Top 10 Things RPG Needs

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
I like all these ideas, but I think the "biggest" ideas in importance and maybe urgency are the CGI and FTP functionality, and the drop-SEU suggestion. The FTP functions Cozzi proposes do not look that far afield from the logic used for SQL API's currently. They also fit what I heard an IBM'er say in a seminar once about RPG, that they are looking to incorporate a lot of functions where currently the RPG has to rely on preparatory CL, or QCMDEXC. An example of this is the re-direct in the more recent releases for files to member names. With SEU out of the picture, companies/shops would be asking their developers what editors they want, and you bet GUI would be it. It's just inertia, but IBM has shown willingness to "shake up" things with its "green-screen" charges while providing a whole lot more development tools with V5 "at no extra charge".

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
I would opt for items that contributes the following attributes to programming code: 1). Consistancy 2). Intuitive 3). Helps that clearly explains how to do something. 4). And Helps that explains what features maybe more appropiate to use. 5). Productivity that helps to lower cost A lot of times, just adding a new feature adds to the complexity and to the learning curve for using these tools. I work hard to keep the applications I write for my users as simple to use as possible. You and IBM should have the same focus for us as programmers.

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
I've long been, for the most part a supporter of many changes to the RPG language. However, we all must face the fact that the numbers of RPG coders are growing smaller each year. Although sad to me, it's all right. We should be encouraging IBM to develop an entirely new language like Microsoft did with C#. Instead of adding C++ and Java to RPG and creating something many people do not care for or even want to use, we should want, no demand, a new universal language from IBM. Why are we limiting ourselves? We have always been leading edge types. So why all of a sudden are we afraid to let go of the past, including what I consider poorly thought out and designed tools such as ILE RPG and Code/400. We really shouldn't limit ourselves this way. Let's be the future and demand something new. Something that will be compatible with Windows, UNIX and whatever IBM has. Something that can generate code just as Graphical as Windows, as business like as RPG/COBOL,and is naturally Web enabled. Something other than JAVA and XML (I could care less what it actually ends up as). WE should insist on something better than what IBM has to offer period.

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Does anyone know the total count of RPG developers worldwide? (From System3x to the Iseries.)

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
In the intro to SunMicrosystems Java manual they state that one misconception is, "Java will become the universal programming language." So even Sun recognizes that JAVA is not the answer to many computing needs. It seems, we aren't following the ideas being presented here. If we want the AS/400 community to grow and prosper, we need something that can make even the Windows developers envy. All we have currently is ILE RPG/COBOL and CODE/400. We need to let go of the old and create the new if we are to have a common 400 community of the future. We need to get IBM to expand their mind set. As well as ourselves. I hope this helps you to understand where I'm going here. Either you want the 400 to survive or you don't. I do.

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Who cares what the compiler generates? Why put a limit on the future?We need new tools in the form of a new language. Something that will attract even the Windows guys. I'm talking about a real universal language that will work for everything a business needs. I'm not limiting my thoughts as much as many seem to be. What's the difference between re-writing legacy code to another legacy code vs a new language? Why can't IBM create the greatest universal tool ever? Check out my post to Ralph's response.

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
In the intro to SunMicrosystems Java manual they state that one misconception is, "Java will become the universal programming language." So even Sun recognizes that JAVA is not the answer to many computing needs. It seems, we aren't following the ideas being presented here. If we want the AS/400 community to grow and prosper, we need something that can make even the Windows developers envy. All we have currently is ILE RPG/COBOL and CODE/400. We need to let go of the old and create the new if we are to have a common 400 community of the future. We need to get IBM to expand their mind set. As well as ourselves. I hope this helps you to understand where I'm going here. Either you want the 400 to survive or you don't. I do.

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
For what it's worth, here are my comments on Bob's list. (I may be on LOA until April, but I still can't resist checking in on these fora now and then!) 1) Here's a wild and crazy idea: Why not write procedures to simplify FTP access? (What next, HTTP access? XML-RPC access? Gopher access? SOAP?) 2) Again, what's wrong with procedures? 3) No argument there. But what's stopping you from using whatever editor you like today? 4) Hmmm, I can write that procedure in 5 notes, er statements! 5) As they say, hindsight is 20/20. Looking back at V3R1, I can think of several mistakes that were made in the first design for RPG IV that now give us significant grief in continuing to enhance the language. At that time, the main goal was to define a new syntax for RPG III, and not necessarily an improved RPG. Looking back, it's amazing what we did actually accomplish for that 1st RPG IV release, but if the language designers were a bit more far-sighted, we could have had a better base from which to take the language further. (But in general, the same problem afflicts probably every other programming language out there. Just look at the work going on to develop Perl 6. Even Python, as great a language as it is, has its own warts.) 6) "...and yet is one of the best languages to use for CGI programming." !?!? My opinions on the subject are known by many in this community, and I won't debate the subject again here. But again, what's keeping you (or anyone else for that matter) from developing the perfect procedure library for CGI? (BTW, the only other general purpose language with specific CGI functionality in the language, rather than in class libraries, is that gawd-awful language Perl. Unfortunately, many of the really useful CGI features in that language would be very difficult to implement in a compiled language like RPG.) 7) Better SQL integration - no argument there. Some improvements are on their way in the next release, and hopefully also in future releases. Not perfect, but there does seem to be a definite push on in Rochester to improve SQL programming in RPG. 8) CCSID independent character handling would definitely be nice to have in the language. But note that it would entail a significant performance penalty. 9) Overloaded procedure parameters is one of our own favorite future enhancements, and maybe in some future release (not the one currently under development) it will bubble to the top of the plan. 10) Bob, it seems like every time I see one of your "wish lists", you include at least one item that is already available in the language. This time, this is that one item. When we added // as an "end-of-line" comment in the /FREE spec, we also added support for // as a comment in non-free specs. This way, in V5R1+ programs, you can have one consistent commenting style for the whole program, rather than one style for /FREE specs and another for fixed specs. Cheers! Hans

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Pardon me if I'm a bit wrong here, while it would be nice to design / edit screen, report, and file formats as well as program and operating system logic (ie. RPG & CL) all within a single utility / language (without having to recompile the whole mess - WOW!), doesn't that sorta flies in the face of modular and object orientated programming? That seems to be a bit of a digression back to the days of huge monlithic programs.

[jquinlan]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Instead of ridiculous panel groups, I wish IBM would come up with a way our programs could use a source document composed of simple HTML tags for help screens. It seems that couldn't be too difficult to do. Anything but evil Panel Groups. Jim Quinlan http://www.AsthmaStory.com

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Hey, I've got an idea.... Let's call this new language Esperanza! ;-) chuck Opinions expressed are not necessarily those of my employer. "builder" <builder@mcpressonline.com> wrote in message news:6ae51335.9@WebX.WawyahGHajS... | Who cares what the compiler generates? Why put a limit on the future?We need new tools in the form of a new language. Something that will attract even the Windows guys. | | I'm talking about a real universal language that will work for everything a business needs. I'm not limiting my thoughts as much as many seem to be. What's the difference between re-writing legacy code to another legacy code vs a new language? Why can't IBM create the greatest universal tool ever? Check out my post to Ralph's response.

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Bob: Why should I write a "find replace" procedure if you've already written one? You're absolutely write that every one of the 6,000,000 RPG programmers out there shouldn't have to re-invent the wheel. And in other language communities, there is no difficulty whatsover in finding code that implements common programming problems. As I've mentioned at least half a dozen times before, in particular, Perl programmers know to look first in CPAN for code. CPAN is an archive of user-written code for many common programming problems. Why can't the Perl language developers do all that themselves? Well, they're busy with language development. Likewise, there is only so much that the RPG compiler developers can do. (And half the team is currently off on LOA!) As you've been told what seems to be a zillion times already, the RPG development team chooses items for implementation that can't easily be done using procedures or any other language technique. If some requested function can be performed by calling an API, the function is considered available in the language. Likewise, if some function can be performed in an easy procedure, it gets shuffled down (low down) on the priority list. The items that get chosen for implementation get chosen because there's some benefit that can't otherwise be easily done in the language. But then, you already know all this - we've debated this ad nauseum in the past already. I agree that this particular community has specific needs, and I have no easy answers. For example, why should any organization put a critical dependency on some code of questionable origin downloaded from the web? And so some RPG development shops can easily justify spending a coupla thou on some useful package of code if it comes with a promise of support, where in other environments equivalent code is readily available for free. Sure, different communities have different needs, and I can sympathise with a "batteries included" approach (like Python has). But then again, where do you draw the line? Find-replace? Statistical functions? Or should the language have built-in functionality for A/R, A/P, and G/L applications? But then again, as the old cliche says, the only constant in this industry is change. Old methods and expectations can't stay the same forever, and the open-source community (in particular) is changing the landscape significantly. Open-source has been one of the biggest influences for at least half a decade now, and the RPG programming community has still largely sit on the sidelines wondering what the fuss is all about. Believe it or not, but many in IBM do understand, and believe it or not, there is indeed a business case for it! "Tabbed dialogs" is indeed an interesting example, and is a great source of bad UI design! To get back to the open-source paradigm, when the need for some functionality is identified, and someone provides a solution, very often that solution can be merged into the official distro of the software (such as the Linux distro's). Or be made available in some common repository (like CPAN). As you've illustrated quite nicely, proprietary software has some significant limitations. We've discussed this before - what the RPG community needs is a user-supported resource like CPAN, the Comprehensive Perl Archive and Network. Based on previous discussions, I was led to believe that you (in particular) were planning such a service for the RPG community.

[Guest.Visitor]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Don't believe the Hype! There won't be one killer programming languague in our career lifetime, JAVA or C#. I agree with Robert Dean there are too many languages already!! There are a lot of us RPG programmers out there and this language certainly is powerful. That's why it is time for a name change to shake some undeserved baggage as only being a report generator. (from programmers whose only exposure was through college RPG II classes.) I use RPG-IV, SQL, Visual Basic, and have looked at JAVA but am waiting. If IBM would create a compiler that would run APG on an X series against DB2 universal that would allow me to stick to the skill set I know best or fits best. (Oh LINUX or MS for the OS.) .NET is comprised of different languages so don't think or act like MS has gotten to one killer language. They are like tools use the one that works best for the task at hand. I know that Visual RPG is there but half-baked. If we could dump the 5250 data stream and go to XML via a browser with go ole APG that would be cool and we wouldn't get that dead end skill set crap or old technology...

[l maartens]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
As for the FTP function suite, look no further. Scott Klement has already provided the complete open source FTP function library. Open sourced , all user documentation, examples and a quick response in assisting the further addition and improvement of functions.

[jmyers@smurfit.com]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
I agree with all of your suggestions. However, I do still use SEU in some cases. I would definitely want it loaded on my system, if only for emergency purposes. It comes in handy when I need to make a quick change from a PC that doesn't have Code400 loaded. Even on a PC with no 5250 emulation I can always telnet in to make changes. In case of a network problem SEU changes can be made from the console. This may not affect all others as much, but we operate a 24x7 production facility and we live and die by uptime. Personally, the change I've wanted the most for many years is multi-dimensional array support.

[rdean400@yahoo.com]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
IBM's in the business of making it easier for their customers (big sprawling monolithic corporations like IBM itself) to control their IT costs. One way they seek to do this is by rallying around Java, so that corporations can run all sorts of different workloads on the iSeries, where they'd have to buy multiple boxes before. This also means the number of programs that run on iSeries increases, as does the number of people who can write programs that run on iSeries. For these reasons, IBM is not likely to develop a new language. About the only scenario I can think of that they would do it is if Java loses its viability. In that event, they'd either latch onto .Net or come up with their own competitor.

[rdean400@yahoo.com]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
There are too many languages already, and I think I've even said that on this forum before. Just because a language may not fit your idea of the perfect world doesn't mean you should run out and create a new one. There are MANY languages out there: RPG, COBOL, Java, C, C#, C++, Objective-C, Ada, Pascal, FORTRAN, Lisp, Prolog, Basic/VisualBasic, Scheme, Eiffel, Haskell, Python, Perl, PHP, PL/I, APL...just to name as many as I could without having to stop and think. A new language won't get you productivity. New for new's sake doesn't get you anything. What IBM really needs is to make the developer tools more productive, and they're becoming much better at it (WSAD5 may be a memory hog, but its a *great* IDE).

[R.Daugherty]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
builder wrote: "Something other than JAVA..." Just curious. Why something other than Java? rd

[R.Daugherty]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Agreed, Robert. Not bloody likely. "About the only scenario I can think of that they would do it is if Java loses its viability. In that event, they'd either latch onto .Net or come up with their own competitor." No one could do anything to make another language viable that hasn't been done with Java. It can't become unviable, yet another language become viable. But I know you were just commenting on the only possibility, etc. rd

[R.Daugherty]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
builder wrote: "I hope this helps you to understand where I'm going here. Either you want the 400 to survive or you don't. I do." And so do I, and I think everyone here. I'm certainly no Java zealot. I trashed OO a couple of days ago and of course got no response. There's no beef, there's only philosophical statements that such and such is true, 'cause everyone says so. They can go into hours of proving that OO is a significant leap for mankind, yet can't point to anything significant, such as an ERP, written in OO, nor show any of the benefits has ever happened anywhere in the history of the world that they spent hours attempting to prove must exist. However... I have no idea what another language is supposed to do that Java doesn't do. I mean, how could the syntax vary that much from Java and be viable? I like assembler and RPG, which have the same basic constructs and thinking, vertical thinking I call it. C and derivatives are horizontal in nature, back and forth, back and forth, back and forth, until my head is snapping back and forth like I'm watching tennis. I can't stand it. It makes me dizzy. On the other hand, no one but me cares what I like. But I digress. IBM has a ton, I mean billions of tons, of effort to make Java *the* language and for a world class Java runtime environment to run on the AS/400. What possibly could a different language for the AS/400 do that Java doesn't do, and what would it look like, and why? rd

[R.Daugherty]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
I know this isn't on topic, but I don't have the time to try to figure out where and how to post someplace appropriate. Ignite/400 sent out a news e-mail about their new AS/400 (call me a bigot) with the new Apache webserver. The e-mail says they love it. Unfortunately, the e-mail says it every few minutes as it keeps arriving over and over and over and over and... you get the picture :) If anyone knows an Ignite/400 admin and can get an e-mail through to them, please pass on from me that their stress test passed... :) rd P.S. a big assumption on my part is that a new POP server came with the package as well... :)

[R.Daugherty]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Ignite/400 says this morning their list mail was responding to an automated response from a list member, and so these computers reliably hurled e-mails back and forth to each other all night. They dropped the one offending list member account and are looking into the issue on their end. rd

[R.Cozzi]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
** This thread discusses the Content article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
0

[R.Cozzi]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Then write that procedure! Hans, I have written a FindReplace() procedure and it's going to be in the next issue of the news letter. But my requirements comes from the point of view that the 6 million (or whatever the number is) RPG programmers out there should not all have to write that routine themselves. Some things can be done more efficiently (I hope!) by you and the gang. For example, everytime I write a new sample piece of code that requires feature X, I have to include feature X in that sample because it is not part of the native language specification. Granted, I am building an RPG ToolKit that has many, many procedures in it, that I hope "everyone" will download, install and use, but history has shown us that on this platform (with few exceptions), unless it comes from IBM, 98% of the population just won't use it. I had this discussion with a Microsoft developer once over the "Tabbed Dialog" boxes you see in Win32. Back then, Win16 (Windows 3.1) still owned the market and many developers wanted tabbed dialog boxes in their applications. Microsoft kept telling us "you can build that code yourself, today, but doing A, B, C and D." Problem is, in that context, that mean 3 million people have to waste their time reinventing the wheel because MS didn't want to bother. So I took his suggestion and I wrote a tabbed dialog class for Windows 3.1. Then I saw him at another event and told him it took me nearly 3 week of 10 hour days and about 6 days per week to get them to work. And I said "So now another 2.99999 million people have to spend a collective half a billion hours duplicating my efforts." Sure I've written a FindReplace() routine, an FTP() procedure set, a convert database to CSV procedure and more. If I publish these procedure (some of which I will) that means about 40,000 subscribers will have the opportunity to see the code and maybe use it. Of those, what 100 may try to key it in. And of the other 200,000 or 550,000 AS/4000 (which ever number you believe) how many of them will be able to take advantage of it? Probably 1 or 2. But if IBM addes feature X, then everyone has it and I don't have to worry about those other quarter million AS/400 shops. Oh, and the Microsoft issue? Well, he finally added it and virtually everyone who built new applications took advantage of it.

[David Abramowitz]

** This thread discusses the article: Top 10 Things RPG Needs (http://www.mcpressonline.com/index.php?option=com_content&view=article&id=3594) **
Builder said "we should want, no demand, a new universal language from IBM. Why are we limiting ourselves?" I hate to use the term "legacy", because it sounds so pejorative. Sort of like a conservative using the term "liberal" in such a fashion that "liberal" actually means something less than despicable. But until a better term comes along, I suppose "legacy" will do. There are tens of thousands of firms who rely on RPG, and even COBOL, to handle their daily business needs. These firms have a great deal invested into every single solitary line of code, and there are hundreds of thousands of lines of code for every firm. The business work, the code works. To ask any business entity to scrap the legacy investment, simply because one does not like the delivery method is bad business. Most firms will move if they are convinced that there is a solid business reason for doing so. Even then, a wholesale move is not usually an option. In a poor business climate, those who have stuck with legacy techniques have fared better in TCO, and ROI than those who have jumped on other bandwagons. When you consider cost per line of code over the entire enterprise (this includes maintenance and time of development) RPG comes out very well indeed. The interesting thing is that (graphics aside) there is very little difference in quality, and quantity of the final information delivered from platform to platform. Most execs, and managers are only interested in what they get into there hands, and not how it gets there. Dave

[Guest.Visitor]

I believe IBM uses the term "heritage" code instead of "legacy" code. Chris

[Guest.Visitor]

I wish we have more RPGIV ILE built in functions like: 1. "%like" that finds a string of text within a text field. 2. "%userid" that returns the current user's Id. 3. "%workstid" that returns the current workstation Id.

[momhodge@aol.com]

Hey, don't just tell us about it... Where/how can I get it?! Karen

[S.Mildenberger]

http://klement.dstorm.net/ftpapi/ I have used this and it works great. Scott Mildenberger

[Guest.Visitor]

Karam, I believe as of V5R1 you can use INZ(*USER) as an keyword qualifier for a character field in the D-specs. That's a really easy way to get the User-Id! You can get the Workstation Name indirectly by defining the Program Status Data Structure and moving positions 244 - 253 to some JobName field. This is always the workstation device name for interactive jobs. We use an externally described data structure for the SDS, so don't have to define it every time. The User Name is also available from the PgmSts strucuture. I would guess that the %Scan function would do some of what you mentioned with a proposed %Like BIF. I personally would like to have a %MoveArray function, but I realize why it would be quite cumbersome for IBM to implement. It would be easy to "roll our own" if there were are really bulletproof way for a subprocedure to know the length of the incoming parms! Hope this helps, Rich Hart rhart@atcdgis.com

[markpejsa@earthlink.net]

I'd like to be able to search up -- instead of alway's having to go down in the source code