programmersI talked with several of my former students who are database programmers over the past few weeks. More and more of them are looking for work. There seemed to be a general consensus that these jobs would reappear as soon as the economy recovers. I hated to be the bearer of bad news, especially since I did my best to make them database programmers when they were my students, but these jobs aren't coming back.

The world has changed. It's not that the need for data collection and processing has diminished; it's that in large part, these jobs have been automated out of existence. Ten years ago, if you needed a million lines of database code to support a large database, several programmers worked more than a year to create that code, and often several programmers were needed permanently to maintain that code. That is simply not the case any more.
In the last ten years, code generation has come of age, and the first places to be impacted are the coding activities that followed simple processes. In 2000, a million lines of code represented ten programmer years and an expensive management structure. Today, one programmer can create that same structure in a few hours. The job now simply requires that one lay out the data structure and push a button. Instantly, you have either a complete application or at least the basic structure that everything else will be built upon. If the customer doesn't like it, I can go back to my computer, work for an hour, and show them a completely different implementation.

There are at least a half-dozen commercial products that do this job, and I have written several variations myself. It's a little tedious (like most programming jobs), but nowhere near as tedious as writing database code by hand. Not only does the customer save a fortune on programming costs, but programming errors with expensive consequences rarely get introduced into the automatically generated code.

I would think that any database programmer who has ever seen a code generator work would immediately strive to develop a different expertise (code generation, for example), but the more typical reaction is to assume that the same people who wanted their skills in the past will want them in the future. They are not actually hungry because they have made a good living for many years, but they are going to be very lucky to get anymore income out of their current skills.

The same thing is happening in all kinds of programming. I watched a demo from Alphacet last week which was quite impressive. We had a simple system consisting of several rules, and the Alphacet representative was able to turn those into an automated system AND do some simple backtesting in less than ten minutes. I could easily come up with rules that I would still have to program by hand, but they were prepared with all of the common structures and market data cleaned and organized. I could get the code generated in my choice of six different languages. The bottom line is that another world of good paying jobs is going to disappear.

What is on the horizon that will replace these jobs? I have no idea, but I spend more and more time thinking about code generation and how to maximize its use. We used to have several programmers on staff, and now we do all that work ourselves. For us, that made the difference between able to survive this downturn (so far, knock on wood) and drowning in our own payroll. For my friends in the programming community, it means that there are hard times ahead.


WordPress database error: [Table './dailyspeculations_com_@002d_dailywordpress/wp_comments' is marked as crashed and last (automatic?) repair failed]
SELECT * FROM wp_comments WHERE comment_post_ID = '4865' AND comment_approved = '1' ORDER BY comment_date




Speak your mind


Resources & Links