> I started working on a near-isolating engelang last year; here is my first
> posting about it:
>
>
http://archives.conlang.info/ghau/goerthin/wuanphukhuen.html
>
> It hasn't progressed very far, unfortunately -- I have a lexicon with
> about 200 root words and a corpus of about 50 or 60 sample
> sentences, but haven't worked on it since last summer. I started
> writing up its grammar but haven't posted that anywhere yet.
>
"I'm probably goingto use frequency analyses on my Toki Pona corpus and my
gzbcorpus, an already available list of the most common morphemesin Esperanto,
and the list of "semantic primes" from the Wikipediaarticle on "Natural
semantic metalanguage" as sources for ideas,along with the first few sections of Rick
Harrison's ULD."
Have you considered using the Swadesh lists?
stevo