Return-Path: <jhawk@MIT.EDU>
Received: from grand-central-station.mit.edu by po11.mit.edu (8.9.2/4.7) id TAA20979; Thu, 4 Apr 2002 19:05:51 -0500 (EST)
Received: from manawatu-mail-centre.mit.edu (MANAWATU-MAIL-CENTRE.MIT.EDU [18.7.7.71])
	by grand-central-station.mit.edu (8.9.2/8.9.2) with ESMTP id TAA15021;
	Thu, 4 Apr 2002 19:05:22 -0500 (EST)
Received: from multics.mit.edu (MULTICS.MIT.EDU [18.187.1.73])
	by manawatu-mail-centre.mit.edu (8.9.2/8.9.2) with ESMTP id TAA08632;
	Thu, 4 Apr 2002 19:05:22 -0500 (EST)
Received: (from jhawk@localhost) by multics.mit.edu (8.9.3)
	id TAA15201; Thu, 4 Apr 2002 19:05:21 -0500 (EST)
Date: Thu, 4 Apr 2002 19:05:21 -0500
From: John Hawkinson <jhawk@MIT.EDU>
To: Alex Coventry <alex_c@MIT.EDU>
Cc: sipb@MIT.EDU
Subject: Re: [orbitee@MIT.EDU: Re: SIPB access to library catalog.]
Message-ID: <20020405000521.GY8045@multics.mit.edu>
References: <200204021546.KAA06286@m2-225-8.mit.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <200204021546.KAA06286@m2-225-8.mit.edu>
User-Agent: Mutt/1.3.25i


I think this is definitely something we should pursue.

Personally, I think we would be _much_ better off with a view into
their oracle server than we would if they exported the data to
us... MARC is not an API, it's a data file format.  We'd still need to
import that data into a database of some sort, and it seems like we'd
be building a duplicate infrastructure for no gain.

I don't think you should encourage her to give us a raw data dump
if there's a possibility we can talk them into oracle access.

It would be good to understand their concerns. Perahps it would
make sense for us to colocate a machine next to theirs that
all the oracle transactions go through, so it can be firewalled (or whatever)
from the rest of the world, or so they don't have to have a publically
beatable TNS listener, or whatever.

I also think that when people want information like loan status, it
would really be extremely lame for them to see a day-old view when
we could have them looking at a current view with a little more advocacy.


As for how hard this is, I can't really imagine it's more than a days
work to produce a simple-but-lame interface that lets users query
for title/author/etc.

I mean, you know, the moral equivalent of:

	read author
	echo 'SELECT * FROM LIBRARY.CATALOG WHERE AUTHOR LIKE'"'$author'"\; |
          sqlplus
	exit

Though of course you wouldn't write it in the shell and you'd enforce
metacharacter checking.

--jhawk
