Re: Refocusing on HTML

Greg A. Smith (gasmith@advtech.uswest.com)
Tue, 03 Sep 1996 11:26:08 -0600


Message-ID: <322C6A30.60B5@advtech.uswest.com>
Date: Tue, 03 Sep 1996 11:26:08 -0600
From: "Greg A. Smith" <gasmith@advtech.uswest.com>
To: Lee Daniel Crocker <lee@piclab.com>
CC: www-html@w3.org
Subject: Re: Refocusing on HTML

Lee Daniel Crocker wrote:
> 
> On the one real HTML issue on the table (tabular <select>),
> I find the proposal not clearly necessary enough to justify
> the incompatibility and burden of implementation, when the
> function could already be better served with Java.
> 
As I have mentioned in some of my other replies, this probably
wouldn't be an issue, if SELECTs had been implemented in Windows
as it had been in UNIX.  I hope that my responses to other messages
in this group have made clear why this is so critical for database
applications on the Web and why simply proposing Java as the 
solution to every problem isn't the answer.  But in case it hasn't,
I'll try to clarify this again.

With HTML forms all of the functionality one needs to create 
enterprise-wide, platform-independent, relational database 
applications is there except that the SELECT element can't be 
made to display multi-column data on Windows platforms.  Since 
Windows is an essential platform in most organizations these days 
for enterprise-wide applications, that means that instead of 
building a simple, cheap, easily-to-implement solution using HTML 
forms, developers have to look for alternatives.

Java is one of the alternatives.  But as soon as we leave the well
understood, easy-to-implement world of HTML forms and CGIs, our 
development costs, time schedule, uncertainty and risk all take 
a big jump. Managers are not eager to enter the unexplored territory
of bleeding edge Java development.  What usually happens is that the 
project gets scaled back to a mostly read-only, non-interactive web
page, or a proprietary front-end such as PowerBuilder is choosen at
much higher cost (with licensing fees, tech. support and training 
costs, etc) for a small subset of the enterprise on a single 
platform, or the project is cancelled altogether.

All for want of a multi-column select.

====================================================================
Gregory A. Smith
303-541-6006
gasmith@advtech.uswest.com