Re: How will the semantic web emerge

Joshua Allen wrote:
> 
>>>    there must be de facto standards of classifing
>>>    information or else the very cause of classification
>>>    is useless.
>>>
>>>People point at a mountain and utter a sign.  That's how language
>>>starts.  People fall into habits of signing and people learn those
>>>habits.  No "de facto standards" are necessary, just habits ...
> 
> repeated
> 
>>>usages ... people repeating things that work for them.
>>>
>>
>>Yep.  Then someone decides to record descriptions of all those
> 
> habitual
> 
>>usages in a document of some kind (often called a "dictionary" in
>>non-computer circles).  And people start referring to that document
> 
> 
> Dictionaries are simply snapshot of current usage at a particular point
> in time [0].  The disease of attributing any more authority than that is
> thankfully constrained to the priesthood; the "vulgate" or "putonghua"
> simply ignore the dictionary and talk to each other.

That dictionaries are a snapshot of current usage is more or less 
consistent with the scenario I described (as is people deciding for 
themselves how much authority to attribute to them).  One of my points, 
however, was that just because someone decides to write down what the 
usage is doesn't imply an attempt to start a new religion, authoritative 
or not.  Re "simply ignore the dictionary and talk to each other", this 
is a bit simpler if there's a human in the loop than for strictly 
machine-to-machine isn't it?

> 
> The argument for ontologies is often that they are a pre-requisite for
> meaningful communication.  Imagine how ridiculous on the face it would
> be to claim that mother could not converse with child without first
> reading a dictionary.  Therefore, ontologies and dictionaries are not at
> all alike.
> 

I've certainly heard that argument for ontologies, but, as you say, it's 
only "often" the argument, and not the only one.  Besides, "meaningful 
communication" between what kind of entities are we talking about? 
(Certainly not between mothers and children!)

I'm not sure the conclusion that ontologies and dictionaries are "not at 
all alike" necessarily follows from your premises ("may be different, 
depending on who's using the terms" I would certainly accept).  It seems 
to me there's a clear relationship, particularly when machine-to-machine 
communication is involved (or how about if the people in the 
communication don't share the same natural language?  Of course, 
gestures and pointing at things work up to a point...!)  The Wikipedia 
entries for "folksonomy", "controlled vocabulary", "dictionary", and 
"ontology (computer science)" might be helpful references in further 
discussion (if we're allowed to use references of course!).

--Frank

Received on Friday, 16 December 2005 22:03:35 UTC