5 - 6 - 98

Programmers? We Don't Need No Stinking Programmers!

by Willy Chaplin

My wife and I used to lament this fact after I had completed each programming contract. Once you have a program, who needs a programmer? The fewer bugs in the finished product, the more nearly true this statement is. I mean, when was the last time you gave a passing thought to how your word processor or flight simulator could be improved by a "good" programmer? These products are mature programs and mature programs need no maintenance nor enhancements.

This must mean...and it does...that a lot of software currently in use is NOT mature. As if you didn't know that! One of the reasons for this is due to the intense competition in the field of software development. Companies feel compelled to offer their wares for sale long before they have been completely debugged. Also, the software itself has become so complex that it resembles a chaotic system...like the weather...more than a finely crafted radio or automobile.

Given enough time, a piece of software that is in continual use by a large number of people will eventually iron out most, if not all, of its flaws. Like word processors and flight simulators. Two forces conspire to make even this maxim false in our current climate. First there is our old friend, competition. Competition for market share requires competing firms to offer ever more features to their products, in order to differentiate them from that of their competitors. This adds to the chaotic nature...since new features have to be debugged as well...plus the fact that the product must still support all the old features, perhaps integrating them with the new as well.

Then there is hardware innovation. Moore's law about processor power applies, in some degree or other, to every type of hardware. Disk drives become faster, smaller and hold more data. Scanner and printer output becomes bigger, while the machines themselves get smaller, faster and with more resolution. And so on and so on. Still, the laws of physics require that this process must someday end. In fact, we are probably within a decade of seeing the end of this type of innovation.

The so-called "productivity paradox" is one of the consequences of this mad race for smaller, faster and better. That is, while companies have been investing vast amounts of money and time into computer hardware and software for the past two decades, the numbers for worker productivity, especially among those who should benefit most from this technical advance, show little evidence of increase. Indeed, the numbers show that the average worker is working more hours, not fewer to achieve the "same" results!

Of course, since productivity numbers by necessity can not include newly created "value," in the form of things that simply could not be done at all before the computer, but can only measure how fast previously defined tasks are now being done, they understate the true worth of these new devices in the workplace. But, it seems apparent that the continual product upgrading of both hardware and software, requiring continual learning on the part of the workers, also has a large effect on the outcome.

The fact that hardware innovation must soon end implies that software products will also soon have time to catch up and become mature. And so it shall. Within a fairly short time, those of you who are sick and tired of learning some new procedure every six months will have some breathing space. Not that this means the end of invention in the computer field. Such a revolutionary change in the way we "do business" will continue to force the evolution of enterprise, culture and the very nature of society, much as the industrial revolution did two centuries ago.

But, it will mean the end of programming. At least the way this word has been used in the past. The useful days for people like me, who have worked with and understand the inner workings of these little beasties are nearly over. I expect "computer science" will move to the history department while those you actually work with the devices will become networkers or other types of information laborers.

In short, in the Great New Age of Information, the very people who have made it all possible...the hardware and software geeks...are rapidly working themselves out of jobs. For the processes most amenable to automation, the work most easily done by a machine is that which we, the dweebmeisters, do. I have known this for as long as I have been working with the machines. For what is assembly language, but an "automation" of machine language? What are FORTRAN, C and BASIC but simplifications of assembly language. Now we have JavaScript, HTML and Objects. Soon, creating "new" programs will be as simple as playing with tinkertoys, available to all but the most technically backward.

The only thing that has kept the field going as long as it has, the only force keeping thousands of programmers in pizza and beer...has been the continual frantic innovation of the past two decades. So how does this bode for the young, trying to divine what the future holds in the way of paying work? Well, for one thing, it strongly suggests that becoming a "computer professional" is about as promising as becoming a buggy whip expert was at the turn of the (20th) century. Just like horses, computers are not about to disappear. But, there are already plenty of trained personnel to deal with the "expert" end of it. As time goes on, fewer and fewer will be needed.

But then, what about those thousands of other professions that have been momentarily eclipsed by the "Information Revolution?" I don't see that plumbers, carpenters, gardeners, street cleaners or child-care specialists are going to become obsolete any time soon. Yes, us computer types have been able to automate and advance even these professions...slightly...but their essence remains exactly the same as it has been for centuries. If the future follows the past, we should expect the interest in and compensation for these "mundane" tasks to once again become equal to or surpass that of the information worker.

Given the Internet and the World Wide Web, there is at least one profession that is about to experience a tremendous revival. Which one is that? How about librarian? I can think of no single task which will take on more importance in the immediate future, than that of the organizing, identifying, sifting and sorting of the vast amounts of information that we of the Order of Bits and Bytes have created. Since all this stuff is only going to be useful to us humans if it can be found and understood by humans. Who better to carry out this task...who else CAN carry out this task...than the lowly...no LOFTY...librarian? Long live the Dewey decimal system!

Talk to you later...

Willy Chaplin is a man who calls himself a libertarian and thinks he has something to say to us all. This rant was previously published on How Can You Laugh at a Time Like This?