perm filename OWL.NOT[E77,JMC] blob
sn#291610 filedate 1977-06-28 generic text, type C, neo UTF8
COMMENT ā VALID 00002 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00002 00002 1. Since the human internal representation of facts is certainly not
C00005 ENDMK
Cā;
1. Since the human internal representation of facts is certainly not
English, as is evidenced by the fact that a person will often utter
an ambiguous English sentence without noticing that it admits different
interpretations, it is not clear that English is a suitable internal
representation of facts for a computer program. In fact it isn't.
English is a communication language designed to get around the fact
thatit is necessary to communicate names associated with network
nodes and not whole networks. It also has to deal with the fact
that different people have different internal information networks.
In getting around these difficulties, it relies heavily on the fact
that the listener is active and knows about the subject under
discussion and can conjecture the right fact using the communication
for suggestions.
2. If English is to be extended to incorporate new notions for the benefit
of physics, why not for the benefit of programming? Why not consider
LISP as merely a technical extension of English appropriate for the
expression of procedures just as formulas are appropriate for the
expression of the law of gravitation.
3. But what are meanings? What guarantee is there that everything
a person knows is expressable in English?
4. It is unreasonable that a world model, however skeletal, can be
a taxonomy of concepts.