Received-SPF: pass (google.com: domain of nictytan@gmail.com designates 209.85.212.52 as permitted sender) client-ip=209.85.212.52;
Subject: [lojban] {moi}3, Predicates, and Conventions
From: tsani nicte <nictytan@gmail.com>
Message-Id: <BEA005D0-59D4-474B-8EC9-14128DF1B538@gmail.com>
Date: Mon, 28 Jan 2013 17:49:44 -0500
To: "lojban@googlegroups.com" <lojban@googlegroups.com>
Mime-Version: 1.0 (iPod Mail 8J2)
Reply-To: lojban@googlegroups.com
Precedence: list
Mailing-list: list lojban@googlegroups.com; contact lojban+owners@googlegroups.com
Sender: lojban@googlegroups.com
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
X-Spam_score: -0.1
X-Spam_score_int: 0
X-Spam_bar: /

A discussion latro'a and I had in IRC the other day led me to want a brivla=
 version of {mei} and by extension, every MOI. I already devised their lojb=
an definitions:
.i ka ce'u me ce'u MOI ...
where "..." represents any trailing places of the MOI.

My proposal is making zi'evla for each. (A lujvo for this purpose is imposs=
ible by my standards.) I just need suggestions for what to use as a zi'evla=
 head.

As for moi3, I'm a bit puzzled as to how we can construct a predicate to or=
der/associate the keys with the values. In truly associative cases, as in {=
lo me mi moi} we can use a simple binary predicate: {lo ka ce'u broda ce'u}=
.=20
e.g. {.i (ko'a goi lo'i toldi fo'a goi lo ka ce'u se melbi ce'u zo'u) lo me=
 mi moi be ko'a bei fo'a cu barda zmadu lo me do moi be ko'a bei fo'a}
(I have put the prenex in parentheses to enhance readability. Also, I have =
decided semi-arbitrarily that the order of arguments for the association pr=
edicate, i.e. moi3, be F(key,value), but I suspect that F(value,key) be mor=
e useful. (It might require less SE conversions. Otherwise the classic tric=
k of the lambda prenex with ce'ai is acceptable.) Anyway, I'm not sure of t=
he ideal order.)

(It occurs to me that some might not know what this type of associative con=
struct is actually useful for. Here's an example: {.i .abu cusku lu lo mi z=
dani cu barda li'u .ije .abu skuspu fi lu .i ua .i si'anai lo me mi moi cu =
cmalu}. Implicitly, the set in this case is {lo'i dinju} that by xorlo magi=
c contains only the relevant elements and the association predicate is {lo =
ka ce'u se zdani ce'u}.)

As for numeric moi, things get trickier, because the predicate needs to ord=
er the elements (which means they must be somehow orderable.) For the purpo=
se of exemplification, let's take {lo'i prenu} as our moi2, and try to crea=
te a predicate that numerically orders them according to the time at which =
they arrive at my house. (This is the part I'm not sure of:) {lo ka ce'u zm=
adu ce'u lo ni ca makau ce'u klamu'o lo mi zdani}. It looks like each argum=
ent of the predicate refers to elements of the set. {moi} is "smart" enough=
 to order the elements of the set based on which elements are superior to w=
hich elements. This is reminiscent of simxu2, isn't it?

My problem with this is that behaviour of moi and moi3 seems to depend on w=
hether the moi is numeric or associative, unless I or someone else can cons=
truct a decent type of moi3 that is universal for numeric and associative m=
oi. I've thought about maybe using va'e in moi3-numeric to directly associa=
te the numeric indices with the values, but I'm not sure how I'll go about =
getting a reference to the appropriate value inside moi3.

Thoughts, comments, hate mail, (spam?) and flame wars are appreciated as us=
ual.

.i mi'e la tsani mu'o

--=20
You received this message because you are subscribed to the Google Groups "=
lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegrou=
ps.com.
Visit this group at http://groups.google.com/group/lojban?hl=3Den.
For more options, visit https://groups.google.com/groups/opt_out.