[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Being "precise" vs being "human"

• To: "'Simon St.Laurent'" <simonstl@s...>, "Clark C . Evans" <cce@c...>
• Subject: RE: Being "precise" vs being "human"
• From: "Bullard, Claude L (Len)" <clbullar@i...>
• Date: Tue, 22 Jan 2002 12:40:06 -0600
• Cc: xml-dev@l...

And we have almost perfectly replicated the long 
thread on the Semantic Web/Ontologies topics from 
last year.  It must be important.

So far, the most illuminating point is that RDDL 
is a simple way to enable the site owner to assert 
their own "meaning" for a signal in a relationship in which they 
dominate the first transaction/sentence.  If meaning 
is first discoverable then negotiable, RDDL is good 
for that first transaction.   It is the negotiation 
that follows which was of interest in the original 
posting I made.  That is, negotiating out noise 
means identifying what is and isn't signal in a 
succesion of exhanges, a conversation you will, the 
point of which is to ultimately exchange value.  This 
is too abstract, I guess.  If I want services, I am 
not yet to the point of negotiation of the file 
types; I need to know if the site supports those 
services.  I don't know if RDDL is too much or 
just enough for that.

What if one isn't at the point of exchanging XML 
or any document?  Isn't a web service discovery 
system contingent on first identifying the types 
of resources one might want to choose from?  What 
one might want is the offering (RFI) before the 
specification (RFP) and definitely before the 
contract (BAFO). It seems to me that focusing 
on RDDL hijacked other possibilities such as 
UDDI when essentially, RDDL is not deployed in 
many sites and UDDI is picking up momentum.

"Only people have meaning" so the saying goes. 
I'll grant that.  Only people really need 
computers too.  A tool is a tool.  What must 
be discoverable is "does he mean what I mean 
and if not, how shall we talk if I need him 
to mean what I mean?"

len

-----Original Message-----
From: Simon St.Laurent [mailto:simonstl@s...]

God aside, I think it's the nature of the subjects being discussed this
particular week.  URIs and semantics are both contentious zones, where
using words in particular ways can madden people quickly.

If you want real fun, try re-reading the xml-uri@w... archives,
uri@w..., or urn-ietf@n... archives.

My answer to the whole mess is simple: there ain't any meaning, just
labels and structures.  Anything else is a bonus or a nightmare,
depending.

• Prev by Date: PerlSAX_on_Win95
• Next by Date: RE: PerlSAX_on_Win95
• Previous by thread: PerlSAX_on_Win95
• Next by thread: RE: Being "precise" vs being "human"
• Index(es):
  • Date
  • Thread