[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] Re: Is there any demand for LoCCan3?
la gleki, On 11/07/2012 08:41:
Let's name
Loglan = LoCCan 1.0
Lojban = LoCCan 2.0
Lojban after xorlo reform = LoCCan 2.1
As I understand it, Lojban was more of a fork from Loglan than an improvement of it -- essentially a Loglan relex with some other subsequent mostly gratuitous changes.
Nevertheless, your terminology is convenient.
1. Backward-incompatible And Rosta's project of CCV-gismu equal to
rafsi and CCVrCCV lujvo where "r" is a buffer consonant.
[I had in mind this "r" being any consonant, not just /r/.]
This lowers
signal-to-noise ratio but makes learning rafsi=gismu much easier (no
separate forms for gismu/rafsi). This can possibly remove the need in
many modal tags that are actually duplicates and compressed versions
of gismu
I have no such project. The CCV thing was just an idea tossed out, as one of the many easy ways of improving on Lojban.
If I was asked for advice on LoCCan3, it'd be this, in the first instance:
Discard the LoCCan2 syntax, then start with predicate logic, and enrich it only with what predicate logic can't express or accommodate. Insist on the syntax being incrementally parsable with no lookahead, where parsability means recovering the logical form, not merely assigning some meaningless structure to surface phonology, as in the current so-called parser.
Apply some sort of Huffman-encoding type scheme to at least morphemes (possibly to morpheme strings). Possibly temper this by the desirability of having some sort of phonosemantic patterning among morphemes, for mnemonic purposes.
That's not be the end of my advice, but it'd be the key bits.
HOWEVER, I think that a LoCCan3 that did that would still not be good enough, and I don't see much point in investing a lot of effort into creating a language whose design is miles better than Lojban's but is still not fit for purpose. (I suppose that if future research were to show that the fundamental loglang problem is insoluble, then one would after all want the best possible LoCCan version, as the language least unfit for purpose.) To apprehend the fundamental loglang problem, imagine Lojban without {zo'e}, and then try to think how it could be redesigned to avoid being impossibly longwinded and impossibly taxing on the memory of (especially) the hearer/reader.
And Rosta's suggests relexing gismu. Still lujvo in eir language will
have no internal predicate structure.
Quite so. Since simplex brivla (gismu) would already have maximally short forms, complex brivla (lujvo) would be necessary only when the forms were not semantically compositional, i.e. when the meaning of the whole is not completely predictable from the meaning of the parts. The morphological translucence (semicompositionality) of lujvo would serve a merely mnemonic purpose.
Rafybri compress some bridi preserving sumti places between two or
more gismu. They are equivalent in meaning to some lujvo but donot
require learning them by rot. However, ordinary lujvo retain broken
vague ambiguous internal structure of tanru with no places. I don't
think And Rosta's CCV suggestion is much better than rafybri (rafybri
can be used immediately as they add new rules but donot destroy any
old rules). But let's wait until ey presents something.
For the reasons given above, rafybri would be redundant in a well-designed loglan. Or, to put it a different way, a well designed loglan would use something akin to rafybri as its basic device.
--And.
--
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@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.