RE: Syntax and semantics.

At 16:18 15-01-01 -0500, LeVan,Ralph wrote:

>I do believe that that most (probably all) things that call themselves
>protocols are locked into a single syntax and transport mechanism.  My point
>is that it is still z39.50, even if it is encoded in a new syntax and sent
>over a new transport mechanism.  So, z39.50 is NOT a protocol.  (That's too
>strong.  Z39.50 IS a protocol today.  I want to separate the z39.50
>semantics from the syntax and transport layer.  Then the protocol is Z39.50
>encoded in BER and sent over raw TCP/IP.  And, in fact, there is an RFP that
>describes just that.  I'm going to come out with an RFP that says how to do
>z39.50 in XML embedded in HTTP and another RFP that says how to do z39.50 in
>URLs.)

I still feel like this is being done (assuming you do it) based on 
insufficient analysis.

I think it was at the San Antonio where Bill Moen first introduced the 
notion of analysing and isolating the (perceived) strengths of Z39.50 as 
something that could possibly be applied in other types of communications 
environments. This struck me as a pretty smart proposal - a good way to 
ensure that our intellectual work survives as the web community (or any 
other community) figures out how *it* wants to do IR.

However, it seems to me that we have never actually carried out that 
analysis. Instead, we have allowed ourself to get whipped into an 
ever-increasing panic over the issue of web-friendliness, and we're now 
approaching what looks like a race to see who can crank out the most RFCs 
(RFPs ?) and whatnot in the shortest possible time. It may all turn out for 
the best (and heck, maybe I'll get into the mood and make a couple 
protocols of my own), but frankly, I am a little concerned. It's like 
burning down your own village before the vikings arrive.

--Sebastian
--
Sebastian Hammer        <quinn@indexdata.dk>            Index Data ApS
Ph.: +45 3341 0100    <http://www.indexdata.dk>    Fax: +45 3341 0101

Received on Thursday, 1 February 2001 09:05:49 UTC