my paper on SLACK has reached "first draft" stage: http://www.nsl.com/k/slack/slack.htm and a summary will eventually follow. meanwhile, i continue to...
sa@...
Jul 5, 2005 2:58 pm
2501
While I was away I have been thinking about flatness, macros(or other definitions), and concatenative semantics. In what follows, by "standard languages" I...
... using ... flatness ... lazy, ... and ] ... level ... conditional ... Consider the following Joy program: LIBRA cpop == [pop] [] branch . cpop is the...
... Right -- I would go on to say that the reason concatenative languages are interesting to work in is that the semantics of the language (sequential...
... From: "William Tanksley, Jr" <wtanksleyjr@...> ... this hovers just beyond my ability to comprehend. [ and ] are functions. but functions of what to...
stevan apter
sa@...
Jul 14, 2005 11:05 am
2506
... now that i think about it, this probably doesn't work. if [ and ] are functions, then how do you produce the quotation [ [ ] i.e. the list whose sole...
stevan apter
sa@...
Jul 14, 2005 11:32 am
2507
the, or at any rate *one* answer is: since [ and ] are functions, quote them. e.g. in XY 2 3 \+ pushes 2, then 3, then + onto the stack. so [ \] ] [ ] ] [...
sa@...
Jul 14, 2005 6:13 pm
2508
i hope no one minds that i continue to post my thoughts on this topic. at first, i thought that flatness was inconsistent with intensional lists (quotations),...
sa@...
Jul 15, 2005 12:47 pm
2509
... To make sure I understand this: in Joy, the square brackets have two jobs: they signify both lists and quotations. You have separated these, so that []...
... Yes, I think so, but I think your way of putting it is confusing, given that in Joy "quotation" means a list suitable for execution. Square brackets ...
in XY (either version) [ \\ ] [\] that is, you can quote the quotation operator. in other words, quotation is first-class....
sa@...
Jul 15, 2005 5:18 pm
2512
... since all lists are suitable for execution, in joy, "list" and "quotation" are just different names for the same thing. i suppose you could make the...
sa@...
Jul 15, 2005 6:05 pm
2513
... Well, not quite. A list some of whose members are undefined symbols is not suitable for execution. ... My intuitions are quite otherwise: a list is a...
... not in Joy, i agree. in XY, the value of an undefined symbol x is x. e.g. undefined undefined [2 foo 3] i 2 foo 3 ... and ... so you think that quotation...
sa@...
Jul 15, 2005 6:53 pm
2515
... From: "William Tanksley, Jr" <wtanksleyjr@...> To: <concatenative@yahoogroups.com> Sent: Thursday, July 14, 2005 12:12 AM Subject: Re: [stack] stack...
stevan apter
sa@...
Jul 16, 2005 12:10 pm
2516
the following is a (slightly edited) conversation which took place on #concatenative earlier this morning. comments? ... <eiz> How is 'flatness' in any way...
stevan apter
sa@...
Jul 16, 2005 5:27 pm
2517
... Why? ... I don't think I made that claim -- did I? I claim only that a lazy concatenative language could be _perfectly_ flat (in the sense that any valid...
... Entirely agree (except you probbably did not reaaly mean "characters"). ... I understand. I used the two kinds of macros (character macros in (1) and...
... This would be a perfect example of my type (2), symbol macros. ... Ingenious. There might be a problem with what could be meant by the resultant [..]...
... Unless one allows error values. In some theoretical treatments in addition to the normal values one has ERROR, and then everything has a value, even [peter...
... From: <phimvt@...> To: <concatenative@yahoogroups.com> Sent: Monday, July 18, 2005 4:07 AM Subject: Re: [stack] stack operations vs...
stevan apter
sa@...
Jul 18, 2005 11:17 am
2524
... From: <phimvt@...> To: <concatenative@yahoogroups.com> Sent: Monday, July 18, 2005 3:45 AM Subject: Re: [stack] stack operations vs...
stevan apter
sa@...
Jul 18, 2005 11:30 am
2525
i believe that my latest version of XY 2.0 arrives at the best possible solution for combining (i) concatenative syntax, (ii) lazy quotation, and (iii)...
stevan apter
sa@...
Jul 18, 2005 9:40 pm
2526
... If taken at face value, and "[" is just another literal, then the evaluator would need to check the stack at every evaluation to see if "[" is on it, and...
... I'm not a big fan of lazy quotation anyhow. Why not just have lazy semantics, and let the quotations worry about themselves? That way your entire language...
