When will Delphi die?

I have a theory. As with all my theories, it probably doesn’t stand up to close scrutiny but I’ll put it out there none the less. And this is it. The more trouble a company is in, the more marketing emails they will send out. and if they are in really big trouble, then they start phoning people. 

If this theory is in any way correct then Embarcadero (purchasers of Delphi) are in big trouble, or their Delphi division is at least. Not only do I constantly receive emails from them or their partners telling me how great the new version of Delphi is, but I’ve also had at least three phone calls from people desperate to sell me Delphi licenses. Unfortunately I have no interest in buying Delphi licenses, since I haven’t touched Delphi code for years. Ten years ago Delphi was a great product, it beat Visual Studio hands down for its power to create Windows apps. But along came .NET, which in its original incarnation was fairly decent and has been getting better and better as time goes on. Now Delphi is essentially an irrelevance, except for people needing to maintain an existing code base. Who in their right mind would choose Delphi to build a new product? At this point I was going to direct you to the JobStats page showing demand for Delphi skills in the UK but they don’t even bother to list it anymore. There may be a small niche area where native compilation is required for performance reasons and Delphi may fit the bill, but I suspect this niche is getting smaller and smaller. Another problem that Embarcadero have is that it is almost impossible to make money from development tools. Microsoft essentially provide their tools as a loss leader and also provide damn good free versions. There are also lots of other good free development tools out there, so why pay for a development tool at all, particularly one for an obscure language that is no longer cool and happening? To be fair, you can pick up a fairly decent version of Delphi for free.

So the only question I have is when will Delphi die? I guess, like Cobol and FoxPro and a hundred other seemingly redundant technologies, it will continue to trudge along for a long time yet. Or perhaps I’m completely wrong and there is some place where Delphi can still be a winning technology, but I can’t think of one myself.