On Mon, Oct 22, 2012 at 3:20 PM, And Rosta
<and.rosta@hidden.email> wrote:
Mike S., On 18/10/2012 02:23:
Okay, prototyping is fine. But things like your initial h- convention are pointless in the light of that. Everything should have an initial h- by that convention.
I realize that, which is why I started using h- for novel word classes and operators hVk which might take up whole C, and using n- and j- for other things.
It would also help to use conventions where stems and particles are indicated by glosses in square brackets rather than by their temporary phonological forms. That would make exx easier to understand, and wouldn't let us be distracted by phonological forms that will be scrapped and will spare us from inadvertently learning forms that we might later be reluctant to discard because we've already learnt them. In sum, use as far as possible a graphical nonphonological notation.
--And.
In order to evaluate this idea, it would be interesting to see what it'd look like if we used a mathematical symbolism augmented with English words for the object language. We'd probably have something like:
person = pr
cat = mlt
black = xkr
& = je
~ = na
x, y, z, u, v, ... = a, e, i, o, u, ...
me = a'a
sth = o'e
...etc.
Then we could write nonphonologically something like:
E-x cat-x black-x
L-x & black-x cat-x happy-x
Definition: Meanwhile[-u] F => L-v Event-v F concurrent-v,Sth[,u]
Nice, but not yet enough: are (incredibly) "meanwhile" and "concurrent" predicates? "sth" is presumably a quantifier phrase of some sort, and other pieces are just scattered about (part of that old problem of not distinguishing connectives from quantifiers from predicate makers from whatever else is floating about).
... versus our current:
sa mlta xkra
la je xkra mlta glka
Definition: ne'u[ke] F => lo fo F cbnoko'e[ke]
I could probably carry on in either format, but I think I'd prefer continuing with what we have, partly because, at the end of the first stage of development, it'd be nice to say that we have a speakable language called "Classical Xorban" (or maybe "Archaic Xorban"), even if we go on to design something better.
I have noticed that Jorge sometimes goes to the trouble of writing the Englishy FOL stuff interlinearly; another solution is to allow people to use both notations freely. It would be desirable to create formal production rules for Nonphonological Xorban.
I am willing to hear other opinions on "Nonphonological Xorban" a.k.a. Englishy FOL. Thoughts welcome.
--
co ma'a mke
Xorban blog:
Xorban.wordpress.com
My LL blog:
Loglang.wordpress.com