... From: "Ehrenberg Daniel" <microdan@...> To: <concatenative@yahoogroups.com> Sent: Monday, February 28, 2005 5:46 PM Subject: Re: [stack] The annoying...
stevan apter
sa@...
Mar 1, 2005 11:31 am
2277
... it's easier than i thought. consider ; +/- [1 -1] * + ; ; quad { [a b c] b -: a c * 4 * b 2 ^ -. _sqrt +/- a 2 * % } ; ... the quadratic formula in XY....
sa@...
Mar 1, 2005 2:55 pm
2278
... An ideal program speaks the same language as the experts who know how to solve the problem at hand. If your problem is mathematical, you want to speak the...
... Heh. :-) Writing a customized language is a good approach for every problem. It's the core of working with Forth; every decent-sized Forth application...
... Thoughtful hackers hear about Unix and try to use it. Ordinary hackers hear about Unix and mess about with it a little. Thoughtless hackers hear about Unix...
... Oops. That's http://www.ccil.org/~cowan/upc , verse 41. -- "As we all know, civil libertarians are not John Cowan the friskiest group around -- comes...
i have, on occasion, found it useful to implement a sublanguage as part of an application, but in every such instance the rationale was that that part of the...
sa@...
Mar 1, 2005 6:53 pm
2283
... We can't make a minilanguage for every domain. If we're making some sort of grammar checker, do you expect people to be able to program it in English? ... ...
... That wouldn't be my question; my question would be whether the sublanguage was closer to the solution domain. I'm not saying that you have to ask the same...
... No; I'd expect to be able to program it by talking about "indirect objects" and "antecedants". I'd expect to see talk about unification parsing, or some...
in k: 0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1 1.0 / looks like 1.0, but: 1~0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1 0 / doesn't exactly match. however: ...
sa@...
Mar 1, 2005 8:48 pm
2288
oops - type less, think more: 1.0~0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1+0.1 1 1.0 does indeed match the summation. but if we set print precision to maximum: \p 0...
sa@...
Mar 1, 2005 9:02 pm
2289
... Dan's code uses Factor's built-in mathematical operators, which support arbitrary precision integers, arbitrary precision rationals, complex numbers, and...
... It does help, but of course it isn't related to numeric instability -- unstable algorithms err by enormous amounts. I know you know that (since you work...
... Not enough -- it all reduces to floating point as soon as you perform the sqrt. And from then on, you've got to watch out for numeric instability. But...
... I don't understand. I'll talk a while to reveal the magnitude of my incomprehension; please step in and correct me. By what definition can you claim that?...
... From: "William Tanksley, Jr" <wtanksleyjr@...> To: <concatenative@yahoogroups.com> Sent: Tuesday, March 01, 2005 4:30 PM Subject: Re: [stack] The...
stevan apter
sa@...
Mar 1, 2005 10:53 pm
2295
it's worth pointing out (i guess) that while shuffle-notation can be eliminated by translating abc-def to stack operations (e.g. swap, pop, dup, &c.),...
sa@...
Mar 2, 2005 3:02 pm
2296
... Well, by "the compiler", I mean "all that program, or those programs, whose purpose is to take a source-code representation of some program and produce an...
... In Factor, immediate words have access to all of the internal workings of the compiler and the parser, and they have some limited access to the internal...
... Actually, Factor's immediate words have access to the parser, and can affect how a syntax tree generation is created from source. The compiler on the other...
Slava Pestov
slava@...
Mar 3, 2005 2:07 am
2299
I've been reading this list for some time, but haven't posted. I have been wondering about something, however: Can Joy or Factor or some other Joy-like...
... Yes, the advantages can carry over; no, not all concatenative languages choose to live within the restrictions that allow such small size. I think that...
... I think I understand your definition. I need to step back for a moment and gain a larger picture, because I've forgotten why I originally said what you...
... The Factor runtime is pretty small (60-90kb on x86) and the image can be stripped down to around 300kb before it stops being recognizable as a Factor...
Slava Pestov
slava@...
Mar 3, 2005 9:21 pm
2304
... This last is how the Joy[01] interpreter works. -- Do what you will, John Cowan this Life's a Fiction...
... You're right, so long as we don't know the stack effect of the words. If we know the stack effect, we can perform the translation easily (that is, the...
