But eventually I became convinced by Philip's discussion of how much more appropriate a scripted language is for web applications than is a compiled language like Java. Philip was particularly derisive about Java "junkware/bloatware" like BEA/WebLogic.
Your derision will only increase if you get to know some WebLogic engineers. They're like cute little bundles of misinformation.
So what happened to this basic architectural belief at aD in the meantime? How is aD's new direction different from what Philip was once criticizing? What about the Tcl argument is no longer compelling?
What's the key word in your question?
"architectural". Ignore the languages, just look at the architecture. Where's the basic belief change? There's no sea of middleware APIs to wade through, no extra servers to support, no n-tiers of applications introducing subtle errors into my data stream. The major difference (after the increase in orthoganality that is ACS4) is language - which is just a tool. We could write ACS in
Unlambda if we so chose.
Why not Tcl? I think someone here said "At first you hate it, then you get used to it." It's only natural to look for a more expressive language, and of the industry 'accepted' languages, Java sucks least - less than Tcl, C++, Perl (for large teams), and C (if you're not bashing bits). Plus there's the toolset - both Aolserver & Oracle run Java already.
I think the Tcl argument is still compelling, but the devil is the details (and in C programs, from what I hear). And in these details, Java trumps Tcl. Personally, I'd like everyone here to learn to love Ruby, but we rarely get what we want. Also, I worry about *any* community acceptance of ACSJava - Java hasn't been shown a very warm welcome by OSS programmers (in my experience - someone please prove me wrong). Probably this is largely the result of oversold hype: every statement about Java is usually followed by a caveat; Java sits midway between Smalltalk & C, excelling in the features of neither; trying to fulfill the hype, tens of thousands of programmer-years have been spent to reduce Java's suck factor - a full-on intellectual battle to drag Java above the waterline of mediocrity.
Obviously I don't have any numbers, but I thought that community support has been building up through now. I wonder what kind of change will happen when ACS becomes solely the domain of a language many have shown an aversion to accepting.
