Re: [lojban] tosmabru test

[Jorge Llambías <jjllambias@gmail.com>]

On Wed, Sep 7, 2011 at 7:17 AM, Bob LeChevalier, President and Founder
- LLG <lojbab@lojban.org> wrote:
>
> fu'ivla are at the bottom of the pecking order among Lojban word-forms, and
> ANYTHING about a possible fu'ivla that could lead to ambiguous resolution as
> something else makes it invalid.  There are a number of possible ways a word
> could fall apart, and we made no attempt to concoct a specific test for all
> of them, and indeed decided at the time that the question of defining all
> necessary tests algorithmically was too hard to be practical (at the time),

The test to determine whether a given string is a valid fu'ivla or not
is not that complicated:

(1) check that it doesn't violate phonotactics (i.e. it doesn't
contain any impermissible clusters).

(2) check that its only stressed syllable is the penultimate one, that
it ends in a vowel and that it does not contain any "y".

(3) check that it doesn't consist of a string of rafsi, and that
adding CV in front of it does not turn it into a string of rafsi.

(4) check that if you remove one or more cmavo forms from the left,
you don't end up with nothing, a gismu, a lujvo or something that
passes (1), (2) and (3).

Any string that passes those four tests is a valid fu'ivla.

The second part of (3) is called the slinku'i test, and (4) could very
well be called the tosmabru test for fu'ivla, even though we don't
call it that.


>  There is no name for the generalized algorithm (which would break down any
> text string marked for stress uniquely.

I call it "the PEG Morphology":
http://www.lojban.org/tiki/BPFK+Section%3A+PEG+Morphology+Algorithm

mu'o mi'e xorxes

-- 
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.