Re: [lojban] Re: Are Natlang the best case for entropy in communication ?

[Escape Landsome <escaaape@gmail.com>]

What is wanted from an engineer is not only to measure things or make
correct proofs of theorem but also to have a right sense of magnitudes
of values.

That is, when I studied physics (and not philosophy, as you would have
guessed), my teachers asked me to prove that no human would experiment
gravity in a hollow Earth, or that some satellite would fall with the
speed V, but *also* they saked tricky questions, such as : "what is
your volume ?", and you had to answer "roughly 70 liters", since you
reason qualitatively thus :

"my body is mainly water (more than 80%) and 1liter of water is 1kg,
and I weigh 70kg, so my countenance is 70 liter"

This is qualitative intuitive estimation, and is accepted as good in
science and engineering !

-

I deeply regret that you have no INTUITION of what linguistic entropy
is, for if that were the case, you would see as SOMETHING OBVIOUS that
S(lojbanic example) far exceeds S(example in natlang)

One acknowledged interpretation of linguistic entropy is that when you
have approximately n responses that could do the guess, and all k
others that are too rare or too strange, as the computation shows, the
k contributions are negligible, and roughly the mean value will be the
same as a choice between n choices.

Since then, you see by immediate intuitive estimation, analogous to
the ones you would do in engineering, that mean value of S must lie
between log_2(1) an log_2(3), for the most often cases have between 1
and 3 possible alternatives, and more than 3 is rare in a paradigm.

Thus, mean value is between 0 and 1.58.

This is rough estimation but this is science.

-

But you insist on having (unnecessary) precise study, this is rather
boring, since you are in the same posture that somebody would would
not accept that the distance from Paris to the Moon is far greater
than from Paris to Moscow, since he hasn't a PRECISE value.

You insist on having precise values instead of estimation of
magnitudes, when the latter suffices vastly.   This is a way to make
yourself boring, and to lock to debate with the kind of trick somebody
could use in a debate where he doesn't accept the evident truth.

-- 
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.