Re: LISP code example matching an abstract construct

Tayssir John Gabbour, Pascal Bourguignon, and Kaz Kylheku many thanks
by this useful clarification. Now i can understand my headaches at this
list.

Basically the problem was we were talking "different" languages; my
apologies! For me something as

(list entries)
(entries list)

has really no sense and, therefore, i was very confused about this kind
of code and part of the criticism i received in the list. Do not forget
that I become from a different community!

I also see that some LISPers would find above code incorrect and even
unelegant, therefore my ideas are not so strange.

<blockquote>
In [CL], deciding whether to use the function or the value is
impossible unless the context is known. These two forms result in
different interpretations of an expression, x:

(x ...)
(... x ...)

To some programmers, a basic 'rule' of Lisp style is that code is
clearest when the least amount of context is necessary to determine the
meaning of an expression. Unfortunately, that rule is violated in [CL].
</blockquote>

Once i understood this (and see that some LISPers think similar) i can
see that probably my original proposal for CanonCode (transformation of
CanonML documents somewhat as XSLT is for transformation of XML) could
work in a minimalist approach.

(X ...)

and

(... X ...)

both mean the same X command, whereas in CL only (X ...) can be
interpreted that way. Therefore, i can write

[[partial rho over partial rho] = L rho]

just as i had thought in a principle and datument could be translated
to TeX for printing and to LISP dialects or some other PL (e.g.
Mathematica) for simbolic evaluation.

I will update a new entry in Canonical Science Today with this
information and my last thoughts about programming module. Many thanks!


Juan R.

Center for CANONICAL |SCIENCE)

.