RE: Init is dead?

We ARE proposing a new protocol.

Let's get our vocabulary very precise.

All implemented standards have three component: A semantic, a syntax and a
content rule.  The combination of the first two is usually called a
protocol.  (Actually, I guess there is a fourth component, the transfer
mechanism, which is raw TCP/IP for us.)  In the case of z39.50, we've been
locked into a single syntax (BER) and have gotten away with calling z39.50 a
protocol.

But, we're now exploring how to use z39.50 in other syntaxes, using the same
semantics.  If we do that, we will be creating new protocols.  Especially if
we use other transfer mechanisms, like HTTP.

So, for you folks determined that sessions are critical for z39.50, you'll
want to stick to the BER/TCP/IP protocol that encourages/requires sessions.
But, the semantics should NOT require an Init service, because it is NOT
critical for other z39.50 protocols.

Ralph

> -----Original Message-----
> From: Pieter Van Lierop [mailto:pvanlierop@geac.fr]
> Sent: Monday, January 15, 2001 11:18 AM
> To: www-zig@w3.org
> Subject: RE: Init is dead?
> 
> 
> > 	 if init is obmitted all the things the client/server learn
> > 	 via init are undefined and all behavior is acceptable...
> > 
> >   bob
> 
> Bob,
> My main argument is that the Init is the beginning of a 
> Z39.50 Session.
> Without Init, the idea of a session disappears. Correct?
> If no session, then no search history, no Result Sets, no 
> Present. Correct?
> If you take all this away, you take the heart out of Z39.50 (sniff).
> 
> If you want a simple Internet search (in Z39.50 terms: Search 
> with Piggy
> back), then why don't you take what you want from z39.50 and 
> put it in a new
> protocol?
> Let's call it Z39 Lite.
> 
> It seems to me that you are only interested in the Attribute Set
> architecture.
> 
> Pieter
>

Received on Monday, 15 January 2001 14:39:03 UTC