[Ex-plain] ZIG presentation

Robert Sanderson azaroth at liverpool.ac.uk
Fri Apr 5 18:29:04 CEST 2002


Overall it went quite well I think.  The following comments were raised:

Bill:   Should we have a Geographic Referrent to describe physically where 
        the server is?  

Juha:   Collection Level Description information would be very useful, 
        rather than just data about the server.

Ralph:  Access point 1 is an unnecessary barrier to implementation.  
        We require a search to look at the structure of the XML, rather 
        than the contents of elements or attributes.

Ralph:  Why have a database? Clients will just download all the records 
        and store them locally rather than updating on the fly.  OAI would 
        be a more appropriate transport mechanism.

Lennie: Why are you doing this at all? What's the point?


My (and others') responses:

1.  Network topology and physical proximity are often unrelated.  Servers 
may be physically distant from the physical items which the records in it 
describe.  This has little practical value.

2.  This is a Big issue which Explain-- shouldn't be concerned with 
-directly-.  On the other hand we have extensibility.  If people want to 
put in further collection level information then we can include a 
namespace which uses EAD or UKOLN's collection level schema or whatever is 
appropriate.  It's outside of our expertise to decide on the value of one 
collection level description or another.

3.  I agree.  There are two solutions:
     1. Change the record to put the information somewhere more easily 
        accessable.
     2. Change the attribute set to instead have an access point called 
        authoritative and has two values which will get hits, true and 
        false.

This also applies to thisServer and authentication.



4.  Databases allow for distributed solutions rather than everyone having 
a federated copy on their drive of all the records, which the client needs 
to sort through to find which they want.  In a world of Many Z servers, or 
more likely many servers between Z, SRU and SRW, this would potentially be 
an unbearable burden.  Also that z servers already do Z, not necessarily 
OAI.

5.  See above, our overview etc.


Rob




-- 
      ,'/:.          Rob Sanderson (azaroth at liverpool.ac.uk)
    ,'-/::::.        http://www.o-r-g.org/~azaroth/
  ,'--/::(@)::.      Special Collections and Archives, extension 3142
,'---/::::::::::.    Twin Cathedrals:  telnet: liverpool.o-r-g.org 7777
____/:::::::::::::.              WWW:  http://liverpool.o-r-g.org:8000/
I L L U M I N A T I





More information about the Ex-plain mailing list