Doug Kaye is talking about Sun being one of the Web Services losers and Jason Whittington mentioned recent troubles at Sun Microsystems.  Sun held a lunch briefing for my employer and one of the leads on my team went.  According to him, half the briefing sounded more like a stock broker pushing the latest pick [1].  The other half was spent telling attendees not to do web services.  The recommended alternative was JINI.  I believe that JINI means mobile code, which is one way to interop, I suppose.  But I think that it also means an entirely new addressing scheme.  So it sounds like JINI is services, without the web.

[1] Did you know that SUNW supposedly has $5B in the bank, no debt, and a market cap of a around $11.6 B.  You could buy Sun for $6.6 billion, and own Java.  Seems like Java's worth 6 billion.

Wow, that's pretty more powerful and certainly leads to the rapid Refactoring that they have always talked about. This example is also a good one of just taking the problem semantics and "just typing it in." [Sam Gentile]

That's what really struck me about the Paul Graham essay that John Lam posted, and also when reading the 1st chapter of Graham's ANSI Common Lisp. Especially in the latter, he sounds like the XP people.  At least he's making similar claims (or is it the other way around?).  Some examples from chapter 1 of ANSI Common Lisp:

1. The phrase "rapid prototyping" describes a kind of programming that began with Lisp:  in Lisp, you can often write a prototype in less time than it would take to write the spec for one.
2. In the old model, bugs are never supposed to happen.  Thorough
   specifications, painstakingly worked out in advance, are supposed
   to ensure that programs work perfectly...The problem with the old model was that it ignored human limitations.  In the old model, you are betting that specifications won't contain serious flaws, and that implementing them will be a simple matter of translating them into code.  Experience has shown this to be a very bad bet indeed.
3. Powerful tools decrease risk, and so decrease the need for planning.

Or, look at a book like The Little Schemer, and observe how the authors create programs: the implementations for a particular algorithm evolve from the concrete to the abstract; i.e. Refactoring.  Except that the big difference is that the language supports genericity on a level that Java and C# don't.

One of the complaints about using "alternative languages" is that you have to learn a new language.  It occurred to me that really good C++ is a totally different language from what's usually written.  I think that's true for most of the popular languages.  At least I learned C++ twice, once as a "better C", and again after reading Scott Meyers and the "Gang of Four", and a lot of that was discovering templates.  I've gained a lot from switching to Java and C#, but Sam's reminding me what I've sacrificed.