W3C home > Mailing lists > Public > www-zig@w3.org > February 2002

Re: SV: Z39.50 on the web (and in print)

From: Rob Bull <bull@crxnet.com>
Date: Mon, 25 Feb 2002 11:05:53 -0000
Message-ID: <00aa01c1bdec$646cada0$2816c3c1@crxnet.com>
To: "Robert Sanderson" <azaroth@liverpool.ac.uk>
Cc: <www-zig@w3.org>
Rob,

>I'm sorry, but Explain Lite is simply not sufficient :(  Perhaps it might 
>please some people, but I'm 100% certain that a better XML definition 
>could be relatively easily agreed upon that could support better 
>definitions of what Z can really do, with less confusion and better ease 
>of autocreation.

there are a number of overlapping issues that Explain Lite addresses.

Firstly - there is the current status quo - 99% of Z servers do not support Explain. History of this would 
suggest that this wont change in the future - and even if there was more takeup, there is still a huge legacy
of servers that will probably never have Explain added.   Explain Lite does not use SOAP or any other
protocol - it has a simple negotiation on Init.  Init request can ask for the XML data to be sent on the Init response.

Secondly, there is the issue of management - whatever format a server returns explain information to a client in
(whether XML, GRS, Explain records) that information still has to be entered by someone administering the
server, although some of that information could be generated automatically by the system itself. We chose to use
XML because it is current trend - and because the same data file can be used for the third point below. We chose
XML because Init allows this to be easily used - no new ASN code, no

Thirdly, there is the issue of discovering servers and their characteristics.  If you do this only within the Explain service
you need a Z client that supports the protocol and all of Explain before you start. And then you still need to find where 
those servers are. Some organisations post this via web pages, there are various registries emerging - no common approach.
We chose to use XML (ie Explain Lite) because it is the same common data as can be returned on Init. Thats
why the registry at RDN is useful - it allows you to find those servers.  

>When I have 30+ databases on my machine, each of which has 20+ 
>possible searches and another 20+ aliases for them, are you Really Sure 
>you want all that info in one go?  Wouldn't you rather be able to search, 
>scan and sort it like a regular database?

and I would bounce this back to you - do you really want all those explain searches and Explain-RS records received as a the
way to find out about a server - whats the difference ?     The whole mechanism sucks.
Explain Lite probably returns a  few tens of Kb extra with XML  - the client still has to deal with the data. Do I really
want to hang around waiting for the scanning/sorting to take place - no.   

As Jacob says - Explain Lite has a very specific purpose for addressing OPACs and its also applicable to smaller systems. It does this very well - and its here and working and is real easy to add on either a server or a web page.  We've been down the full Explain path years ago - we've shouted about it from the rooftops before - but we've also listened to the deafening silence and the message is clear - folks in the main are simply not going to do full Explain - not now, not ever.

Rob
Received on Monday, 25 February 2002 06:10:56 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:13:27 UTC