[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
On Sun, Dec 15, 2002 at 02:27:58PM -0000, And Rosta wrote:
> Jordan:
> > On Sat, Dec 14, 2002 at 11:07:16PM -0000, And Rosta wrote:
[...]
> > > The convenience of ke'a is exactly when you want to use one variable
> > > more than once. It saves having to use ce'u subscripting (also an
> > > unofficial convention), or nei/no'a subscripting (also unofficial),
> > > which also won't even work if neither instance of the variable is
> > > an argument of the selbri. The method that definitely will work is
> > > ko'a-assignment with goi, but that is cumbersome. So, ke'a is much
> > > more convenient than the alternatives
> >
> > Ok; you ignored the main point though: people almost *never* need
> > to repeat these things in practice
>
> None of us have very much basis for judging this, seeing as we have
> all of us written and read so little, but you have had more
> experience than me. All the same, in what I have read and written
> I have noticed that other people are generally happy to use anaphora
> that relies on glorking, while I am not and consequently find the
> problem to be a severe one. (It's not that I think glorking is bad,
> but the very thing that attracts me to a logical language is the
> extent to which it can feasibly reduce reliance on glorking.)
Ok I can just tell you, that I see ka stuff a lot, and probably 1
in 100 (or less) have more than one ce'u in it. You can take my
word for this (I have about a conversation a day in lojban on irc)
or not.
Anyway; I don't find "goi ko'a" particularly difficult. I would
use ce'uxipa though.
> > WRT to subscripting; those are obvious and (afaik) uncontroversial
> > conventions which may very well be standard in a while
>
> Is your subscripting scheme on the wiki? I remember us discussing
> subscripting schemes (for nei/no'a) on one of the lists a while
> back and not reaching agreement on the details of the scheme or
> on what counts as a bridi (though iirc it was pc who disagreed
> with others about what counts as a bridi, so maybe that disagreement
> can be considered defunct).
I was speaking of ce'u. Subscripting on nei is probably useless,
since you can just script no'a (which *is* book specified I think).
> > > That said, if there are poi/poi'i embedded within poi/poi'i, ko'a
> > > assignment remains necessary, which is what led me to propose a
> > > range of experimentals to abbreviate KOhA-assignment
> >
> > Most of those abbreviation things are a bit of a joke though. You
> > have things like "goi'a" IIRC, which saves *one* syllable, and needs
> > preprocessor changes to work properly.
>
> If any require preprocessor changes then that was inadvertent error
> on my part, as I have striven to distinguish between what is and
> isn't baseline-compliant. (Preprocessor changes obviously aren't
> baseline-compliant.) I will check "goi'a" when I go online; my
> recollection was that I had proposed such a thing but moved it to
> the page for obsolete proposals, but I may have intended to move
> it but forgotten to actually do it.
It would've required processor changes to do it right. Putting it
into KU is a hack.
> Because the scope for introducing glork-free anaphora is so limited,
> because of tight constraints on innovation, it is inevitable that
> experimental cmavo can ameliorate the problem to only a slight
> degree.
I'm not sure what the problem is?
> Regarding the saving of a single syllable, there is a sense in
> which the saving of words is desirable, regardless of whether it
> saves syllables. There are two reasons for this. The first is that
> since syllable-count was ignored in the design, the best way to
> do justice to the design, when exploring it through usage, is to
> ignore syllable-count. The second reason is that one can measure
> length not only in syllables but also in words (which are the units
> of input that the parser operates on).
I highly disagree. Unless there are stops between the words for
some reason it doesn't count. I see no benifit to being shorter
in terms of words (esp. since a lot of words can get compounded
together also ("goiko'a", "lenu"). And having a relatively hackish
shorting cmavo just to save *one* syllable is pretty laughable.
> > Furthermore, the entire idea
> > of proposing them now before there's any usage (particularly usage
> > by yourself) suggesting them warranted goes against what you were
> > saying about your views on shortings on the main list
>
> If we had as a community agreed on a programme of usage that would
> explore the design and deliberately ignore syllable count and, to
> a certain extent, word count (e.g. allowing us to compound cmavo
> that we wish were a single word, and letting the compound count as
> a single word by stylistic criteria), then I would not be proposing
> these abbreviatory experimentals. The one thing is what I think
> the community should do, and the other thing is what I do given
> what the community thinks it should do.
[...]
I don't think any such thing is needed. If shortings are to be
properly done, they should be done after there's *at least* 40-50
fluent speakers and it should be done by frequency counts of text
with the intent being to convert the most common 5-6 syllable strings
into 1-2 syllables (and this only if the most common 5-6 syl. string
is common enough for this to work). Doing it by what we *think*
is going to be useful is a bad idea.
--
Jordan DeLong - fracture@hidden.email
lu zo'o loi censa bakni cu terzba le zaltapla poi xagrai li'u
sei la mark. tuen. cusku
Attachment:
binmq6HQsJyuz.bin
Description: application/ygp-stripped