much less the starting blocks. (03)
People do know the ontology implicit in their data, but the current
tools don't help them discover it. I like to quote the following
point by Immanuel Kant (citation and longer excerpt below): (04)
Socrates said he was the midwife to his listeners, i.e., he made
them reflect better concerning that which they already knew, and
become better conscious of it. If we always knew what we know,
namely, in the use of certain words and concepts that are so
subtle in application, we would be astonished at the wealth of
our cognitions. (05)
What we need are tools that can play the role of Socrates in drawing
out the implicit ontology. And those tools should use language and
notation that the subject matter experts understand. (06)
People keep talking about how the Semantic Web is growing, but the
most rapid growth is in RDF -- which has no semantics, no ontology,
and almost no structure. The notation does nothing to guide anyone
toward semantics. (07)
In fact, the definition of RDF is just "a set of triples." Since
many people are using RDF, there is a bandwagon effect that brings
more people into the game. But most of the users don't use OWL,
and they don't even use RDFS. (08)
In fact, the biggest use of RDF could be replaced by CSV (Comma
Separated Values). For the way they use it, CSV is an *upgrade*
from RDF: (09)
1. It's more readable. (010)
2. It's easier to parse. (011)
3. It's more compact. (012)
4. It's more general, since it supports arbitrary n-tuples. (013)
5. And it's more widely supported by everything, including
legacy systems. (014)
Some people have added the option of a "blank node" in RDF
to support unknown or omitted options. CSV had that in the
1950s: two commas with nothing between them. (015)
Some people say that RDF is an advance over CSV because it uses
URIs. But those are just based on the Unix naming conventions,
prefixed with the name of some computer in the style of the
old ARPANET from 1969. You can put them between the commas. (016)
Some people say that CSV doesn't have self-describing data.
But you can add that option, and many people do: (017)
FirstName:Barack, LastName:Obama, Occupation:President, ... (018)
It's pitiful to see that the most widely used version of the
most widely hyped semantic system is a step down from a format
from the 1950s combined with a naming convention from 1969.
What's worse is that the old format is more readable, compact,
efficient, and easy to program than the new one. (019)
And by the way, I say this as somebody who was very hopeful about
the Semantic Web when Tim B-L proposed it in the late 1990s. I
still believe that real semantics can be added to the Semantic Web,
but the first step is to get beyond the obsession with notation. (020)
Fundamental principle: Semantics must be the focus. Any and
every notation that anybody prefers should be supported, but
the goal is *semantics* -- notation is secondary. (021)
John
____________________________________________________________________ (022)
Source: Immanuel Kant, "Vienna Logic," in _Lectures on Logic_,
translated and edited by J. Michael Young, Cambridge University Press,
pp. 297-298 (023)
Socrates said he was the midwife to his listeners, i.e., he made them
reflect better concerning that which they already knew, and become
better conscious of it. If we always knew what we know, namely,
in the use of certain words and concepts that are so subtle in
application, we would be astonished at the wealth of our cognitions. (024)
E.g., philosophers and jurists have not yet been able to develop and
explicate the concepts of justice and fairness. But we need only
give one of them a case _in concreto_ , and he will quickly say what
is just, and to what extent it is fair. In my concept of fairness,
then, something must lie hidden that is distinct from that of
justice; and I make use of this mark, which is clothed in obscurity
in the case... (025)
To make a distinct concept is the synthetic method, to make a
concept distinct is the analytic one... (026)
Mathematical distinctness is wholly synthetic. The mathematician
says that whatever people may have represented by a circle, I think
only that it is to be where a straight line moves around a fixed
point. Here I have not made the concept of a circle distinct,
but rather have made a distinct concept. (027)
_________________________________________________________________
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 (028)
|