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

Mike,

- 2 answers to your question.
In the shape of Web Clarity its a very useful registry, so long as its kept up to date. The users of
the latest Bookwhere client (v4) can periodically get their local database list updated automatically by pressing a button in Bookwhere.  

However, part of the reason the ONE-2 project developed Explain Lite was to represent inclusion of some metadata - it is not formally added in any form like Dublin Core - but could be so.

Look at www.bibsys.no/z/bibz.xml  for an example.  In addition, Explain Lite is being used in a UK academic registry by RDN - this is being set up at the moment, to replace a former registry that was out of date !. In the shape of the UK DNER infrastructure, such a registry will be more useful in the next year or so for the various DNER projects - where sometimes these details are difficult to get hold of.

And of course, a list of web sites - is just that - a list of URLs.  Z39.50 as we all know is another matter - port numbers, database names, record formats, search attributes etc etc.  But I agree - the other metadata points you mention are very useful as well - what is the database about - medical, scientific, community information etc.

Rob


-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
  Rob Bull    bull@crxnet.com         Crossnet Systems Limited
  tel +44 (0) 1635 522912             Unit 41 Bone Lane, Newbury
  fax +44 (0) 1635 522913             Berkshire, RG14 5SH, 
 United Kingdom
-+-+-+-+-+-+-+-+-+-+-   A member of the DS Group  -+-+-+-+-+-+-+-+-+-+-+-
  ----- Original Message ----- 
  From: Mike Taylor 
  To: quinn@indexdata.dk 
  Cc: www-zig@w3.org 
  Sent: Friday, February 22, 2002 11:47 AM
  Subject: Re: Z39.50 on the web (and in print)


  > Date: Thu, 21 Feb 2002 14:30:28 +0100
  > From: Sebastian Hammer <quinn@indexdata.dk>
  >
  > One concrete piece of criticism is the lack of an automatic way to
  > discover Z39.50 hosts, which I'll pass on to the ZIG. It might be a
  > good idea to think of some protocol-level way to access databases of
  > targets -- such as Z39.50, LDAP, or even "something XML".

  Hmmm.  (I notice that this seems to be my initial response to most
  issues these days :-)

  Philosophically, what use is access to a list of Z39.50 servers?  What
  can you do with them apart from pick one at random and connect to it?
  How would that be any more useful than (say) a list of web sites?

  What's needed is (of course) some metadata concerning each Z39.50
  server that tells the potential user _what_ is there (e.g. what
  application domain it serves, how many records it has) as well as
  _how_ it's accessed (attributes to use in searches, etc.)

  That sounds to me like a job for what we Z39.50 types call a
  "profile", but which no doubt has a different name out there in the
  wider world -- maybe an RDF schema or something like that?

   _/|_ _______________________________________________________________
  /o ) \/  Mike Taylor   <mike@miketaylor.org.uk>   www.miketaylor.org.uk
  )_v__/\  "Welcome to Mousebat, Follicle, Goosecreature, Ampersand,
  Spong, Whapcaplet, Looseliver, Vendetta and Prang ..." --
  Monty Python.