Mark’s rant is red and meaty, just the way I like it, but in the end I have to disagree.
The fundamental wisdom at work here is not, “stay away from proprietary platforms”; it’s “don’t marry the platform… any platform”. If you think that open source is insurance against your technology choices going wrong, I have to ask: how many job openings do you see these days for Tcl/Tk hackers? And doesn’t Perl seem to be going down the same path? (And no, I am not a Perl hater; on the contrary, I am a Perl lover who’s feeling kind of unrequited these days…)
I’ve been itching to dig into Flex (darn SCWCD certification still to do first…), and I feel fairly confident that the time will not be wasted. After all, at its base Flex is just ECMAScript and XML. Learning Flex is, in fact, good practice for programming the next generation of browsers: it’s the first available implementation of ECMAScript 4, the standard to which the browsers will eventually hew. And learning a new XML dialect is something that a professional programmer is going to have to do over and over and over in the decade or so to come. So where’s the risk?
That Adobe is going to hose the runtime? Backward compatibility with gazillions of Flash apps is their main selling point; and keeping it free-as-in-beer is how they (okay, Macromedia) established their amazing level of market penetration. Why the hell would they give any of that up? And they are open-sourcing the SDK, so no sale there either. Where’s the downside, honestly?
(Now Silverlight is another matter; Microsoft has neither an installed base to appease nor any commitment to open source anything; the only thing keeping them semi-honest is, well, competition from Adobe. So I’ll be steering clear of that.)