[
Lists Home |
Date Index |
Thread Index
]
1/25/2002 4:46:15 PM, Mark Baker <distobj@acm.org> wrote:
>It only seems fuzzy because the principles behind it aren't well
>understood. In actuality, it's anything but; it's the most powerful
>distributed software infrastructure that's ever been built.
I meant "fuzzy" in the fairly rigorous sense of fuzzy logic/set
theory, see http://www.sciam.com/askexpert/computers/computers8.html
Definitely not in the perjorative sense of something that isn't
powerful.
From the SciAm page: "Fuzzy logic is a generalization of standard
logic, in which a concept can possess a degree of truth anywhere
between 0.0 and 1.0. Standard logic applies only to concepts that are
completely true (having degree of truth 1.0) or completely false
(having degree of truth 0.0). Fuzzy logic is supposed to be used for
reasoning about inherently vague concepts, such as 'tallness.'"
My point is simply that many things we discuss here such as "what the
XML spec says" [c.f the discussion of what non-validating parsers are
supposed to do with DTD information or the lack thereof], what "The
Web" is, what the relationship between Push/Pull and event/tree
processing models are, are "inherently vague concepts." Maybe the
XML spec could be formalized to the point where it is no longer
intrinsically vague, but I doubt if we'll ever fully agree on the
other stuff.
I find "fuzziness" in this sense a very useful organizing concept:
rather than debate what "The Web really is," propose criteria to
define the "Webness" of a technology. HTTP, HTML, and URIs probably
have a 1.0 "degree of membership" in the fuzzy set "The Web" whereas
UDP, FTP, PDF, etc. have a somewhat lower membership. The "Webness"
of XML is an interesting question ....
There are products (mostly designed in Japan) that take the "degree
of truth" numbers and the math very seriously here and do powerful
things; I would not even attempt to argue that we should do so here,
I just find in a useful heuristic device for thinking about nasty "is
X a Y" problems. For example, in some sense it doesn't matter what
the intent of the XML spec says: if all processors implement some XML
feature in the same way, it is really and truly part of "XML." If
there is agreement in principle, but non-interoperability in
practice, it has a high, but not complete degree of membership in
"XML." If the people here scratch their heads and can't agree on
what the spec implies, and the processors do different things, it has
a low but non-zero degree of membership in "XML."
So, a practical person might use this heuristic and say "Hmmm, that
feature has a non-unity degree of membership in XML, I think I'll
stay away from it" rather than "Hmmm, the consensus of the gurus on
xml-dev is that that feature is really and truly part of XML, so I'll
use it."
|
