RE: Webfacing reactions from users

> From: Vern Hamberg
> 
> Now, it _is_ possible to create an appropriate tab order and even map
some
> key combination to the "hook" key. And with Change and various Key*
events
> in textboxes, one could get a lot of what you say, so I'm with you
there.

Vern, we do most of this already.  PSC/400 is designed from the ground
up to support as cleanly as possible the 5250 interface.  We even have
support for a quasi-field-exit key (even though it's impossible to
COMPLETELY emulate it in the browser because it would require
implementing keyboard scan codes which currently no browsers support).
Pure cursor position keys are a little harder, and we haven't yet gotten
a firm requirement for them, but they can still be done.


> I think, however, that this kind of work is rarely done, esp. by
> Windows-minded folk trying to replace that "antiquated" interface with
> something better (being translated to "prettier") - unfortunately, the
> better is usually tied to graphic methods, hence to the mouse, hence
to
> the eye, hence slow.

Our goal is that while you can indeed "prettify" the interface (with the
proper use of style sheets, our applications look like they are written
for the web), it should still perform like the original, so that someone
with 15 years of experience on your application doesn't have to
completely relearn their keystrokes.


> So let us all do better than this. I assume Java and other tools can
> specify, for one thing, the order of things when you tab. But, again,
I've
> seen a lot of pages that are a mess in this regard. Add tabs also
include
> the stuff that is not directly on the page itself.

As I said, we ALREADY do this.  The browser can be used as an input
device for all but the most demanding of data entry tasks.  Yes, some
people don't understand that, but WE do!

Joe Pluta
Pluta Brothers Design, Inc.
http://www.plutabrothers.com