Message-Id: <199603210146.TAA23518@library.wustl.edu> Date: Wed, 20 Mar 1996 17:42:36 -0800 From: Andrzej Kowalski <mailto:andrzej@DINGO.COM> Subject: Re: Texpress (was Re: WEB catalogs that handle MARC records) To: Multiple recipients of list WEBCAT-L <mailto:WEBCAT-L@WUVMD.WUSTL.EDU>
Since Kevin Marsh's questions about Texpress were posted to this list, I will also answer them in that forum. If people feel this is not appropriate, Kevin and I can move this discussion off line.At 14:13 03/20/96 -0600, you wrote: >>As a software vendor, my response to this question will be targeted towards
>>my own products, but I will try to be objective as possible in the
>>circumstances.
>
>And I think you did a pretty good job.
Thanks
>Let me ask you a couple of questions
>specifically about your product.
>
>1. How much control can the library exert over the look of the HTML you
>generate on the fly? Can they include a link back to their home page?
>Background images? In-line images? Links to a digital version of the
>cataloged item, perhaps using the 856 field?
You have complete control over the look and feel, images, formatting, links etc. Any HTML tags that are understood by your browser will work. Texhtml does not care whether it is HTML 2.0, 3.0 a frame or whatever.
The paradigm we have adopted is essentially that of the report writer that you might use with any common database product. To us, each HTML document produced by Texpress consists of a header, body (usually for the data returned by the engine and a footer. At its simplest you design a standard WWW form for querying the database which has our cgi-bin program Texhtml as the form action. You design an HTML template for the header, body and footer and a script to generate the Texql query language and the Texhtml software takes care of the rest. The Texql script may be generated without any programming knowledge. Here is a simple example based on a database of 3 million business records which took about 30 minutes to put together:
SCRIPT: select all from northwest where true and name contains '#name#' and ubi=#ubi# and bizid=#bizid#
This script is actually a simple text file. The Texql syntax is very similar to SQL. Texhtml will replace the #field# placeholders with the values a user provides in a Web form, execute the query and return an HTML document to the user based on the following header, body and footer components:
HEADER DOC: ----------- Content-type: text/html
Northwest Location and Investigation Services Search Results< /title></head> <body bgcolor="FFFFFF"> <H3 align="center"><p> Northwest Location and Investigation Services Search Results</H3> Selecting a UBI number will retrieve all names for that UBI<P> <HR size="3" width="60%"><br clear=all> <H3>#TexRowHits# Matches</H3> <UL> Again, we have a simple HTML document. The #TexRowHits# is a placeholder for an inbuilt Texhtml function that will return the number of matching records. BODY DOC: --------- <LI><b>#name#</B>, <A HREF="/cgi-bin/texhtml?form=nw/nw.all&ubi=#ubi#">UBI</A> <b>#u bi#</b>, BIZID <b>#bizid#</B> Another simple HTML document which contains the #fields# we want to display to the user. Texhtml will automatically instantiate these values once for each matching record returned by the database. The HREF tag calls our cgi-bin program texhtml which would then generate another query based on different parameters and cycle through the scripts, headers, bodies, footers etc. FOOTER DOC: ----------- </UL><br clear=all><hr size="3" width="75%"><P> <h4 align="center"> <A HREF="/bin/texhtml?form=nw/nw">Search Again</A> </h4> </body></HTML> Simple HTML document. We also offer a post-processing option by way of a filter. You may define a filter i.e. any executable program e.g. Perl script, which is invoked, if it exists, after the above document is assembled. This gives you the opportunity to manipulate the document, perform validation, logic - whatever. We have further extended this paradigm to support full database inserts, updates and deletes using standard Web forms. So you can essentially have a complete online DBMS handled entirely with Web forms. Demo databases and full product information is available at our Web site: http://www.dingo.com > >2. Have you implemented Z39.50 compliance and tested inter-searchability >with DRA or other Z39.50 servers? Which Z39.50 clients have you used for >testing? It is under development. Expected release is 2nd quarter 96. > >3. Can you give an example of what a typical Web site implementation of your >product would cost? Say for a site that got less than 5,000 searches per day. You would probably only need a 2 concurrent user Texpress licence with Texhtml. You would use one licence for Texhtml (i.e. Web queries) and the other for applications development and db admin. 2 concurrent user Texpress costslicence $4,000.00 and Texhtml costs $1,600.00. It has been our experience that a 2 user licence c<!-- body="end" --> <p> <ul> <!-- next="start" --> <li> <b>Next message:</b> <a href="0147.html">mailto:staff@CVCOM.COM: "A Place to visit"</a> <li> <b>Previous message:</b> <a href="0145.html">Elizabeth Roderick: "Library of Virginia Digital Projects"</a> <li> <b>Maybe in reply to:</b> <a href="0144.html">Kevin C. Marsh: "Texpress (was Re: WEB catalogs that handle MARC records)"</a> <!-- nextthread="start" --> </ul>