To whom do we sell

[cbmvax!uunet!I2Wash.COM!dlb (David Barton)]

Lojbab writes, in a recent message:

>If the language is going to ever have practical application (i.e. if
>it is ever to be a real langauge), we have to sell the language to one
>of the three 1) an international language community 2) the artificial
>intelligence mnatural language processing community 3) the linguistic
>communtiy.

This seems as good a time as any to agitate for another community to
whom we should be selling the glories of Lojban.  I think we should be
selling Lojban as a professional engineering language, in which to
design and verify complex systems.

The IEEE, the government, and some in industry are engaaged in the
difficult task of setting up a graded series of languages in which to
design complex systems.  This includes languages oriented towards:
software (Ada and C), digital systems (VHDL), RF and microwave systems
(MHDL), and a systems design language.

I submit (and am starting to push in this community) that the
legitimate "top" of this pyramid is a spoken logical language, i.e.
Lojban.  A truly nasty step in top level design, which causes major
heartburn and is an impediment to formal systems design, is the
translation of specifications and requirements into formal notations.
I suggest that if design discussion among the engineering team takes
place in Lojban, this barrier is much smaller and may even be null.
Any place where ambiguity exists is a place where design effort must
be focused to eliminate a lack of information, and is not a result of
the language (English) being used for disucssion.

The nice thing about selling to this community is that Lojban (or at
least logical languages in general) has unique benefits.  Esperanto
and other languages may be artificial and more regular, but are not
constructed in such a manner at to assist the transition into formal
systems specifications.

The NLP connections only enhance this connection.  Picture, for a
moment, a system that monitors a spoken discussion concerning a
complex system, translates this into written Lojban, and derives from
this written description a formal specification of the system that can
be decomposed according to existing design techniques.  Imagine
interacting with a design environment in a spoken language, to the
level where semi-automated synthesis converts the description into a
true system design (consisting of executable notations, as listed
above).  Now imagine that this system is formally verifiable,
according to a formal semantics of each notation.

With Lojban, I think that this is achievable.  I think we should be
selling this to the appropriate people in government and industry.

For that matter, I am trying to sell it to you.

Any comments?

						Dave Barton
						barton@i2wash.com