RE: Web enablement options (previously private)

Nathan,

Been following this thread, somewhat...  See inline...

jt

| -----Original Message-----
| [mailto:midrange-l-admin@midrange.com]On Behalf Of Nathan M. Andelin
| Sent: Friday, January 04, 2002 12:46 PM
|
|
| From: "Adam Lang" <aalang@rutgersinsurance.com>
| > I agree.  Your RPG program SHOULDN'T write the html.
| > That would mean everytime you want to redesign your pages,
| > you need to recompile.
|
| Web development with ILE has moved past the Dark Ages, Adam.  I'd
| encourage
| you to look at my Relational-Web product, for details.  The HTML is in one
| file.  The RPG in another.  Change the HTML as much as you want.
| It likely
| won't affect the RPG program.

I really like your ideas about separating the presentation from the logic.
(This is based more on your discussion, rather than looking at your
website.)  I recall someone referring to a webpage that showed some
outstanding performance, as well.  Pretty hard to combine elegant solutions
with improved performance, but you seem to have found the ticket...  (Just
as you told Walden) you sure have a keeper, and wish you great success in
the market.

|
| > It also means your RPG programmer has to be your web deveoper
| > too, or at least have extremely good communication with him.
|
| It can be as simple as having the page developer tell the application
| developer that the HTML is ready.  The application developer takes it from
| there.
|
| I'll borrow a quote from Walden:
|
| "From CS we went back to host controlled, web browsers.
| Sure, they're graphical and 'pretty' but at the end of the day,
| they're host-controlled green-screens that aren't green."
|
| In my experience, application developers have a much better handle on the
| requirements of program flow and "user interface" control logic.  Web
| developers usually have a better handle on page layout.  My
| preference is to
| have the application developer control the user interface, not visa versa.
|

However, small point of disagreement:  In my experience this issue of
control is best resolved by viewing it from 3 separate views.. you've left
out the view of the user, themselves.  So to pick any one, to control the
others...  Well, no matter which you pick, you're gonna have the tail
waggin' the dog.

I agree that app developers *generally* have a better handle on the logic
end, but at the same time, they frequently have the *worst* handle on what
the user needs to do their job.  This is the proof in the pudding, whether
the user can accomplish their job better, right?  So frequently the person
doing the webpage layout is in a MUCH BETTER position to act as the
middle-man, to see where computer logic can assist the end-user.

Picking the app developer to control this dynamic doesn't make any sense, to
me anyway.

But to have the person doing the page layout control the UI makes as little
(or less) sense, again IMHO.  Obviously the user can't control the UI,
however they would (presumably) have the best handle on what they need to do
their jobs better.

This is nothing more than the age-old problem of how you "negotiate" the
optimal systems design.  Some believe it's best to have one of these three
elements in the system control the thing, but that hasn't been my
experience.


OTOH, the best part about your product is that it allows a lot more
flexibility.  One GREAT thing is that you're product can be used by a single
individual, in a small shop, or it can scale to where the UI and app logic
can be controlled by two entirely different departments, in a large
enterprise.  It is my understanding that this is commonly done, as well...
So your product seems to scale from the smallest to the largest, quite
well...!

<snip>