SpencerBarron.com

or at least trying to. I’ve been busy expanding some ideas I have with internet technologies. As such, I haven’t been completely focused on writing much lately but I’m still here. I’ve been working on expanding my very limited programming background. I’ve been learning web development and have been trying to pick up as much as I can before it all changes. Outside of the classes I’ve been taking, I think I’ve actually made some progress. I don’t really enjoy formal classes. I find that it’s usually structured in such a way as to not teach you enough to move forward on your own. When it comes to technology, it seems a typical local college just doesn’t have the ability to keep up with what’s new. And of course, my biggest peeve: colleges like to separate crucial components in such a way so as to fill their 30-40 credit requirements for a certain program of study overcomplicating the learning process so they can sell you more classes. You would think you would go through a class that covered the subject broadly early on, perhaps building a web application without going into much detail of every little thing. That way, as you continue down the road, you’ll have a better grasp of where each technology comes in to play. Of course, if you knew that, you could probably just pick up a book on the subject.

So what have I learned so far? I’ve learned enough C++ to know that I hate when things become very tedious. That really helped with pushing me towards learning more about Ruby and Ruby on Rails. I was able to use C++ for an actual application though. I wrote a parser to convert an online directory of addresses to a comma delimited file so it could be used to create mailing labels. Pretty exciting stuff isn’t it.

I learned more HTML and CSS. No time wasted there. Besides all the CSS you come across playing around with blogs, you can also use it to help with designs of actual websites. Who knew? Hopefully, my site, ValueRents.com will be online sooner than later. That’s the website I’ve been working on in a round about manner for a couple of years. I dusted the concept off and developed a frontend for the site but still need to plug in some sort of backend. One day ValueRents.com will help landlords find the best properties then manage their rental properties from one location. I apparently have a few more things to figure out before it all comes together. I might make some attempts to redevelop the theme for this site. How hard could it be?

And to wrap it all up, a few classes in database design and SQL. I got to tell you, there’s nothing like bending raw data to your every whim. Hours of fun. I can’t even begin to explain how happy I am to finally stop using Excel to run calculations on the thousands of foreclosure records I have. Mysql and SQL Server are so much more fun. So maybe I can start using Excel for what it was intended–spreadsheets and quick and dirty charts.

So that’s what I’ve been up to while the real estate megalith cracks and crumbles. I’ll have more on that in coming posts. I can’t help but think though, where there are problems, there is an opportunity. I haven’t seen such great deals in Denver real estate since I moved here 8 years ago. Hopefully, the web development education I’m giving myself will help me take advantage of the dozens of opportunities I see every day. Time will tell.

Google “RPG”, and your first hits will be about role playing games. But before there were Role Playing Games, there was RPG (Report Program Generator). This was the name of a computer programming language developed in the 1960’s by IBM. It was designed to generate reports (thus the name) in a terse, simple way. It used a cycle of input, process and output that was built into the language, so that you did not have to issue formal instructions to the computer to read the data files; you just described the file using fixed -format File specifications and Input specifications, entered the necessary Calculation specifications, then output the report in the format described in the Output specifications. Your humble correspondent, without a lick of Computer Science education, learned the language (RPG II) from the manager of a small RPG shop (in a class he held in his kitchen) well enough to write non-trivial programs in a matter of weeks (one class per week). It was easy.

Today, that language has become RPGIV, reflecting continuing enhancements that make it an extremely effective language for manipulating data. You will not likely use it to write an operating system or make pictures dance around on your screen, but computers used by tens of thousands of businesses have it as their base computer language, generating reports and entering and displaying data.

The future of RPG today, though, is in doubt, even though programs totaling billions of lines of code are running today on hundreds of thousands of machines. This is a reflection of the state of computer science today, as well as an example of the fact that, too often, what is often serviceable and effective is set aside.for what is fashionable.

In programming, the fashion today is centered around the Web, and languages associated with the Web - Java, Perl, Python, Ruby, C++, Visual Basic, Javascript, and so on - visually oriented languages, for the most part. Can RPG do the Web? Yes, but you have to jump through hoops to make it work. There is no facility in RPG that allows it to use HTML as effortlessly as its customary screen display files. And there are forms of RPG from IBM and ASNA that resemble VB and Java with their Windows displays, but they are by no means nearly as ubiquitous as VB. The kids learn VB or teach themselves Java or C- they do not learn RPG. It is not sexy; RPG is like its cousin COBOL - they are functional business languages, but Windows and Web processing have to be grafted onto them to work on Windows or the Web. RPG can be used in conjunction with Java on the AS/400; it works quite well to use Java for the user interface and RPG for the core business logic; but people tend want to forget the RPG and use something (anything) else.

In short, all RPG has going for it is that it works. It runs on a machine (the AS/400 or iSeries or Series i) with a database (DB2) incorporated into the hardware- no extra database software, though it can run SQL (and RPG can call and run SQL statements). The operating system (OS/400) makes even Linux seem to run flaky by comparison, without the need for systems administrators, and the machine it runs on seems to run forever. There is no such thing as the “blue screen of death”; at least, I haven’t seen one in almost 30 years. (For a long time, Microsoft ran its business on the AS/400, well into the latter half of the 1990’s.) RPG can call C++ and Java programs, and they can call RPG routines. It can also run HTML, with a little bit of extra effort.

Perhaps RPG and the AS/400 are too good at what they do. You can run 35-year old RPG programs unchanged using RPGIV. (Try running 1980’s vintage GWBASIC programs unchanged using VB. It likely will not be easy, perhaps not even possible.) Planned obsolescence is not implemented, as what seems to be true in the programs of Microsoft and other software producers. (And frankly, I believe IBM has done the right thing, purely on principle. Programming should be done to benefit those who pay for it to be done, not to entertain the programmer. Change the program if you need to improve or fix it, not just for the sake of change.)

RPG is being taught by very few schools; some community colleges have given up on their RPG-AS/400 courses due to lack of interest. Few new programmers are being educated. The RPG programming population is getting grayer and grayer. Though it can be used quite elegantly to produce powerful data-crunching programs, the young are not learning it. Crunching data is not entertaining. On RPG forums the topic of the future of RPG is argued quite violently; perhaps it will survive as long as people care about it. It would be a shame to see vast numbers of programs rewritten because people no longer care to use this highly functional language.

I find it difficult to believe computer science grads find it difficult to learn a language that I could learn in a very short time under less than ideal conditions. (I typed my first programs into the computer on 96-column punch cards and fed them into a card reader.) I have read opinions of obviously educated people who feel that the RPG logic cycle and the use of RPG “level break” handling (of processing when a designated field changes in value) is incredibly complex; I read about it in my RPG textbook, said “Hey - that’s pretty cool” and didn’t need to give it any further study. It is incomprehensible that these people find it difficult. Evidently, it’s sometimes easier to learn something when you know nothing about the topic than when you think you know something about it and your mind is halfway frozen shut.

I wish I had a solution; far wiser heads have yet to come up with a solid solution for the growing shortage of RPG programmers. But it would be such a waste if someone doesn’t.

 Guest Post - Curtis Barron

When Bill Gates was 13, a parents group enabled his school to buy a Teletype machine and computer time on a remote GE computer that used the BASIC programming language; he, Paul Allen, and others became intrigued by the technology, including obviously the immediate feedback they got from programming in an easy programming language. He went on to learn other languages, found Microsoft and become a gazillionare.

Does that mean that if your sons or daughters learn about computers using BASIC that they too will earn googol dollars? Well… probably not. If not, might they understand the computer better if they do? Well…. maybe.

Being interested in programming by profession and by inclination, I came across by sheer chance an article in Salon.com entitled “Why Johnny Can’t Code” by David Brin. By 12/31/06 it had attracted 297 comments and over 300 blog reactions. Read the rest of this entry »