|
What does the future hold for software developers?
Some people seem to love bad news. And I must confess, reading the news and blogging every night probably hasn't made me a cheerier person. If it isn't terrorism, it's the war on civil liberties. If it isn't war, it's the economy. At least the job market is going great guns <g>.
A couple days ago I ran into this gem, a depressing riff on the future of software developers by Phil Wolff, which he wrote after reading an InfoWorld column by Bob Lewis. Lewis talked about how programming jobs are being exported to India and other countries where there are high quality programmers who will work for substantially lower wages than American developers. Lewis' advice to software engineers: Find a different field of endeavor. Unless you're in the top rank, there's little future for you in IT. [of course that doesn't explain how to deal with engineers' seemingly congenital belief that they are all in the top rank.]
Phil Wolff's riff has a lot of good reasoning as to why it is easier than it has ever been to export programming jobs overseas, which can be summarized in one word - Internet. Plus there are an increasing number of good software engineers overseas. While he makes the point that technical people with so-called soft skills (like project management) or jobs that require a lot of face time don't face as much competition, he basically concludes that Lewis is right, and that American software engineers are going to face rapidly increasing competition and downward wage pressure from overseas engineers.
I've been noodling on this for a little while and decided to try to get some unformed thoughts on this out there into the blogosphere to see what others think (one of the advantages of writing for free is that you can do stuff like this, instead of waiting until you have all the loose ends tightened up.)
It seems crazy to be predicting a long-term surplus of software engineers when three years ago I was offering newly minted college grads what seemed to me to be obscene amounts of money plus bonuses to come work for my company (and they thought they had somehow earned the right to that much money). And we have heard this warning before -- I have a book in my bookshelf (unread, I confess) called The Decline and Fall of the American Programmer, published in 1992, that warned of exactly this future. It certainly didn't come true in the late 1990's. I have learned to be very wary of trying to predict the future using reasoning - there are always factors that you don't account for which end up having huge effects. People thought the internet would be big in 1996, but who would have predicted the excesses of the internet boom before Netscape went public? Finally, my career in high-tech has all been with startups working on new products and new technologies, where the spec and the dates get revised frequently, and the ability to communicate and to change rapidly when new data comes in (or the boss gets some bright idea :-) is critical to success. Attempts to outsource that kind of work seem doomed for failure. I can't imagine having outsourced the development of a new embedded OS for instance, although I did successfully outsource things like the IrDA stack, and had an outside company do a great job. In my experience, one of the keys to success in outsourcing is tightly defining the deliverable, and understanding the costs, direct and indirect of making changes to the deliverable. In running a consulting business, one of the most lucrative though frustrating situations is having a client who constantly changes his or her mind. But when you develop something that really is new, you have to change a lot on the way.
It seems to me that there will be downward wage pressure from the availability of qualified software engineers overseas, as well as from the availability of high quality Open Source software, which none of these authors mention. But the likelihood that the American Programmer will "decline" is dependent on how fast the rate of change in the software business is, and if it remains led by the US. I have worked through two booms, the rise of PC software in the mid-1980's and the rise of the net in the latter half of the 1990's, and in both cases the rate of change in everything that a developer needed to know was extremely high. New tools, new languages, and new platforms rose and fell by the wayside rapidly. Large numbers of new products for users were imagined, developed, marketed, copied and sometimes sold. Companies grew rapidly and failed rapidly. These were not conditions under which managers could outsource work, and it is hard to imagine that changing. I also worked through the doldrums of the early to mid 1990's, where it seemed like most development was incremental or me-too, and the lawyers were the one who got to be creative (remember the Lotus-Borland and Apple-Microsoft look and feel lawsuits?). A lot of that development could easily be outsourced. So which is the future going to look like? I don't know, although I fear that given the economy and the excesses of the internet boom, for the immediate future it will look a lot more like the doldrums. However, I'm betting that the roller coaster will come back around again in a while, and I'm ready for another ride.
Sorry about the length. I'm gonna learn how to do that >>>more<<< thing.
|
|
