Re: Trial for RDi SOA

Joe Pluta wrote:

Jim Steil wrote:

I'm not here for an argument. I'm here to try to learn what is out there and what EGL can do for me in my environment.

Sorry, Jim. My bad. I've been fighting the architecture fight for nearly ten years now; I get a little too defensive.

I know lots of people prefer it for business logic but it just doesn't feel right to me for web programming.

Agreed.

What do you see as the barriers for adoption of EGL? Here are my thoughts:

1. Cost of entry -> I'm not sure of the cost, is it $750/developer
2. Availability of information - 1 book available at Amazon at this time (one more available by pre-order)
3. Availability of playground. - Need to have access to an AS/400 with necessary tooling (apache at the least) to try it out
4. General developer mindshare - I read a number of trade publications and blogs, and the mainstream (not AS/400 based) publications don't seem to give it any attention

There are trials of EGL that are free. They run on Windows, require no additional installation. The RBD installation includes a self-contained, pre-configured test environment. You may not find books on it, but there are tutorials and information available, as well as distance learning classes.

The last issue, one of mindshare, is one I can't help you with. If you want mindshare, you're not on the i to begin with. You're using Microsoft or PHP or whatever the fad of the day is. I know that's not much of a response, but it's the truth: if programming is a popularity contest, the System i is the sitting out in the parking lot getting drunk. (But it's a heck of a lot more fun to hang out with <grin>.)

Compare that to Ruby on Rails (which I know little about, but hear about all the time)

1. Cost of entry -> $0.00
2. Availability of information - 200+ titles available on Amazon
3. Availability of playground - Any Windows, Linux or Mac computer
4. General developer mindshare - Can't avoid reading about it in the press

This is the interesting bit about fad technologies: they fade as quick as they rise. For example, with Ruby, you're already late to the party - it looks as though Ruby already peaked. According to TIOBE it's greatest popularity was late last year. I predict you won't see one successful commercial installation of a Ruby-based application for the System i.

Obviously the second scenario is going to rely on ODBC for i data access, but I agree that is too slow. Maybe IBM needs to focus on speeding that up and playing nicer with others. But I'll say that I agree with you, that EGL seems to be a great way to web enable i-based RPG business logic, and that it is capable of producing rich front-ends. But, you'll have to look to the vendors of other IBM i web toolkits to take up your challenge.

The point of my challenge was simple: to silence the naysayers who say that PHP is better or RPG is the way to go, many of them without even having looked at EGL.

I was a bit too strident, I agree, and I apologize.

Joe

And, I probably joined up on the list too late to see what was being thrown at you a couple weeks ago. But hey, you've helped me better understand what EGL has to offer, and it may be important for us down the road. Our ERP vendor has converted all of their old green-screen to a tiered implementation (front ended by green screen or Net.Data and now working on PHP front-end) and I assume that the business logic is available for me to wrap in EGL. When I get some time (yeah right) I hope to play around with it to see what they've exposed to me. I'm assuming that you can consume the services and get JSON data for Ajax requests and such.

Well that's about enough for one week, hey? Have a good weekend everyone....


-Jim