Just a Few More Vacation Thoughts

Wow. I really didn't expect this kind of response. I see that my original Flash rant is up to 29 comments. I'm a little scared to read the ones I haven't read when I finally get back from vacation. I really seem to have pissed off a lot of people. That honestly was never my intention.

To some extent, I regret the "Flash Sucks" part of the discussion. Not because it's necessarily untrue, though: It's my honest opinion of Flash on Mac OS X - it's leaky, crash-prone and, well... sucks. But, I regret it because that statement has detracted from the real discussion that's been going on underneath it all. It seems to me that many Flash proponents are trying to write me off as "just another run-of-the-mill Flash Hater" predicting the demise of Flash "yet again". The same old story, nothing to see here. Move along, move along.

But I'm not a run-of-the-mill Flash hater, and I'm not predicting Flash's demise because I hate it. I'm not even predicting Flash's total demise, just the demise of Flash cum general purpose web-development tool. And my predictions are the result of a fairly thorough knowledge of the history of web development and a reasonably educated guess about where things are likely going next.

And you know what? Flash is the completely wrong approach for a general-purpose web development tool. Flash's current approach is anathema to what has become fairly widely recognized as good web development practices. The fact that Flash is fun or easy, or that the tools are great doesn't change that it is fundamentally flawed approach to doing web development.

And, believe it or not, I'm going on about this for the opposite motivation from hate. I'm dwelling on this because I don't want the many excellent Flash and Flex developers I've met to get screwed if the walls do come crumbling down. I can tell you stories about programmers I've known who spent a career working in PL360 or COBOL, only to find themselves without a job and with a skillset that was completely obsolete and not marketable. Take a look at this list and tell me how many languages you recognize. Languages go obsolete all the time. Some of the ones that have gone obsolete were bigger stars in their day than Flash is now.

Right now, we have fifteen years of web development evolution to look back upon. It's now a fairly mature space, and we are at a point where we have a pretty good idea of what things have worked and are sustainable and scaleable. We also have a pretty good idea of what things have not worked. And though you might disagree with me, when I look at the web technologies that have survived, the Flash plug-in is an anomaly. If the web is headed where it looks like it's headed, it's hard for me to see how Flash can continue to remain as relevant as it has up until now. That's not hate, that's just reality.

You can do so much more with HTML and Javascript today than you could a decade ago, or even a couple of years ago. With the exception of certain types of online games, the vast majority of what Flash is used for on the web can already be done in HTML and Javascript without waiting for HTML5 (combined, of course, with server-side development of some sort). The situations where Flash is truly necessary are dwindling, and Flash's uptake in the area of greatest growth (the mobile web) has been lackluster at best. "Dismal" is probably a better adjective than "lackluster" to describe the mobile Flash uptake.

There might be a silver lining here, but I don't see it. Adobe's done too little too late when it comes to opening up Flash, and has done too little too late in the mobile realm period. Adobe can send out press releases about "billions of Flash-enabled mobile devices" till the cows come home, but it's simply not reality. It's just marketing types playing with semantics. Companies only mince words when they have something to hide. To a consumer, a Flash-enabled mobile device is one that can access most any Flash content that's anywhere on the web just as if they were at home on their computer. There aren't billions of devices that can do that. There aren't any devices that can do that yet.

And think of this. If Flash somehow maintains or increases its prevalence as a web development tool going forward, I'll have been wrong in a couple of blog postings on the Internet. If I'm right, an awful lot of people are going to be finding their skills less and less relevant and less and less marketable.

I've been wrong before. I can live with being wrong. But what if it turns out I am right? What if it turns out that after 26 years of being online¹ and keeping abreast of the technology involved in online communications I actually know something about it? What if these rants aren't the result of hatred but are rather just an honest assessment of Flash's current place in a complex and ever-changing technology landscape? If you're a Flash developer, especially if Flash development is your only marketable software development skill, well… just make sure you can live with me being right. I'll continue to be able to put food on the table if I'm wrong about this. Please make sure you can say the same thing.

Just as an aside: I left my last full-time gig (at PeopleSoft) in 2002. I'm not sure if you remember, but 2002 was a pretty crappy economy. Not as bad as today overall, but it was a pretty rough time for programmers and software consultants because the dot com boom and subsequent bust had created a glut of tech workers. I started my first project as an independent consultant only four days after walking out the doors at PeopleSoft (two of which were a weekend). After that project (knock on wood) I've found steady work with very little unplanned downtime. The work hasn't always been in my preferred language or using my preferred tools, but I've always found work quickly because my most marketable skill is that I have a varied toolbox, an understanding of what tools exist outside of my toolbox, and a willingness to expand my toolbox if that's what I need to do to get a job done right. Languages and environments come and go. Don't tie your fortunes too tightly to something that can become obsolete unless you want to risk becoming obsolete yourself.

---

1 - I got my first modem in 1983, a hand-me-down acoustically coupled modem from a friend's dad who was a college professor. It was replaced less than a year later with a Hayes-compatible 1200 baud model. I had my first internet e-mail address in 1986 (a fidonet address through a BBS I called using PCPursuit), and I got first true Internet account in 1988 (a Unix shell account through my university that I accessed using Kemit). I had my first PPP internet account in 1992: a metered account through a company called holonet.

Posted in: Rant