Re: [engelang] Xorban lexical design considerations - engelang messages

While I think the lexicon is far and away the least interesting part of this project and a genuine obstacle to the interesting parts, I throw in my penny's worth in hopes it'll move out of the way.

Yes. The Loglan/Lojban rationale is suspect and untested and costly (and mostly lost now with the drpped vowels); a more philosophic approach would be as likely to achieve the desired goal of easy learnability and be easier to expand or contract. I would suggest (for syntactic purity) that the number of places for a predicate be encoded.
Sent from my iPad

On Oct 13, 2012, at 7:33 PM, "Mike S." <maikxlx@gmail.com> wrote:

This thread is for the long-term goal of completing Xorban's lexicon. Place anything related to the lexicon here.

I have been sifting through gismu somewhat randomly to start with, and an idea occurred to me. The basic idea is that stems could generally get relexed such that "CC" chunks give a mnemonic clue as to the stems' meaning, and the last C or CC places the stem in a semantic class. Without hashing out all the details, I'd like to hold a straw poll:

[Yes] I would like the lexicon to generally break down schematically into CC mnemonic clues (in other words, let's reorganize & overhaul the lexicon).

[No] I would like the lexicon to be random/naturalistic like the current gismu stems are (in other words, let's just remove the homonyms and add new stems as needed, but otherwise import the gismu as they are and as we have been using them).

I am looking for your initial reaction and thoughts here, nothing else.

-- ECG
co ma'a mke

Xorban blog: Xorban.wordpress.com
My LL blog: Loglang.wordpress.com