In a previous post called Is RPG Dead? The Autopsy, I suggested I'd probably have more to say on the subject. This offering is in response to Scott Klement's editorial, RPG is Dead? Are You Serious?! I'd like to respond to a couple of points he makes.
First, he brings up the fact that people have been debating RPG's death for the past 20 years, and yet RPG is still around. But why bring up the question at all if there was any doubt as to RPG's future? Does anyone ask if Python is dead? Or PHP? Of course not! No one has to.
Next, he raises the point that most application development on IBM i is done in RPG. No one is denying that. However, no one can deny that IBM i is a platform in decline. The numbers just don't lie. The fortunes of RPG are intimately tied to the fortunes of the platform. As IBM i slowly dies, so does RPG. (I said more on this theme in my blog posting Is RPG Dead? Do We Still Have To Ask?.)
Skipping past some history, Scott talks about what's important to programming business logic. Let's look at each in turn:
Business logic uses numbers heavily. Well, this is not unique to business applications. However, most arithmetic in any business application, like in any other domain, is integer arithmetic. Business logic does make use of decimal numbers, but even there, most operations involving decimal numbers are simply the movement of data. Actual decimal arithmetic is a small part of the logic, and in most business apps, could be implemented fully in SQL code. That said, other languages do have support for decimal arithmetic. Scott mentions Java for one. Python is another. Even PHP has support for decimals.
Business logic uses databases heavily. Business applications don't have a monopoly on using databases. Many different domains, from gaming to on-line social networking to statistical analysis make at least moderate use of databases. And so you can find support for databases in practically every programming language out there. RPG gives you two choices: built-in I/O operations and pre-processed SQL. Other languages often provide higher-level access to databases. For example, in CakePHP, you define your high-level relationships (1-1, n-1, 1-n, n-n) between your tables, and one find() operation can bring in all related data for a transaction. That is, the framework figures out the SQL for you, including joins. Among the available RPG frameworks, which one can provide that level of power?
Business logic uses dates frequently. And so do many other types of programs. And so you can find date support in practically all modern programming languages. RPG is certainly not unique. Python has the datetime class. PHP has an extensive list of date/time functions and classes.
Business logic uses a lot of string manipulation. I disagree with that assertion since strings are most commonly needed in the presentation layer of an application. But let's assume it's true for business logic. Again, what modern programming language lacks support for strings? Languages like Python, Perl, and PHP (to name a few) all have very powerful string support. Much more powerful than RPG's support. In those other languages, strings are all objects, with a rich set of methods. RPG, in comparison, has a rather limited choice of string operations. Consider the ease of doing regular expression matching in Perl, Python, and PHP. And compare that to what you need to do in RPG.
If these four categories define what's important in programming business logic, there are clearly languages other than RPG that have a definite advantage. Whenever I see someone claim that RPG is the best language for business logic, I always interpret that to mean that the author simply knows RPG best. And Scott certainly is an expert on RPG.
Finally, Scott makes the point that RPG is still being enhanced and supported. True. However, other languages have long ago surpassed RPG in terms of features and capabilities. In my previous missive, I listed half a dozen features common to other programming languages that are still lacking in RPG. And I could have listed more. And other languages continue to grow and gain in popularity.
RPG will not disappear overnight. To say that RPG is dead is an exaggeration. However, it is becoming increasingly irrelevant in the I.T. world. I'm reminded of a job interview I had back in 2008 where the interviewer looked at my resume and asked: "What's an iSeries?". (RPG wasn't discussed at all in that interview.)
Cheers! Hans
Showing posts with label IBM i. Show all posts
Showing posts with label IBM i. Show all posts
Monday, June 11, 2012
Thursday, May 31, 2012
Is RPG Dead? The Autopsy
In a previous post, I promised to say more on the subject of RPG's death. But I'm finding it tougher than I expected. Don't misunderstand me, it's not for lack of content. It's just that I worked with RPG for a long time. I don't want to see RPG go away. Writing this blog posting feels like I'm kicking an old friend while he's down on the ground. On the other hand, it's just a programming language. And the facts need to be understood.
When I first learned RPG III, 31 years ago, I was impressed by a couple of things. First, I was impressed with the externally-described file. As far as I knew, no other language had such a powerful feature. Second, I was impressed with the syntax-checking source editor. At my university, some professors were just beginning to think about something like that.
But other aspects of RPG III were definitely goofy, such as the fixed-form syntax and the indicators. Fortunately, the language has grown since then. Indicators can be largely avoided, and calculations can be coded in a free-form syntax. However, although RPG has progressed, other languages have progressed faster, and new languages have cropped up with even more powerful features. More and more, RPG looks like an anachronism, rather than a modern programming language.
I'll list some features that are commonplace in other, modern programming languages that RPG still lacks:
1) Namespaces. That is, what happens when you want to import two libraries with the same name? In languages like Java and Python, imports can be put into a directory hierarchy. And if there are conflicts, things can be renamed.
2) User-defined data types. Sorry, data structures and LIKEREC just don't cut it.
3) Object-oriented programming. What language doesn't have OO support these days? It's something we now take for granted elsewhere.
4) Extensive list of built-in functions or classes. Just compare what RPG has to the Python Standard Library or PHP's Function Reference. No comparison.
5) Frameworks. Most modern programming languages have a goodly selection of frameworks that simplify programming web applications. These frameworks have various levels of functionality, however, typically they take care of a lot of nitty-gritty details. Using a framework, you should never see a query string, or XML, or JSON, or SQL.
6) Operating system independence. Programs written in languages like Java, Python, and PHP can easily be ported across different operating systems.
Some would argue that RPG could be further enhanced to address these short-comings. But why bother when there are already other languages that have these features? The planners at IBM certainly know what's lacking in RPG, and they still seem to have the resources to make the occasional enhancement. But over the past decade, there seems to have been little effort to address these specific short-comings. Looking at the situation from that perspective, one could argue that IBM itself also sees no future for RPG.
So far, I've written about 500 words, and yet there's still more to be said. Stay tuned.
(Update: Next installment: Is RPG Dead? The Wake)
Cheers! Hans
When I first learned RPG III, 31 years ago, I was impressed by a couple of things. First, I was impressed with the externally-described file. As far as I knew, no other language had such a powerful feature. Second, I was impressed with the syntax-checking source editor. At my university, some professors were just beginning to think about something like that.
But other aspects of RPG III were definitely goofy, such as the fixed-form syntax and the indicators. Fortunately, the language has grown since then. Indicators can be largely avoided, and calculations can be coded in a free-form syntax. However, although RPG has progressed, other languages have progressed faster, and new languages have cropped up with even more powerful features. More and more, RPG looks like an anachronism, rather than a modern programming language.
I'll list some features that are commonplace in other, modern programming languages that RPG still lacks:
1) Namespaces. That is, what happens when you want to import two libraries with the same name? In languages like Java and Python, imports can be put into a directory hierarchy. And if there are conflicts, things can be renamed.
2) User-defined data types. Sorry, data structures and LIKEREC just don't cut it.
3) Object-oriented programming. What language doesn't have OO support these days? It's something we now take for granted elsewhere.
4) Extensive list of built-in functions or classes. Just compare what RPG has to the Python Standard Library or PHP's Function Reference. No comparison.
5) Frameworks. Most modern programming languages have a goodly selection of frameworks that simplify programming web applications. These frameworks have various levels of functionality, however, typically they take care of a lot of nitty-gritty details. Using a framework, you should never see a query string, or XML, or JSON, or SQL.
6) Operating system independence. Programs written in languages like Java, Python, and PHP can easily be ported across different operating systems.
Some would argue that RPG could be further enhanced to address these short-comings. But why bother when there are already other languages that have these features? The planners at IBM certainly know what's lacking in RPG, and they still seem to have the resources to make the occasional enhancement. But over the past decade, there seems to have been little effort to address these specific short-comings. Looking at the situation from that perspective, one could argue that IBM itself also sees no future for RPG.
So far, I've written about 500 words, and yet there's still more to be said. Stay tuned.
(Update: Next installment: Is RPG Dead? The Wake)
Cheers! Hans
Tuesday, May 29, 2012
Is RPG Dead? Do We Still Have To Ask?
Guess what, folks? It's that time of year again when people debate the future of RPG. Is RPG dead? Think about it: Does anyone ever ask that question about any other programming language? Why do some people still insist that RPG does have a future?
To start with, let me make one assertion: Of all participants in this debate, I'm one of the very few who is both knowledgeable about RPG and who does not have a financial stake either way. Some commentators make their living by moving people off of IBM i. Others make a living doing RPG programming or training others. If anything, I don't relish the idea of RPG disappearing because of my long involvement with RPG development.
RPG is very much unlike practically all other programming languages in use for application development. It is one of the few programming languages that is available on only one operating system. That's right! If you want to do RPG programming, you need a system with the IBM i operating system. And IBM i is one of the last of the traditional proprietary operating systems.
What do I mean by that? Look back at the history of computers: In the 1950's and 1960's, there were a lot of computer companies, colloquially called "IBM and the Seven Dwarfs". After a couple of mergers, they became "IBM and the BUNCH" (Burroughs, Univac, NCR, Control Data Corporation, and Honeywell). Each one had their own mainframe computer products. And each computer ran its own proprietary operating system, each incompatible with all the others. Over time, almost all of these proprietary operating systems have disappeared. Today, there are just two left: z/OS and IBM i.
Today, most computers run either Windows, or some variant of Unix or Linux. Even the machines that run the O/S relics z/OS and IBM i now also run Linux, with significant cost savings over the dinosaurs.
So here's the current situation: The fact is that the main reason for running IBM i is to run RPG applications. And to be fair, a lot of them are still running. Without RPG, there's no reason to choose IBM i. And without IBM i, there's no way to run RPG apps.
Is there a future for IBM i? When I was looking for work back in 2007, one headhunter specifically told me that the bottom had fallen out of the i job market. Indeed, over the course of about 8 months, I saw only three listings for iSeries jobs in the greater Toronto area. If there are so few IBM i job openings in the 8th largest metropolitan area in North America, what hope is there for IBM i programmers elsewhere?
Frankly, in this day and age, if you want to develop a new application, it just doesn't make any sense at all to limit yourself to one particular operating system. Especially one that is clearly in decline. Other languages like C, C++, PHP, and Python can run on practically any operating system. If you want to protect the value of your software development investment, clearly, any of these other languages is the way to go.
I have more to say about RPG, but I'll save that for another blog post. Stay tuned!
Cheers! Hans
To start with, let me make one assertion: Of all participants in this debate, I'm one of the very few who is both knowledgeable about RPG and who does not have a financial stake either way. Some commentators make their living by moving people off of IBM i. Others make a living doing RPG programming or training others. If anything, I don't relish the idea of RPG disappearing because of my long involvement with RPG development.
RPG is very much unlike practically all other programming languages in use for application development. It is one of the few programming languages that is available on only one operating system. That's right! If you want to do RPG programming, you need a system with the IBM i operating system. And IBM i is one of the last of the traditional proprietary operating systems.
What do I mean by that? Look back at the history of computers: In the 1950's and 1960's, there were a lot of computer companies, colloquially called "IBM and the Seven Dwarfs". After a couple of mergers, they became "IBM and the BUNCH" (Burroughs, Univac, NCR, Control Data Corporation, and Honeywell). Each one had their own mainframe computer products. And each computer ran its own proprietary operating system, each incompatible with all the others. Over time, almost all of these proprietary operating systems have disappeared. Today, there are just two left: z/OS and IBM i.
Today, most computers run either Windows, or some variant of Unix or Linux. Even the machines that run the O/S relics z/OS and IBM i now also run Linux, with significant cost savings over the dinosaurs.
So here's the current situation: The fact is that the main reason for running IBM i is to run RPG applications. And to be fair, a lot of them are still running. Without RPG, there's no reason to choose IBM i. And without IBM i, there's no way to run RPG apps.
Is there a future for IBM i? When I was looking for work back in 2007, one headhunter specifically told me that the bottom had fallen out of the i job market. Indeed, over the course of about 8 months, I saw only three listings for iSeries jobs in the greater Toronto area. If there are so few IBM i job openings in the 8th largest metropolitan area in North America, what hope is there for IBM i programmers elsewhere?
Frankly, in this day and age, if you want to develop a new application, it just doesn't make any sense at all to limit yourself to one particular operating system. Especially one that is clearly in decline. Other languages like C, C++, PHP, and Python can run on practically any operating system. If you want to protect the value of your software development investment, clearly, any of these other languages is the way to go.
I have more to say about RPG, but I'll save that for another blog post. Stay tuned!
Cheers! Hans
Thursday, May 17, 2012
OS/2 - 25 Years Later
I almost missed an anniversary. Last month was the 25th anniversary of OS/2. I suppose that's understandable. After all, who cares about OS/2 anymore?
Well, I still remember OS/2. I used it on my home computer up until June 1998. By then, the writing had been on the wall for years. But rather than turn to Windows, I put Linux on my home computer. My first reaction after booting up Red Hat 5.2 for the first time was: What the heck am I getting myself into? However, KDE version 1 was released just three weeks later, and that made Linux much easier to use. Not as easy as OS/2, but still acceptable. And unlike OS/2, interest in Linux was increasing.
At IBM, I worked with someone who was at the meeting where Microsoft effectively told IBM that their OS/2 partnership was over. And I still remember his description of the meeting. Both sides presented their status. After Microsoft presented their status, the IBM'ers present slowly began to understand the implications of Microsoft's position. Meanwhile, on the other side of the room, the Microsoft employees were smiling giddily, knowing full well that they were shafting their loyal partner.
After the "divorce", Microsoft did everything in their power to stop OS/2 from gaining any traction. But OS/2's failure in the market wasn't entirely Microsoft's fault. IBM's sales division didn't know how to sell it. And later on, Windows was adopted as the standard workstation for all IBM employees. Clearly, the OS/2 supporters within IBM were a small, albeit dedicated, minority.
There were broader implications. At the time, I was working on the AS/400. Penetration of OS/2 within AS/400 shops was practically zero. Microsoft had pretty much 100% of the desktops within that market. And yet, few people listened to the warnings that the desktop was a beachhead for Microsoft's incursion into the SMB market. And over time, as many of us predicted, Windows became more and more prevalent in that market, displacing AS/400 installations (and later, iSeries). And now, interest in IBM i is way down. Perhaps the only thing keeping IBM i alive now is the fanatical devotion of it's remaining users.
What would have happened if Microsoft didn't go their own way? Would we all be using some form of OS/2 today?
Cheers! Hans
At IBM, I worked with someone who was at the meeting where Microsoft effectively told IBM that their OS/2 partnership was over. And I still remember his description of the meeting. Both sides presented their status. After Microsoft presented their status, the IBM'ers present slowly began to understand the implications of Microsoft's position. Meanwhile, on the other side of the room, the Microsoft employees were smiling giddily, knowing full well that they were shafting their loyal partner.
After the "divorce", Microsoft did everything in their power to stop OS/2 from gaining any traction. But OS/2's failure in the market wasn't entirely Microsoft's fault. IBM's sales division didn't know how to sell it. And later on, Windows was adopted as the standard workstation for all IBM employees. Clearly, the OS/2 supporters within IBM were a small, albeit dedicated, minority.
There were broader implications. At the time, I was working on the AS/400. Penetration of OS/2 within AS/400 shops was practically zero. Microsoft had pretty much 100% of the desktops within that market. And yet, few people listened to the warnings that the desktop was a beachhead for Microsoft's incursion into the SMB market. And over time, as many of us predicted, Windows became more and more prevalent in that market, displacing AS/400 installations (and later, iSeries). And now, interest in IBM i is way down. Perhaps the only thing keeping IBM i alive now is the fanatical devotion of it's remaining users.
What would have happened if Microsoft didn't go their own way? Would we all be using some form of OS/2 today?
Cheers! Hans
Thursday, January 12, 2012
Should Everything Run on IBM i?
re Shouldn’t Everything Run on IBM i?
To start with, I'm reminded of a couple of things told to me when I was job-hunting four years ago. First, a head-hunter told me that, at the time, the bottom had fallen out of the iSeries job market. Second, at a job interview, I was asked: "What's an iSeries?"
In the referenced blog, Aaron mentions that the IBM i has just seen 4 quarters of growth. That's good to see. I worked for 22 years on the iSeries and its predecessors, AS/400 and S/38. And I really don't want to see my good work go to waste.
But in his blog, Aaron suggests that everything should run on IBM i. The i is indeed a fine machine to store your mission critical data. You really can't go wrong choosing IBM i for your database. But I disagree that it should be used to run everything. (Those who like the i, tend to like it a lot, which says a lot about the system. The only people more devoted to their computers seem to be Apple fans.)
First, Aaron brings up the common strawman argument that the choice is between IBM i and Windows. Clearly, there are more choices out there, especially in the Linux, Unix, and BSD realms. For example, for the top tier in a web site, you can't go wrong with OpenBSD running on a couple of inexpensive Intel servers. You can easily configure OpenBSD to offer 100% availability, for at least your static pages, with rock-solid security.
But let's look at the application tier. Many shops choose an IBM i solution for one simple reason: The application they need only runs on i. If the application is implemented in RPG, then it's basically limited to the i. What if you need to develop your own custom application? Is RPG the best choice? Frankly, no. Thirty years ago, when I started working on the RPG III compiler, it was already an oddity among programming languages. It has advanced quite a bit since then, and I'm proud of my own contributions to its evolution. But it has not kept pace.
Is there an ideal choice of programming language? Again, no. Actually, I think that's the wrong question to ask. Today, when choosing a programming language for a particular project, you don't look at the characteristics of the programming language. If the attributes of the language were a primary consideration, Python should be the most popular programming language on the planet (in my opinion). Today, one does not program using just a programming language. Today, you use a programming language and a framework.
Consider PHP. By itself, it's a rather ugly programming language. But it's popular because of the rich frameworks implemented using PHP. In the i community, Zend is commonly used. Elsewhere, frameworks like CakePHP and CodeIgniter are popular. Other programming languages have their own rich frameworks: GWT, .net, TurboGears, Ruby on Rails, to name just a few. Lately, I've been implementing a web-based application using CakePHP, and I can attest to its convenience and power. But, as far as I know, the RPG community has nothing like these frameworks. IBM i does support a number of modern tools and languages. But then, unlike RPG, those tools are not limited to IBM i.
That is, when using an interpreted language like PHP, you have something that a language like RPG can never have: platform independence. For my own modest CakePHP application, I can run it on Windows and I can run it on Linux. I'll bet that it could also run on IBM i! In other words, by using many of these rich frameworks, you can run on many different operating systems, and so, you can render the choice of operating system platform irrelevant. To get back to the original question, should or shouldn't everything run on IBM i? I say this: It doesn't matter. And hasn't mattered for a long time.
Hans
To start with, I'm reminded of a couple of things told to me when I was job-hunting four years ago. First, a head-hunter told me that, at the time, the bottom had fallen out of the iSeries job market. Second, at a job interview, I was asked: "What's an iSeries?"
In the referenced blog, Aaron mentions that the IBM i has just seen 4 quarters of growth. That's good to see. I worked for 22 years on the iSeries and its predecessors, AS/400 and S/38. And I really don't want to see my good work go to waste.
But in his blog, Aaron suggests that everything should run on IBM i. The i is indeed a fine machine to store your mission critical data. You really can't go wrong choosing IBM i for your database. But I disagree that it should be used to run everything. (Those who like the i, tend to like it a lot, which says a lot about the system. The only people more devoted to their computers seem to be Apple fans.)
First, Aaron brings up the common strawman argument that the choice is between IBM i and Windows. Clearly, there are more choices out there, especially in the Linux, Unix, and BSD realms. For example, for the top tier in a web site, you can't go wrong with OpenBSD running on a couple of inexpensive Intel servers. You can easily configure OpenBSD to offer 100% availability, for at least your static pages, with rock-solid security.
But let's look at the application tier. Many shops choose an IBM i solution for one simple reason: The application they need only runs on i. If the application is implemented in RPG, then it's basically limited to the i. What if you need to develop your own custom application? Is RPG the best choice? Frankly, no. Thirty years ago, when I started working on the RPG III compiler, it was already an oddity among programming languages. It has advanced quite a bit since then, and I'm proud of my own contributions to its evolution. But it has not kept pace.
Is there an ideal choice of programming language? Again, no. Actually, I think that's the wrong question to ask. Today, when choosing a programming language for a particular project, you don't look at the characteristics of the programming language. If the attributes of the language were a primary consideration, Python should be the most popular programming language on the planet (in my opinion). Today, one does not program using just a programming language. Today, you use a programming language and a framework.
Consider PHP. By itself, it's a rather ugly programming language. But it's popular because of the rich frameworks implemented using PHP. In the i community, Zend is commonly used. Elsewhere, frameworks like CakePHP and CodeIgniter are popular. Other programming languages have their own rich frameworks: GWT, .net, TurboGears, Ruby on Rails, to name just a few. Lately, I've been implementing a web-based application using CakePHP, and I can attest to its convenience and power. But, as far as I know, the RPG community has nothing like these frameworks. IBM i does support a number of modern tools and languages. But then, unlike RPG, those tools are not limited to IBM i.
That is, when using an interpreted language like PHP, you have something that a language like RPG can never have: platform independence. For my own modest CakePHP application, I can run it on Windows and I can run it on Linux. I'll bet that it could also run on IBM i! In other words, by using many of these rich frameworks, you can run on many different operating systems, and so, you can render the choice of operating system platform irrelevant. To get back to the original question, should or shouldn't everything run on IBM i? I say this: It doesn't matter. And hasn't mattered for a long time.
Hans
Subscribe to:
Posts (Atom)