Message-Id: <m0t0BdR-0000ZUC@xiron.pc.helsinki.fi>
Date:         Tue, 3 Oct 1995 18:53:41 +0100
Reply-To:     ucleaar <ucleaar@UCL.AC.UK>
Sender:       Lojban list <LOJBAN@CUVMB.BITNET>
From:         ucleaar <ucleaar@UCL.AC.UK>
Subject:      h for apostrophe
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 2654
Lines: 52

> >The motivation is partly aesthetic, and partly practical.  The {'} is
> >typographically ugly when preponderant.  Most software doesn't recognize
> >it as a within-word character.  Omitting it altogether makes the text
> >briefer and more pleasing to the eye.  I could go into more details, but
> >won't bother.
> I understand all of these except one.  Most software used in processing
> language had better accept the apostrophe as being an acceptable
> character in a word, since it is fairly common in English (probably more
> common than 'z', 'q', or 'x').  You even used it in "won't" in the
> paragraph above.

Sorry, yes omost softwareo is exaggerated. What I should have said is
most software doesnAt recognize it as an alphabtic charcater. You canAt
use it in unix or some dos filenames, and when I call up my jbovlaste
using Write (which comes in Windows) & search for whole words it treats
a as a word-delimiter.

> As far as I know, it is only in programming languages that apostrophe is
> a special character.  We use 'h' instead of apostrophe in the YACC
> grammar, therefore, since YACC takes a text file that meets C language
> conventions.

So not a great step to making it officially sanctioned (in the sense of
allowed, not in the sense of forbidden-on-pain-of-being-ignored-by-the-
parser).

> A lot of the rest is just aesthetics.  >I< happen to like having a
> language that looks unlike English - it helps remind me to avoid
> malglico usages.  But that wasn't much a factor when we made the
> decision - the goal was to have a consonantal sound that was not a
> consonant for morphology purposes, and making it look different from a
> regular alphabetic character seemed like an excellent way to go about
> it. But we've been over this many times. %^)

Yes qquite, we donAt need to rehash this. IAm not saying the apostrophe
is stupid or Really Bad. Everything in lojban design has a fairly
respectable rationale, and even its worst features are defensible.

> I don't like having to exclude And's texts - I just processed 3 years
> worth of archives into a new set of lujvo for processing into the
> dictionary.  But all my means of recognizing Lojban words pretty much
> require that I shrink down the words to process by eliminating those
> with non-Lojban letters like 'h', and to expect certain patterns of
> consonant and vowel.  In only a couple of cases did I see a word that
> was clearly an And Rosta Lojban word and manually "saved it" from
> oblivion by putting the apostrophe back in.

Thanks for the dedication. I am not a great zbasu be lujvo, so the
dictionary will not be the worse for their absence.

---
And