[Mark E. Shoulson: la <letteral>]

["Mark E. Shoulson" <cbmvax!uunet!cuvmb.cc.columbia.edu!shoulson>]

>Date:         Mon, 20 Jan 1992 19:29:42 TZONE
>Comments:     Warning -- original Sender: tag was
>        cortesi@CRICKHOLLOW.INFORMIX.COM
>From: David Cortesi <cortesi%INFORMIX.COM@cuvmb.cc.columbia.edu>
>X-To:         lojban mailing list <lojban@cuvmb.cc.columbia.edu>

>[me]   So at least: la ga'e ty moi == "that which I call T'th"

>[mark]
>> Yes, that'd work, but think about it.  It really doesn't seem to be what
>> you want.  Would you go around calling the number one "that which I call
>> the first"?  "moi" is probably the wrong word, you'd do better with "mei",
>> and eve that stinks.

>Not at all: moi is for making a label out of an ordinal.

No, "moi" is for making an ordinal selbri out of a letteral-string.  "mei"
is for making a cardinal selbri out of a letteral string.  "pamoi"==x1 is
first... (other places left up to some nice prescriptivist).  "pamei"==x1
is a singleton/one/solitary/whatever....

>This is what I've been trying to achieve: a label as a sumti.  It does
>sound strange when you use a number ("that which I call 4-th" -- well,
>is it 4th or isn't it? perhaps it *was* fourth in an order that has
>since been disarranged).  But if you are going to label variables with
>letter (strings), I think this is how you'd have to talk about those
>variables as variables.

You are trying to get a letteral as a sumti.  The cmavo for that is "li".
That's what it does for a living.  It *should* sound odd to require that I
use "that which I call fourth" for 4, just as it should sound odd to
require "that which I call nth" for n.  If you're going to require "moi"
for variables, how will you distinguish the nth [element] ("da poi nymoi"
or something) from n (li ny)?

>Don't know about .e; Bob? But if LI is the evaluator function (lambda?)
>which returns the value stored under a label, or applies a function to
>get the value it returns, then it is the complement to the LA exercise.
>LA names the variable, LI names its contents.  Both are required, yes?

.e is grammatical; check the grammar.  It's okay.  I'm unsure about the
semantics though.

The referent/symbol dichotomy is handled not with LA and LI, but with
"lu'e" and "la'e".  Thus, "lu'e li ty" would be the symbol refering to the
variable T, and "la'e li ty" would be its value.  This leaves the unadorned
"li ty" kind of ill-defined, so I *hope* somebody like John (poke, poke)
will step in and clear this mess up.  Personally, I feel that "li ty"
refers to the value of T, just as "li ci" is the value of 3 and so on.
"la'e li ty" seems redundant.  It may be undefined, it may be a pedantic
way of specifying "li ty" for people who want to be absolutely clear, or in
case there are more levels of abstraction involved in a particular case, or
something else.  "lu'e li ty" seems to me to be the variable name (not the
letteral "ty"/"T"; if I wanted to refer to that as a symbol I'd use "lu'e
zo ty" or something).

I seem to be muddying things more and more the more I speak.  Basically, if
you're treating "ty" as a value, then treat it like other values like "ci".
Would someone who actually designed the MEX stuff (and presumably had
semantics like these in mind) please let us know what the intent of the
language design is?

~mark
      o o     o   o             o o   o     o        o   o     o o
              o                       o o   o      o       o     o
  o   o     o     o         o     o     o   o      o     o     o     N2KOT
Mark E. Shoulson:  shoulson@ctr.columbia.edu