It's always hard to fill the shoes of someone who has gone before, and Bob
Cozzi's decision to step away from this newsletter has left MC Press Online with
a challenge: How do you replace a teacher who has proven to be as intelligent,
thoughtful, and experienced as Bob Cozzi?
The answer is "You don't!" Instead, you take a look at what the readers need
to know, and you aim to find the best talents available to promote the
mission.
What is the mission? RPG, of course! And everything it touches!
Modern Languages and Modern RPG
Let me tell you a story that recently piqued my
interest.
A person high up in the echelons of the corporate IBM world (who shall remain
nameless) sent me an email that smacked a bit of typical big-IBM chauvinism. He
asked me rhetorically, "How can we move the cult of RPG programmers into the
modern world?"
By this, he evidently meant moving RPGers to the so-called "modern languages"
that his corporation has spent years promoting. What these individual languages
are and have been, of course, has evolved over the years: First there was PL1,
then C, then C++, then SmallTalk, and currently, Java—in support of IBM's
WebSphere middleware platform.
Now, don't get me wrong: We're not in a programming language war. Each of
these languages has its merits, and each has had strong, knowledgeable
proponents. In fact, I've heard some of these proponents tell me, every five
years, that RPG is dead, only to have them come back to me later to inform me
that though they'd now moved onto yet another programming language, in
their view, RPG was still dead!
Dead? This so-called "modern world view" of programming languages seems
strangely divorced from the real world of System i programmers! What is
considered "modern" seems to be merely "hot." And the proof is that RPG is still
here and still thriving, while many of those IBM-sponsored languages are but
distant memories. Why? Because RPG is what academics call a post-modern
language: a classic with a lifeline that debunks the myth of obsolescence.
Legacy Versus Cadillac!
So let's get to the bottom of the "legacy" versus
"modern" debate right up front.
To my way of thinking, RPG is clearly not a legacy. It's a Cadillac!
Not only is it a vibrant and highly productive tool used by literally thousands
of developers day in and day out to maintain legacy code, but it's also a luxury
vehicle used to build new and productive applications that are vital to the
business community. And it is being used because, when the dust settles, it is
the heart of the System i developer community and all that this community has
come to expect in a luxury language!
Getting the Message Out
Of course, getting this message up the IBM food chain
is not our goal here at RPG Developer. Instead, our message is to let you
know not only that RPG Developer is alive and well—and still
dedicated to the RPG language—but that we intend to transform this
newsletter into something that you and your organization can better use.
In the coming months, we'll be transforming RPG Developer to do more
than merely fill in the blanks of your personal RPG knowledge. We'll
evolve this publication to help expand the use of the RPG language within
your organization.
How?
Best Team of Educators
First, we have nearly finished lining up one of the
hottest RPG editorial teams of educators on the planet. These are professionals
who not only know and use RPG on a daily basis, but have built their careers
around expanding the use of the language in business environments just like
yours.
Second, we've engaged one of the best educators in the industry to prepare
step-by-step tutorials that your organization can use again and again to train
and educate the next generation of RPGers—first, in the use of the tools
used to construct RPG source, and later in the implementation of the key
functions that make RPG so powerful to the business community.
Beyond the RPG Blog!
Finally, we're opening the doors of this publication
to you, our readers, to send us your ideas for new articles, tips, and
techniques.
In the past, the value of RPG Developer was that it was almost a
personal blog written by one of the industry's most renowned RPG authors, Bob
Cozzi. Our goal is to move beyond the personal blog to enhance your involvement
in the RPG community at large.
So if you have questions or queries or ideas about RPG–related areas
that need to be explored, we welcome your input. We'll try to build our
editorial schedules with those ideas in mind. And if we don't know the answer to
a particular question, we'll do our best to find it for you.
Our mission statement is the same as it has always been: RPG all the time!
But we're doing a bit of remodeling as we move forward with the new RPG
Developer: Now it's RPG and all it touches all the time!
Drop us a note with your suggestions! I think you'll find the new RPG
Developer will meet and exceed the expectations of all you've come to love
about this newsletter.
Tom, <p>Thanks. I already read that newsletter too. I think my point was missed. If RPG is in the *title* of the newsletter, I would expect it to be centered around RPG, not PHP or Java or any other language. Other languages could certainly be on the "fringes". <p>I do like the idea of discussing RPG in terms of other languages too, so non-RPG people can relate to it. I do this as well. For example, instantiating an object in Java is a foreign concept to RPG'ers. But if you explain that the class is like an empty physical file and you have to "instantiate" a record in it before you can read a record, then instantiation makes more sense. <p>Chris
Chris, <p>We already try to cover lots of different language topics in a specific newsletter called iApplication Developer. Don't know if you're a subscriber, but if not, check it out. <p>Tom
Mike, <p>Absolutely no need to apologize. <p>I asked the question because I want to know! Yes, PHP is big, and many of us are really interested in how that's going to be incorporated in the application toolkit. But there are some developers who just want to get better at what they know. RPG is one language, and RPG Developer will cover that. But ... Well, there's a lot happening on the System i, and it's important to bring in a base of understanding by which more and more potential readers can consider RPG. <p>What do I mean? Let's see if I can explain myself. <p>Let's say for argument's sake, you're fresh out of college and you know some Java and PHP, but -- because they don't teach RPG much anymore (or should I say, "Not Yet!") -- you've never been exposed. (People in Rochester used to call these uninitiated "The Unwashed") Yet you find your first real job and it's in a manufacturing organization that runs a System i. Your task, as a new kid in town, is to start learning RPG and get exposure to the language by maintaining code. "No problem!" you say to yourself. "I know Java! I know PHP! I even know C#! This should be a snap!" <p>But then, after about two weeks, you start hating RPG! Why? Not because it's simple, and not because it's "old"! You hate RPG because you have no conceptual framework by which you can understand the language, its integration with other System i services, the open memory model, etc. All your previous educational experiences in languages are from a completely realm. And it's driving you nuts! <p>So you go to your IT manager and you say "Hey! You know I could do this a LOT faster in PHP or Java!" <p>And that wise old sage of an IT manager eyes you up and down and says "Well, YOU could do it faster in PHP or Java, but Tom we don't have those languages here. So go and sit over next to Mike there, and learn how he thinks about RPG." <p>Now Mike is a really smart developer. He's had a lot of experience in a lot of different languages, but he LOVES RPG because he understands it. So, he acts as a mentor to you, putting concepts into something you can understand. Concepts that relate to Java and PHP, and which describe RPG in a syntax that you can understand. And sure enough, in about two months, you start to understand Mike (because he speaks your "language") and you start to think "Hey! Maybe this RPG thing isn't what I thought." <p>My point is, to attract new blood to the RPG language, the writers of this newsletter have to be able to speak in a syntax that newbies can understand, not just what we "experts" understand. <p>So your comments are not out of line, but they do, in fact, open a window to a strategy: The RPG Developer newsletter can learn to attract new readers by talking about RPG in different ways, to attract a different breed of developer who needs to learn RPG. That might be a PHP programmer, or a Java programmer, or even a COBOL programmer. <p>My educational background in languages was originally in Fortan and COBOL (Yes, I'm that old!) and my first experiences in RPG were rotten. Fortunately, I had a really good mentor (named Dave Bailey) who got me started and up over the first hurdles. And then the good old System/34 -- which was so easy to write code for -- helped me transition. <p>Of course, the System i and ILE is not that easy for a newbie today. And old guys like me are not going to be around to help them transition to the language much longer. Yet you know that many MANY companies find that RPG is exactly what they want and need to run their businesses. They will not be forced to change platforms because they can't find new RPG programmers. They'll train them in house if they have to. <p>So part of our task as editors is to educate those new users -- as well as challenge the rest of us. Not only to provide the RPG Developer with the best RPG articles, but to also provide a wider context and syntax to help others gain access to this great resource to BECOME RPG Developers. <p>Tom
Jon, <p>Are you proposing that the newsletter be renamed to "i5 Developer"? I would not restrict the letter to just RPG, but that sure would be the focus since RPG is in the *name* of the newsletter. <p>Chris
While I consider myself primarily an RPG Developer, RPG is is not all that I do. <p>Don't most of us code some CL? DDS? SQL? A great many of us (myself included) also code some Java (Yes Joe - really!), some PHP, and even a little COBOL from time to time. <p>My point is that perhaps one shouldn't take too narrow a view of what an "RPG Developer" is. Isn't it as important to explore avenues that can fill in the gaps in RPG (yes there are some) as well as discussing tools and techniques to extend one's RPG skills? <p>One other point. The "RPG Developer" of today (or at least those interested in reading about anything they couldn't do on a System/36!) is doing very different things with very different tools today than just a few years ago. Shouldn't an RPG Developer newsletter explore potential future avenues? <p>Jon P.
Tom, <p>I apologize for cluttering up this thread with a discussion about PHP. I read the by-line "Beyond the RPG blog" and assumed you were fishing for ideas beyond RPG. It appears that you were looking for RPG ideas beyond the "blog". Please feel free to transplant the PHP discussion somewhere else, I would be happy to cease and desist until then. <p>btw: Still a HUGE RPG fan, too. Can't we all just get along... <p>Regards, <p>Mike
Yes, I agree. I'm thinking along those lines also but you have to start small with shell programs that integrate into a larger system. A small ERP is a goal but too much to start with. <p> I would think you'd even want to keep the app artificially simple to have some logic to display infrastructure methodologies but not be a distraction. <p> The key is a flexible infrastructure in place. Adding business logic would just be more of the same. <p> rd
Okay, this is getting dangerous, I thought a little more. <p>Take that book and base it on a small ERP home grown application that we create in modules. <p>Partner up with either IBM or a an existing AS/400 timeshare that will allow all the bells and whistles, and make this a FREE tutorial course, and yea, go one step further and let's create an open source library of Service Programs and Procedures that 'NEW' and existing RPG programmers can use in their everyday work day. And let's only then after we get this modest baseline use of RPG in todays world, or where it should be.... <p>Let's then get 'Creative'. <p>But, it would interfere with life after 5 o'clock, and could be contagious, and could invigorate the RPG community, and could increase your net worth...for who: MC Press, any Developer that gives up a few soccer games, any employer whose Developers play in this sandbox, any Soccer Mom or Dad who has to go to the soccer games without Daddy or Mommy (you never know today). And what about NEW BLOOD, it could happen if you think and act positive, don't try and you will never know. <p>Is it worth it? <p>It has been for me. But, then as Joe does with interweaving song lines in his 'Weaving Websphere', 'I may be crazy', or 'Don't shoot me, I'm only the Piano Man'...you can learn to play too. <p>Craig
I thought about a little more. <p>I want a roadmap. I want all of the service stations listed, I want all of the exits listed, I want all of the attractions listed, I want to know all of the exits that have little stops from hell listed, so I can avoid them at all costs... <p>What about a column that could double for a book later. Broken into chapters starting from the beginning. <BR>
1) What is the current state of RPG and where should I be. <BR>
2) Convert your source to RPG IV and free form and remove all of those (MOVE)'s. <BR>
3) What is ILE? What is an Activation Group? <BR>
4) What is a Service Program? What is a Function? a Procedure? a Sub Procedure? and Why do I want any? <BR>
5) Binding Directories and why do I want any? What will they save me? <BR>
5) Okay now we have RGPIV, free form, ILE, Service Programs, Procedures, where do we go from here? <BR>
6) Extracting your Business Rules to Procedures, or the JOY of write once, use many times, and don't recompile. <BR>
7) Okay, now we have to convert all of your code because we are starting to experience cases of mixed Default and Non Default Activation Groups, the Joy of existing Triggers and Activation Groups and Scoping, so we now need to understand Scoping and their use in Activation Groups. <BR>
8) Are we ready to now Develop an Application in a true ILE Environment? <BR>
9) Are we ready to use API's. <BR>
10) Let's have a cup of Java. <BR>
11) Let's talk to Joe about why we should have been using WDSc all along. <p>I think it would make a good series, and a good book. A year ago I might have written it myself. <p>Just some thoughts, <p>Craig
My answer to your question. If I start to read an article in a slot that says RPG Developer, then my expectation is to read about the forward progress of RPG in words and examples. <p>I could care less about the doubters and the naysayers. I make my living off of good clean RPG in use in a progressive environment. I want to increase my net worth with usefull information that I can either use today to solve my immediate needs or use to affect the direction of my education or the direction of the environment I work in. <p>I want to learn how to work RPG with PHP, RPG with Java, RPG with Cobol, RPG with etc... <p>If I want to compare RPG with PHP or Java or etc...how about a column titled: RPG vs etc...? This could be something fun as a diversion, but what I have come to your place for consistently over the years is to further my career with knowledge gained in RPG. <p>Thanks, <BR>
Craig
Thomas, <p>I think the RPG Developer news letter should be about RPG programming - topics/code we can read and use right away. It does not have to be purely technical, but can contain programming techniques. <p>For example, I just coded an RPG batch program that if it crashes during updates, the next time it runs it detects this and fixes the data and continues on. (not using commitment control). Downstream processes will ignore the data until fixed. <p>I think a separate PHP newsletter would be awesome. <p>Chris
Not to spoil a very good discussion about the merits of PHP vs. Java, but I want to remind everybody that the newsletter is called "RPG Developer". So it's an interesting to ask the following question regarding the newsletter: <p>a. Should the newsletter be about RPG? <p>or <p>b. Should the newsletter be about RPG Developers and their needs? <p>It's one thing to speak to RPG Developers and their prejudices and preferences in terms of language structure and functionality. <p>But it's quite another thing to actually talk about RPG -- the language -- and how it can be efficiently used. <p>And there's a temptation -- when lining up content -- to run for a while creating articles that speak to the audience and its particular needs. But when push comes to shove, sooner or later, the newsletter will start to lose focus, and ultimately readers, while failing to attract new readers. <p>The debate about PHP in the RPG Developer is a good example: Is PHP really essential to this newsletter's topics? RPG coders find a natural and understandable attraction to PHP because its structure and format are easy to comprehend. It's "obtainable." <p>But -- despite the call for PHP content in this newsletter -- is that what RPG Developer be about? Or should you (dear readers) be looking to one of the other of MC Press's newletters (Like iApplication Developer.) Or would you be interested in a iPHP newsletter, which is targeted to System i Developers who want to learn more and more about PHP? <p>Just asking. I'd appreciate any insight into your thoughts. <p>Thomas M. Stockwell <BR>
Editor in Chief <BR>
MC Press Online
I teach RPG programming at Gateway Technical College in Kenosha WI. I wanted to comment on two things. <BR>
One, the students in my classes actually like RPG. I teach Free format RPG ILE this includes procedures modules and service programs. I’ve actually heard students say they like RPG better than Java! <BR>
Secondly, this years graduates (with the exception of one) got jobs this year based on their knowledge RPG not their knowledge of VB.net or Java. <BR>
Check out the Programmer Analyst web page at Gateway <BR>
<a href="http://www.gtc.edu/pages/displayProgram.asp?pid=10-152-1">Gateway System i Programmer/Analyst </a> <p>Jim Buck
Jacobus asked a good question the other day, what makes a good business language, and I don't know about others but for my part unfortunately I haven't had time to try to answer. <p> But I know for a fact that we for the most part could do just about anything in any language on any OS running roughly equivalent hardware. There are extremes we could push it to that would make some tasks too long, but we've all got real innovative in the past to get around the worst of obstacles through the years with software of all sorts. <p> So it's true PHP could be compiled, and it's true that dynamically interrogating the data to determine if it's character or numeric is perhaps not the most robust of business solutions. <p> On the other hand, PHP has added variable declarations, and we've done plenty of ambiguous character-numeric operations with MOVE through the years, haven't we? <p> The answer is not syntax, it's infrastructure. A language is a communications to infrastructure, and each infrastructure usually arose in symbiosis with its language. <p> RPG is the language to talk to a world of packed decimal, record level IO or embedded SQL, and 5250 displays. It can generate text strings that represent other worlds, like HTML web pages, but so can anything else. <p> PHP is a language to talk to the infrastructure of the web server. It has thousands of built in functions oriented around a web server infrastructure with which to hold that conversation. <p> Java is even more fluent, but yet another infrastructure unto itself. <p> If one were not attempting to leverage existing logic, and in my opinion trying to extract logic to be called might as well not be called existing logic, would one create an ERP in PHP? <p> There are one or two CRM focused systems in PHP, but last I looked portions had been written in one of them in Oracle PL-SQL "for performance reasons." But let's take that further and say, ok, a good portion of logic is IO bound, so we'll push that to the database infrastructure, and perhaps the language best suited to talk to that infrastructure. <p> RPG could be said to be the database language of DB2/400, with record level IO and embedded SQL with equivalent DB2 extensions. <p> The crux of PHP is that it is usually used as an embedded module in the web server, such as Apache. This makes it part and parcel of that infrastructure, running in that existing space. <p> Is there a best of both worlds answer where the shell PHP programs make calls to ILE and/or Java programs, ideally already up and running and communicating through data or message queues but melding the power of business systems with an in space web server module? <p> One could argue similarly for the Java web server module, Tomcat, but perhaps argue PHP is simpler, or requires less hardware, but depending on need, the cost effectiveness of Java could be compelling, which gets back to those built in functions. <p> Personally, I would like to see us provide RPG with equivalent ILE functionality rather than as a DB2/400 database language for other crossplatform environments such as PHP or Java. I don't think there's really a question of choice until we also have thousands of functions to call upon in ILE, and in doing so provide high speed native business solutions. <p> Unless we're doing that, we're only an equivalent alternative to many other choices with greater visibility. But if we do that, we could be more than a viable alternative, we could be a compellingly more performance effective alternative. <p> rd
Joe, <p>My expectations for PHP are quite high, actually. I don't think IBM and Zend would have worked as hard on the port of Core if there were not additional plans. 'Watch this space...(:-)' Plus, the way PHP has been implemented on System i facilitates the ability for Zend to tweak PHP without having to wait for an IBM Version release cycle. <p>Truly, I never said PHP would be a Java killer. RPG did not kill COBOL, therefore I do not believe PHP will kill Java. It's not black and white. In fact, while I have not seen any statistics, I would bet there is still more COBOL out there than RPG, except on System i. Also, RPG in the early days could be compared favorably to a scripting language. RPG I & RPG II for example. (All hail the logic cycle!) Todays RPG doesn't compare closely to yesterday's. So let's pretend that Zend and the open source community might continue to develop PHP to the point where, dare I say it, it creates a compiled object? I don't know, but why not? <p>And why not use PHP for business logic? I love being a heretic since my Dad had similar discussions with the COBOL zealots back in the 80's while he was pounding out his RPG on 96 column cards and laughing while he needed only a fraction of the time on the keypunch machine. All I'm saying is that if you need to get your stuff on the web quick, and you are a one man shop, pick your tool. There are plenty of viable contenders, including third party stuff like Lansa, et al. I still like RPG and CL will always be my first love. (especially when compared to JCL!) I think PHP is as viable as CGI or Java. But, I Never said PHP would be a Java Killer. Please do not misinterpret my words. Sheesh, now I know how it feels to be Jon Paris...Actually, I have far too much respect for Jon to compare my self to him, but you know what I mean... <p>C'mon, more healthy discussion, please!!! <p>Regards, <p>Mike
Mike, first you call PHP a "21st century equivalent" of RPG, and then you say that "PHP will do to Java what RPG did to COBOL." I'm not sure what your expectations are for PHP. <p>Do you plan to write business logic in PHP? I hope not! PHP is not a particularly pretty language for business logic; it's more of a C derivative than a BASIC derivative, and it's got a lot of weird baggage. Take a look at how clumsy the PCML calling conventions are. I prefer free-form RPG to PHP any day. <p>As for being a Java killer, I'm not sure. Java does a whole lot that PHP isn't really designed to do, including thick client code. It might be more appropriate to position PHP as an alternative to JavaServer Pages (JSP), since PHP is very much designed to be an HTML generation script language. And for that particular task, largely because of its huge existing user base, Zend's PHP implementation vaults up into contention as a browser UI option for the System i, right alongside RPG-CGI and JSP. I've got an article coming out next week on that very topic. <p>Joe
Ralph, <p>So, what do you mean by "a convergence of RPG and PHP?". Zend PHP already has a toolbox similar to the Java toolbox that can call *PGM programs, call commands, use data queues, etc. <p>Chris
Why do say request? Who do you think I'm talking to? Not IBM. I don't request anything. <p> I've had a PHP web site for several years now. You don't need to tell me anything about it. <p> Any script can be compiled with some combination of a compiler and a conversion. <p> rd
Ralph, <p>How many lines of PHP code have you written? That just seems like a very strange request to me. RPG is compiled into modules. PHP is a web scripting language, interpreted on-the-fly, has no clue what a module is, runs in PASE. PHP 5 has some OOP in it. RPG does not. The syntax is different. PHP looks like a combination of Java and Javascript. And so forth... <p>Chris
JavaBoy wrote: <i>Finally, management could do a better job by obtaining a clear vision (I prefer one toward being platform agnostic) about how to evolve the RPG-based applications into a browser-based world. </i> <p> Well, you'll need to be platform agnostic. :) Because you know you hear the same thing from any platform group you go to. <p> I'm sure you didn't get far on Windows with it, and Unix Perl programmers laughed in your face, and PHP programmers said Java SUXORS or whatever it is they say, and Linux C programmers said you're not open enough, so what, that leaves the iseries with the IBM Java Daddy to run it by? <p> How many billions did IBM spend on Java with their San Francisco Java fiasco? I've posted so many multi-hundred million dollar Java J2EE failures it got tiresome. <p> I think it's a great cross platform client platform now though. <p> rd
JavaBoy wrote: <i>Again, from my observation and experience, there are too few of these coming out of our domestic colleges. Mostly, they come from off-shore...</i> <p> When's the last time you saw a company even hint at considering a college graduate for RPG programming? It was before 2000, I can tell you that. <p> As for coming from off-shore, do you or anyone else who chants this mantra believe or have any clue that the off-shore programmers were experienced in RPG, or much of anything for that matter? Of course not. <p> Please don't mix up sending work off to cheap overseas labor in hopes of saving a ton of money and on the fast lane to a corner office with who does or doesn't get hired by a company that is run by people that need to count on the end product developed. <p> The airhead headed to the corner office isn't one of them. He just needs PowerPoint to work. <p> rd
#118634
There are too many comments to list them all here. See the forum for the full discussion.
