"essential characteristics"

AWWW says:

"The distinguishing characteristic of [information] resources is that all of
their essential characteristics can be conveyed in a message."

Now I have always taken this to be complete gibberish, some kind of
tragic artifact of design-by-committee. Even supposing you can tell,
for any particular characteristic, whether it's essential or not,
"convey" seems wrong. If my essential characteristics are birth date,
birth place, parents' names, and name (none of those appearing to be
accidental, and the aggregate being unique to me), or even a subset,
certainly those characteristics can be conveyed in a message? They are
conveyed in my birth certificate, and that to me qualifies as being a
message. Am I therefore an information resource? Similarly, it seems
obvious that the number seven has conveyable essential
characteristics, yet according to Tim it's not supposed to be an
information resource. It just doesn't make sense.

Henry's lecture in Edinburgh in September, by taking AWWW more
seriously than I do, forced me to take a second look, however, and
maybe AWWW, instead of being a total miss, is just a near miss. If you
look at my elaboration [1] of Tim's "generic resource" idea it says
that properties (characteristics) of a generic resource are shared
with the specific resources (messages) that specialize it. That says
these characteristics can be *possessed* by messages. So if you switch
"conveyed" to "possessed by" you get a plausible definition.

"... all of their essential characteristics are ones that a message can have."

"... all of their essential characteristics can be possessed by a message."

In other words: Any essential-characteristic-class that contains an
information resource also contains a message. (E.g. if messages don't
have mass, and having mass is an essential characteristic, then IRs
don't have mass.)

I wonder if this is just yet another error of indirection, and that
maybe the 2005 TAG would have agreed that we're talking about
characteristics a message might *have* as opposed to ones a message
might *convey*. Then IRs become similar to messages, which is what
Tim's architecture needs.

The question of which classes / characteristics are "essential" and
which aren't is a weakness in my theory as well, and I'm trying to
figure out how to answer it.

Jonathan

[1] http://www.w3.org/2001/tag/awwsw/ir/20110625/

Received on Wednesday, 5 October 2011 14:19:27 UTC