RE: RPG, 10 years from now

>they have to keep their plans a secret so the competitors dont steal their
ideas?? I dont think so.

I think you mis-read my sentence. The sentence was meant to invoke the idea
of IBM not wanting to commit to any particular implementation of the next
generation of UI in RPG (e.g. so they can change it once it goes live), not
that they are trying to protect secrets.

>More public input might be helpful.
100% agree with you here.  The proof is in the pudding as to why languages
like PHP, RubyOnRails, Java, etc are so successful.  They have tremendous
community support, and that support comes from geeks getting involved with
the compilers and supporting technologies/frameworks that are more often
than not open source.

>Aaron, why does a programming language have to be concerned with the
devices attached to the computer?  That is what APIs are for. 

Man Steve, I feel like you are taking my statements and running with them
(and you are running with them in the wrong direction). I didn't say that
the RPG program would have to know it is talking with a blackberry, I said
it would give us the ability to talk to a blackberry. Note that this can be
done easily right now with RPG and CGI, but it would be great to make it
even easier with some technology specific support built into the compiler.

>C# knows nothing about talking to a blackberry.  why should RPG? 

It sure does, or rather the programmer that implements the code does. A C#
programmer needs to know the end "screen" that will be used to display the
data so they can appropriately format the data. Screen real-estate is very
important to have in mind as one develops code. Sure you have frameworks
that supposedly do this for you, but I have yet to see a good example of how
you can take a ui for a browser and make it work on a device xyz.  Your C#
application could code to the lowest common denominator (say blackberry) and
never include more than three input fields or a table with more than 10
records, but that is a waste of space for the person who is trying to use
the exact same program from the browser.

I know where you are coming from and I was of the same mind and thought it
was great that frameworks like JSF allowed you to "plug'n'play" different
render kits so you wouldn't have to rewrite your code when a different
display was used, but it is so uncommon that you could use the same ui for a
browser and handheld that it doesn't justify the reason for the render kits.

>It needs namespaces, function overloading, try finally blocks, classes,
delegates for event programming, data struct member functions and of course
managed code for memory managment. 

I would love to see a handful of these in RPG.  The thing that the compiler
people need to be careful of though is to not make a purist language.  By
that I mean keep it productive and business minded.  A comparison I like to
make is with PHP.  PHP has OO capabilities, but not everything is based on a
"root object" like Java is (i.e. http://java.sun.com/j2se/1.5.0/docs/api/).
PHP instead chose to implement a feature rich set of compiler functionality
without taking such abstract routes as Java (e.g.
http://www.php.net/manual/en/funcref.php)

It has been at least a year since I have played with C# so I can't comment
on the latest developments or whether it follows a purist or production
minded route, or both.

Aaron Bartell


-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of Richter,Steve
Sent: Friday, May 26, 2006 10:22 AM
To: RPG programming on the AS400 / iSeries
Subject: RE: RPG, 10 years from now


-----Original Message-----
From: albartell [mailto:albartell@xxxxxxxxx]
Sent: Friday, May 26, 2006 10:01 AM
To: 'RPG programming on the AS400 / iSeries'
Subject: RE: RPG, 10 years from now


> >I doubt if it will be enhanced any further in 10 years, but it will 
> >still
>>be around. A compiler written today will still work in 10 years.

>To the contrary, it will be changed more in the next 5 years than it 
>has in the past 15!  I am working on a blog entry for imho.midrange.com 
>on this exact topic.  I attended the rpgworld.com conference this past 
>month and had the chance to hear George Farr discuss "potentially" 
>where the RPG language could be going. Of course he has to be vague 
>because they haven't completely and publicly committed to the feature 
>list, but what he was brining up got me excited!

they have to keep their plans a secret so the competitors dont steal their
ideas?? I dont think so.  The enhancements to CL were kept secret and the
result were subroutines when what was needed were CL procedures.  I respect
the people who work at IBM, but their record of creating great programming
languages is not very good.  More public input might be helpful.


>My favorite bit of info he shared was their thought process of 
>determining how the next versions of RPG will interact with the user. 
>Specifically speaking they were mentioning new OPCODE's like EXUI (vs. 
>EXFMT) which would give us the ability to "execute" a ui that was a web 
>page, or maybe a WML app for a blackberry.

Aaron, why does a programming language have to be concerned with the devices
attached to the computer?  That is what APIs are for. 

the great success .NET is having shows how run time frameworks and
programming languages should work together and what features they should
have.  RPG needs to be like C#. It needs namespaces, function overloading,
try finally blocks, classes, delegates for event programming, data struct
member functions and of course managed code for memory managment. 

C# knows nothing about talking to a blackberry.  why should RPG?   

-Steve




--
This is the RPG programming on the AS400 / iSeries (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxx To subscribe, unsubscribe, or
change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives at
http://archive.midrange.com/rpg400-l.