Re: [xml-dev] Mailmen, POST, Intent, and Duck Typing

[Chris Burdess <d09@hush.ai>]

Peter Hunsberger wrote:
>> Semantic disambiguation by context is only one aspect of pragmatics.
>> There's also paralinguistics, such as "what are you communicating by
>> sending this message above and beyond the content of the message?",
>> "what are you communicating by not sending a message?" (insert
>> interferometer analogy here), "what can I infer about your intent
>> when you send this message?"
>
> Ugg.  Makes sense, but now I fear that we'll end up with Web
> Psychiatrists:  "and tell me Mr. System, how did it make you feel when
> BigCo declined to communicate with you".
>
> Or worse, Web mediums: "I sense someone is attempting to communicate
> with you about an order for 1000 widgets. Do widgets sound familiar?
> Yes? That will be $200 please."
>
> Seriously, do you see applications for that aspect of the field to the
> Sem Web (other than perhaps a reason for celebration by the lawyers)?

Not really, except to assert that meaning is a lot more complex than  
semantics would have you believe. I have felt for many years that  
this traditional AI approach doesn't help to understand how humans  
represent meaning and is therefore of limited use in designing  
systems for the benefit of humans. The fact that "dumb" Bayesian  
networks with no semantic formalisms have been much more successful  
than expert systems in classifying spam, and therefore much more  
useful to real people, is perhaps a beacon in this regard.
-- 
犬 Chris Burdess
   "They that can give up essential liberty to obtain a little safety
   deserve neither liberty nor safety." - Benjamin Franklin

This is a digitally signed message part