[
Lists Home |
Date Index |
Thread Index
]
- From: "Bullard, Claude L (Len)" <clbullar@ingr.com>
- To: "Thomas B. Passin" <tpassin@home.com>, xml-dev@lists.xml.org
- Date: Fri, 20 Oct 2000 08:26:19 -0500
Because it gets us into the same trap of trying to have one service
in the system dominate the design and direction. That is a bad idea
from the start. It would be disastrous if all of the specifications
became bogged down trying to satisfy the requirements of meta-vocabularies.
It is a stone of sysyphus.
Perhaps one should step back and ask the question, if semantics are
the dominating requirement for a next generation web, what requirements
emerge easily from that? If you can't get that straight and explain
it, then the terminology is worse than useless; it actively defeats
the ability of developers to understand what they should do next.
If you say however, we have a set of standard XML application languages that
inference engine services can exchange by a stateless asynchronous protocol
of
messages, you have a well-established professional cadre of engineers who
understand what needs to be done. They have worked on such since Turing
and VonNeumman talked about such and nothing about cs has changed except
the acceptance of commodity protocols and markup has changed.
Len Bullard
Intergraph Public Safety
clbullar@ingr.com
http://www.mp3.com/LenBullard
Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h
From: Thomas B. Passin [mailto:tpassin@home.com]
How about "Associative Web" instead of "Semantic Web"? I
think that "associate" probably covers everything everyone's
brought up, but at the same time is more specific and
evocative than "semantic". Services associate a provider
with a consumer. You could associate a meaning with a term
or resource (if you could figure out how to specify it).
And assigning a property/value pair to something can easily
be looked upon as asserting an association.
|
