Re: EGL and RPG - The Fastest Path to the Future

"What we REALLY need is a technology so we can put a nice GUI on our EXISTING apps."

But the reality is that even if you did, would you *want* the UI to remain the same? So the screen is a Windows Modal window, then what? You can do that with HATS using the Rich Client. I have several examples of HATS transformations (with a few tweaks) of a 5250 application called CIMS at http://www.i5gui.com but you and I both know that a nice UI on a 5250 screen is still limited. So your idea of having an EXDLG opcode in RPG that outputs to an Rich Client (or HTML) application still leaves us short of the mark.

OK. You want *more* than just executing a 5250 screen format in HTML or a Rich Client. My guess is, as soon as you make the decision to add new op codes and capabilities, you require new skills (like HTML and new UI design skills) and you start redesigning your existing 5250 app. And, also, since you now have a new client-side expression (HTML or Rich Client) you'll need new tools to design them. Now your "easy" path to an RPG web solution is gone. Very soon you have a tool kit that looks very much like RBD. Sure you may continue to write some base code in RPG, but you'll be dealing with Javascript, JSON, REST and a bunch of other web technologies before you know it. It would be great if all of those things could be written in RPG, but I am not sure that RPG even with all of it's capabilities could morph into a "one size fits all" business logic/web/rich UI language.

It *sounds* like what you want is RBD/EGL to be RBD/RPG but with RPG being now a "meta" language being able to handle business logic as well as UI logic (scripting, presentation, AJAX). Seems to me that EGL is a better fit as the meta language in such an environment. However, I have tried (and liked) EGL, so I am a bit biased now. And, if you are NOT going to have RPG be the meta language for your top to bottom application development, what would you use? RPG for business logic and HTML/Javascript for client presentation? Well, there you go: might as well jump into EGL which will make that transition more productive.

Pete

john e wrote:

Joe,

I think you have to ask yourself "what real problem do we have now that we have to solve".

I know for a fact that it is not "we have one source in which we express the UI and the business logic".

A real problem we have is that our apps still look like they were built in the 70's. And there is no "simple" way to give our apps a modern interface. We could'nt even use EGL for that.
Because EGL (and browser apps in general) is client/server. That means it can only be used for new apps. Our existing apps are all monolithic, and in fact are the client (where the terminal is the server). It is too much of an investment to re-build these apps because we have to factor out all business logic so that it can be called from the server. In practice, this is impossible. Not impossible, but very expensive and risky. So it's not going to happen. Because if this was an option it already happened with for example the ASNA products or whatever.

What we REALLY need is a technology so we can put a nice GUI on our EXISTING apps. And this is possible, but we need a new technlogy for this. One which merges the event-driven approach of GUI with the host-driven approach we use today with 5250. So we can take our existing app, simple replace the EXFMT with a EXDIALOG (which presents a nice looking windows MODAL screen). And so we can EXTEND our monolithic apps with event-driven functionality if that's what the user wants.

But instead, IBM gives us..... EGL!
Bummer


So apparantly we have to do it ourself. Like X-Windows (which is a mess by the way) for Unix, which was also not created by AT&T but the Unix user community.

Date: Wed, 18 Jun 2008 11:16:32 -0500
From: joepluta@xxxxxxxxxxxxxxxxx
To: egl-i@xxxxxxxxxxxx
Subject: Re: [EGL-i] EGL and RPG - The Fastest Path to the Future

Aaron Bartell wrote:

I am not trying to start a heated debate, but isn't this second best?

Wow, no Aaron, it's not. It's absolutely the best way to develop applications. In most of my reading, the idea of distributed or collaborative development is pretty much the Nirvana of application design. If I read you right, you're advocating the old school of architecture where one person owns the application from user interface to database design.

Why
should I need two completely different people (read knowledge sets) involved
to create a single business application?

I don't. I could have done it myself, if I wrote the rich client part. I mean, it's just EGL. But at the same time, the division of labor - allowing different programmers to focus on different tasks - has long been recognized as a productivity enhancement. In this particular instance, I didn't have to worry about images or CSS or the iPhone - I just wrote business logic.

Seems like we are going backward in efficiency.

Nope. Disagree. I showed that when I responded to Nathan's application. I put together a nice little screen in minutes. It was actually easier to do then it would have been with SDA, since EGL painted not only the fields but also the prompts.

If that is the case I would say the technology has become
too complicated to quickly/effectively build business applications.

Well, I think a lot of people would agree that technology has gone past the day when one person could easily run an IT shop. But that's a primary argument *for* EGL - you can write all the interfaces (thick, thin, even 5250) with EGL, and all your business logic in RPG. One person can still run the shop.

Again, I understand your desire for a single UI that runs from RPG, but we don't have one, Aaron. So until we do, what do you suggest developers do?

Wait?

I guess
this points back to my opinion that people will find greater efficiency by
dropping RPG if EGL is the facing language because they have one knowledge
set to maintain.

Fair question?

No, not fair at all. We've always used a couple of different languages. We use DDS and CL and RPG. Thiis argument that EGL will force RPG out has no basis in fact. I mean, it's your opinion, but it seems awfully far fetched and it's a terrible, terrible reason to try to keep EGL out of the hands of developers. I mean, if you don't want to use EGL, fine. but don't badmouth it because you think it's going to displace RPG. Heck, Microsoft's been trying for decades, and they still haven't done it.

Joe
--
This is the EGL on and around the IBM i (EGL-i) mailing list
To post a message email: EGL-i@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/egl-i
or email: EGL-i-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/egl-i.

_________________________________________________________________
Check je Hotmail nu ook op je mobiel!
http://windowslivemobile.msn.com/BrowserServiceHotmail.aspx?lang=nl-nl