Re: Typesetting Lojban [was: Lojban word processor for Windows?]

[Ivan A Derzhanski <iad@xxxx.xxx.xxx]

Yes, I am on the list, and here's what I have to say:

(1) Like Mark, I prefer monospace to proportional for Lojban.
I don't know whether Lojban is hot or cold, poetic or prosaic,
fluid or solid, nor whether it needs to be any of those things.
Its first and strongest claim is to being a *logical* language,
and in this it can't afford to fail.

Forget about typewritten English; think on a larger scale.
The Chinese script is always monospace, and yet looks gorgeous,
doesn't it? More to the point, it is a logical thing: 1 unit
of width = 1 syllable = 1 morpheme. (With very few exceptions.)

Do you like the fact that all Lojban gismu are of equal length,
measured in phonemes/letters (5 in all, 3 consonants and 2 vowels)?
Well, I want the next logical thing: I also want them to be of
equal width on paper. That doesn't have to mean monospace as in
all characters being of the same width, but I'd like every consonant
to be as wide as every other, and every vowel as wide as every other
(perhaps less wide than the consonants). It makes sense for {' , .}
to be much narrower than the letters.

(2) _cm_ does occur in English, but is very rare; try a search
for `*cm*' in <http://www.m-w.com/cgi-bin/mweb>. I don't mind
ligatures in general, but I don't think I'd want any in Lojban
-- for the same Vulcanish reason as above: the distinction
that would be introduced between ligatured and unligatured
sequences would be illogical and artificial, and by virtue
of that fact most undesirable in the context of Lojban.

(3) Lojban in tengwar should work, although, in fairness, there are
natlangs whose consonant structures are more regular, and which are
better candidates for being written in tengwar (I'd nominate Nivkh,
or any of a number of Australian languages such as Aranta).

But no primary-order tengwa for {'}, please! Writing it as if
it were a voiceless velar fricative would make it more similar
to {k} or {x} that many other consonants would be. If it must
be a tengwa, let it be a very distinctive one such as yanta.
(This is why I've always opposed {h} for {'} in Roman script
-- it should be very prominent, but it is a Good Thing that it
doesn't look like a consonant.)

And if you want to take advantage of the logicality of the tengwar,
you shouldn't dream of writing {x} as if it were a voiced consonant.

-- 
"mu' Dajatlhpa', reH DajatlhlaH, <soxan tA nagoftI, tawAnI-^s goft,
'ach Dajatlhpu'DI', DughatlhlaH" walI gofteH rA bAz natwAn nehoft>
(Sheikh Muslihuddin Abu Muhammad Abdullah Saadi Shirazi)
Ivan A Derzhanski <http://www.math.bas.bg/~iad/>
H: cplx Iztok bl 91, 1113 Sofia, Bulgaria <iad@math.bas.bg>
W: Dept for Math Lx, Inst for Maths & CompSci, Bulg Acad of Sciences