On 6-Nov-07, at 8:25 AM, Joe Pluta wrote:
I re-read that post several times, and I swear I don't see where I
said that
you said PHP was better.
That was the whole point I was trying to make. I made a simple
comment about my personal opinion on EGL and you launched into a
tirade about PHP. You are correct - I did not mention PHP - why did
you?
And I still fail to see in what way an RPG-CGI app is any more open
to the internet than WAS or any other option.
I will state simply some of the basic facts that form the foundation
for my current opinion. I reserve the right to change/adjust that
opinion as time unfurls.
Actually I will start with my opinion first.
Jon's Opinion: EGL is a very good product with a lot of potential.
It will fit well for a lot of shops. I have every intention of using
it and promoting it to those of my customers where it makes sense.
Why I don't think it will take over the world:
1) I have many customers where .Net is the selected "solution". The
System i folks were too late to the party and .Net has the political
mind-share. EGL can't be "sneaked in" because you have to pay for
it. In these shops even getting WDSC in the door can be tough (in
one case impossible) because of the WebSphere label. The .Netters
see any WAS intrusion as an attempt to displace their beloved .Net
(MS have trained them well) and will fight _anything_ that smells of
WAS.
2) RPG-CGI can often be sneaked in (as sometimes can PHP albeit for
different reasons) and with a number of customers such action has
been the foundation upon which the dominance of .Net was successfully
challenged. Once this has happened then tools like EGL become viable
options - but not until the door is open.
3) I suspect that IBM will price EGL out of reach for many shops. I
just don't see most of my customers paying $1,000+ a seat let alone
the $2,000 figure that I've also seen floated. If the fee is that
high - why not go for a third party option?
4) IBM tend to be hot-and-heavy promoting a product for 12 - 18
months and then moving on to the next big thing. In a market where
many have yet to move to RPG IV after 12+ years that is too short a
runway to get momentum.
5) Very few of the senior folks in Rational have a clue what a System
i is. This is actually true of the whole of Software group. The
existence of the Java toolbox gave them "free" access to i5 function
- so the tool can look a lot more "native" than it otherwise would.
Even though you and I know it makes no sense, many people will balk
at the fact that it can't generate RPG. As I told George Farr
yesterday when we were discussing this - it shouldn't matter - but
the fact that it can generate COBOL says that non-OO code gen is
acceptable/desirable. It is natural for an RPGer to feel more
comfortable with a tool that generates his language. The fact that
COBOL gen is probably more an accident of history than design is not
relevant to them.
6) The notion that you can use EGL and forget the Java (or the AS for
that matter) is according to people who's opinion I trust a complete
fallacy. Bleddyn Williams for example has stated publicly that it
was only their vast knowledge of Java and WebSphere that made EGL
viable for them. It is hard enough debugging problems with an AS
when you know Java inside and out - I simply find it hard to believe
that the tool makes it so transparent that you can simply use it
without any Java/JSP/JSF/AS knowledge.
7) I believe in choices. I don't believe that EGL or indeed any
solution works for everyone. I do believe that getting the customer
moving on the web front is critical and I make customer
recommendations as to how that should proceed based on that. Since I
teach everything from RPG-CGI to PHP to (well not EGL yet) I have no
axe to grind on one solution vs another.
8) I have been in the midrange industry for over 38 years - my
opinions are in part based on that experience.
Jon Paris
www.Partner400.com
www.SystemiDeveloper.com
midrange.com
