Re: WebFacing - Is it ready for primetime?

We have used WF/HATS on a number of applications. We needed to get
them running on a zero-interactive box right away. Performance is
acceptable for inquiry applications. Not so much for heads-down data
entry. We are looking at rewriting those in JSP/Servlets. We've
already had great success with writing JSP/Servlet apps that do
high-volume data entry. Once the keyers got used to some of the new
keys, they were able to tear through the browser-based app as quickly
as they did on their old 5250 app.

The main thing we've run into with WF/HATS is the "futz factor." If
you have pretty good developers that can write browser based apps, you
don't want to spend too much time tweaking the settings in WF and/or
HATS. You can spend hours if you feel like it. Let it get you into
the browser quickly, then spend your time separating your presentation
and business logic (I'm assuming they are inextricably wedded,
otherwise it's pretty simple to wrap RPG/CL in stored procedures and
call them from Java servlets and not bother with WF/HATS) and moving
to JSP/Servlets. Or you could move to .NET, or PHP, or whatever.

One thing I have played with but not yet tried in production is using
HATS to turn a 5250 app into a web service. I did the lab at COMMON,
but have not done anything else with the knowledge. Only took about
90 minutes to do the lab, so it might not be too terrible of an idea.
At the very least, it would force us to write a decent JSP and
Servlet, and then we could swap out the back end later. Neat concept.
Our WebSphere rep is on my back to try this in production.

On 8/8/07, Shannon ODonnell <sodonnell@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote:

Hello List,



I have a couple of clients who currently use Seagull's JWalk product.



They are happy enough with it, I guess, except that they dislike greatly the
update process of the activex component of it as that clogs their entire
network while it is happening.



The biggest problem they have with it though is that they now have an
antagonistic relationship with their old vendor or their primary
application, which JWalk is used to GUI-ize, and they want nothing more to
do with that vendor.

To complicate things, that vendor has the developer client to make changes
to their Jwalk screens and to add new Jwalk panels, etc.. for new screens.
And that's not a good thing because there is no chance at all of these

Clients working things out with the vendor and so soon, jwalk apps are going
to start failing as the client has other programmers create new 5250 screens
and modify existing screens, etc..



I personally would like to take them the WebFaced route instead for a
variety of reasons:



- Free - No licensing fees like they have with Seagull

- Easy to convert existing and new screens

- Customizable

- No client software (Which is a big big plus for them)







My question for the list is: For those of you who use WebFacing, is it now
stable enough and robust enough to be used for daily, heavy volume,
production use? I personally have done little more than play with it for
the last several years. The last time I really spent a lot of time with the
product was about 5 years ago and at that time, it wasn't the most stable
tool in the world and it required you to do all they manual "fixes" and
coding of config files, and such, to make it work (even to get it to compile
sometimes).



Does it now support most DDS keywords (Print supported yet?), and will it
crash under heavy volumes? Is it still buggy?



Your experiences with the product would be most appreciated.



Thanks,



Shannon O'Donnell







--
This is the Websphere Development Studio Client for iSeries (WDSCI-L) mailing list
To post a message email: WDSCI-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/wdsci-l
or email: WDSCI-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/wdsci-l.