Data Goob wrote:

>The real question is why are you still using 4GL at all?
>
>
Because there's lots of legacy code and maintenance work to do. And
because, in spite of it's many shortcomings, it's a very good and simple
language for business apps. Unfortunately it lacks the expected
characteristics of a fashionable development tool:

* Traditional GUI & Web interfaces
* IDE
* Additional instructions to make simple tasks like sending e-mail,
parsing XML, manipulating files, etc..
* Object orientation

>Why would you use a payware version of 4GL when you can get an open source version?
>
>
Because the open source project doesn't give the same assurance that a
brand like IBM does and because they don't have the funding to invest in
product development that should be done.

>Who will be around to maintain the code if you depart the project?
>
>
Unlike Java and other technologies in fashion, 4GL is very easy to
learn. The problem with old projects is not the programming language!
Just look at the number of J2EE orphan projects in the last years!
Picking up a 4GL project without documentation it's easier than a J2EE
project in the same situation. There's many more ways to do it wrong in
J2EE than in 4GL! ;-)

>4GL is interesting and useful but would you really deploy it on new projects?
>
>
Ask the many customers that keep having big teams of programmers in 4GL.
They sure don't do maintenance only!

And I hear you asking: "Why do you resist converting to EGL, that seems
to have all the things you say 4GL needs?"

Personally, I don't resist it (although it still lacks some things). Our
company is very interested in this new product from IBM. But I know many
who will resist, for many different reasons, and IBM could, easily,
create a product called the "New Informix 4GL" with the work that's
already in place (the I4GL translator and the EGL compiler). So,
*everybody* would be happy.

Fernando

--
Fernando Fernandez
http://www.moredata.pt

sending to informix-list