When I was a lad, I wrote an implementation of Scheme in a new-fangled language called Java that nobody had heard of. You certainly couldn't get a job in a bank with Java experience at the time.
I'm glad I did so, the Java experience gave me enough money to take time off my career and the Lisp experience gave me something interesting to do with the time off.
Slowly, yeah. Slava is still working hard on improving it and the community is slowly expanding.
However, Factor still has one problem: it takes a bit of a time investment to get used to the Factor idioms and until you do, its horribly difficult to code in. Once you do though, its an amazingly powerful language. Pity I need to spend a little longer at it.. time which I don't currently have to spare. Hopefully in a few months though :-P
Hm, yeah. Concatenative languages haven't had much attention, undeservingly. I'm going to keep an eye on Factor I think. I'm already plunging into functional programming these days though, there are so many languages to learn!
I'd say there's some passion and curiosity about concatenative languages. (Summary for those too busy to read it: concatenative languages are direct implementations of combinatorial logic).
I played around with Forth a while back, and while I thought it was a great experience and an interesting way of 'seeing the world', I think that keeping track of what's on the stack seemed like a route to madness. Even stuff like C doesn't make you do that.
The Factor community likes to use the following phrase to answer that question and I think its quite correct: "Idiomatic Factor does not use the stack".
That is, well written idiomatic Factor makes use of higher level abstractions which hide the stack from you. The stack should not be used as a data structure, but rather as a scratchpad to allow pointfree programming. Factor provides a load of abstractions which can often handle all of that low level stack handling stuff for you, making the code easier to read, follow and understand (once you know the abstractions). Factor also provides local variables, for use when it makes sense to name your data. Its not shunned to use them over the stack - when it makes sense to. Finally, Factors macros allow you to modify how the code is parsed (Factor calls them "parsing words"), so you can actually change the syntax. Core Factor provides an "infix" parsing word which allows you to write infix expressions and it is advised to do so when it makes sense (eg, for mathematical formulas). Another example of parsing words put to great use is Jeremy Hughes' inline C/C++ library, which basically allows you to write C/C++ code inline in Factor source files. During compilation, Factor will compile this code with GCC and simplifies writing of FFI stubs for wrapping C/C++ libraries.
Basically, Factor is very different from Forth. Slava likes to say that "Factor has more in common with Lisp than it does with Forth" and its often said that Factor only has a syntactical resemblance to Forth, but is otherwise very different.
My quick take, from playing around with it a bit a few months ago: you can put much, much cooler stuff on the stack. Closures, for instance. (Or at least the Factor equivalent of closures.)
Oh, I was able to get coding very quickly too. The REPL is very good for learning and development. Still, I find it dificult to read other peoples code and some problems take me significantly longer to work out in Factor than another language. Having said that, I'm not familiar with a lot of the common higher level abstractions, which I really need to learn to make the most of Factor. I think its easy enough to get started, but dificult to become fluent. YMMV.
