Re: Proposal to deprecate OPAC Record Syntax and XML

Ray wrote:

> Thank you both for bringing this up. I would not like to see this item
> discussed without input from people who have a stake in this, and I don't
> think either of you is coming to the meeting.

No, I don't get to go on jollies to America, however, I hope to make the
next one in England.

Leif Andresen wrote:

> This doesn't make existing implementations of 102 illegal, but is a signal
> to new implementors of Z39.50.

I realise it doesn't make it illegal and I don't think it will ever
disappear now it is here, but I have found it useful for retrieving
circulation info from INNOPAC targets. I've had mixed success with
other makes of target and not mentioning any names, one well known
make of target sends back opac records that cannot be decoded (bad
BER encoding I guess.) Another type of target sends back a opac record
but sometimes not all the holdings statements are there -- in this case
it turns out I can get all the circulation info from a Full MARC
record anyway.

So, we've had mixed success with OPAC, but where it works it seems
to work well and gives us the data we need -- ie location, class mark,
and whether or not it is on loan.

> This proposal is motivated by the enthusiasm of the proposer (Leif Andresen)
> for the Holdings schema and his desire to see it widely implemented.  Perhaps
> as an alternative to deprecating the OPAC definition, we might consider
> instead adding a note within that definition, pointing to the Holdings
> Schema, so that people who pick up the standard and see the OPAC format will
> at least be aware that the newer and more comprehensive definition exists.

If the zig thinks the new holdings schema is better then so be it --
we just have to work with whatever the targets want to throw at us.

Now to digress. It seems that in the past the only rec. syn.
that an origin had to know how to parse (for better or
worse) was a MARC record. The future is now looking like a
combination of MARC, GRS-1 and XML. I don't mind GRS-1 as
that is easy to parse, but XML is far from trivial and I
would say is over the top for presenting the data structures
we need, eg Explain-lite. So, it's looking like I'm going to have
include an XML parser in my origins -- I'd rather not. Could
we perhaps use one of the simplied forms of XML that I've been
hearing about (if we go this route) -- all we need is a
heirarchical tagged structure -- we don't need free form text
markup facilities. Or do we?

Regards,

Ashley.

-- 
Ashley Sanders                                a.sanders@mcc.ac.uk
COPAC: A public bibliographic database from MIMAS, funded by JISC
             http://copac.ac.uk/ - copac@mimas.ac.uk

Received on Wednesday, 22 November 2000 10:36:31 UTC