Xref: utzoo comp.databases:7927 comp.sys.mac.hypercard:5082
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!van-bc!ubc-cs!uw-beaver!milton!biostr.biostr.washington.edu!kraig
From: kraig@biostr.biostr.washington.edu (Kraig Eno)
Newsgroups: comp.databases,comp.sys.mac.hypercard
Subject: Re: Accessing an Oracle database from the hypercard interface
Message-ID: <11568@milton.u.washington.edu>
Date: 22 Nov 90 04:15:28 GMT
References: <499@n5pph007.UUCP>
Sender: news@milton.u.washington.edu
Organization: University of Washington
Lines: 25

>front-end on the Mac using Hypercard.  The database itself will be
>running on an HP unix machine, with the access from the Macs being
>through SQL*Net.

I cannot tell you how to solve your problem, but I can offer some hope in 
a general way.  We decided against Oracle because it was too expensive 
(not free, I mean) and too huge and complex on the Mac end.  I've rolled 
my own with the following pieces:  HyperCard (and SuperCard) running the 
MacTCP XCMD's to access a Sybase SQL database on a NeXT via TCP.  
Very similar to your situation but with different products.  If I click on 
a button which requires data, I start seeing results in well under 2 
seconds -- that includes composing the SQL command, sending it to the Unix 
database server, executing it, and getting the first row of results back 
in my HyperCard field. Small rows come back at a rate of about 3 per 
second, though that has not been optimized at all yet.  

So the problem is not inherent in the Mac interface.  There MUST be a 
solution.  If you can get Oracle data from a C program on your Unix 
machine, then I am sure that our method would work for you, but you may 
not want the pain.  (I had to write a TCP-accessible daemon on the Unix 
side which fields requests, does the queries, and sends the data back.  
Easy once you know how, but hard to figure out!)

Kraig Eno, kraig@biostr.washington.edu
"Problems generate new knowledge." -- M. Usui