[lit-ideas] Re: Correction [2]

  • From: Robert Paul <robert.paul@xxxxxxxx>
  • To: lit-ideas@xxxxxxxxxxxxx
  • Date: Mon, 16 Jan 2006 21:51:33 -0800

About problems about numbers, Peter Junger wrote;

These are issues of extreme practical---or, at least, legal---importance since they lie, as far as I can tell, at the heart of the questions of whether computer programs can be patented or copyrighted or both or neither. General purpose binary computers---and Turing machines---distinguish between one of two marks on a magnetic tape or an optical disk or wherever, which marks we can---and often do---interpret as numerals representing the numbers 0 and 1.

I wonder if this is really what computers do? Yes, they may be said to distinguish between marks on a tape, etc., but what a computer does
(I thought) is to do something (or not) as it 'reads' a pattern of marks; that the marks parse to numbers in a binary code is irrelevant to the computer, which just trots along on/off/etc. That the program is to the programmer in numbers matters not to the machine. But this may be a side issue or worse yet wrong and a side issue.


The question of software patents turns on whether algorithms for processing one ordered set of binary digits into another set of binary
digits (the two sets perhaps being the same) can be patented. The
Supreme Court has said that they can't be because they are ideas or
mental processes; the lower courts have, however, of late been ignoring the Supreme Court.

The copyright question turns on whether a numerical encoding---usually
as binary digits---of a program implementing such an algorithm is
an idea, and thus not copyrightable, or the expression of an idea,
in which case it is.

Are algorithms copyrightable? The algorithm itself is no more a 'mental process' than the recipe for Coca-Cola is a mental process, and if the latter is copyrightable (patentable?) I don't see why an algorithm couldn't be or shouldn't be, questions about the free exchange of information aside. I'm surprised to learn that the Supreme Court finds any algorithm whatsoever uncopyrightable. I mean: how can Microsoft in light of that have a claim on Word, or any other application?


I think that there is a lot of ontological confusion lurking in these
issues. Is a numeral the same sort of thing as a number? Neither
is, after all, tangible. Are numbers ideas? Are algorithms ideas?---after all, every algorithm can be assigned a Goedel number.

Numerals are not numbers. 5 and V refer to the same number; they are not the same numeral. (A proof.) My hunch is that programmers needn't have a definition of number to work with numbers and that they could do what they do even if it turned out that numbers were Fregean objects or the notes to the music of the spheres. I wish that Alito had been asked what he thought numbers were. He might have said, 'My earlier view was that Plato was right,' and refused, as he did with so many other of his replies of that general form, to say what his current view was. Perhaps we each mean something different by 'tangible.' A Roman date carved in stone is tangible, and its elements are numerals. So with marks on a chalk board.


Lawyers---even academic lawyers---are not supposed to have to worry
about such matters and I am convinced that we don't want judges to
worry about them. (Imagine Judge Alito being asked by the Senate, What is a number?")

So I would greatly appreciate any help that you can give me.

I can give you none, obviously, because if I were actually to help anyone I would be admitting that philosophy had some practical value, an admission that would violate any number of principles I've sworn to uphold, e.g. Morgenthau's extension of the Auden Rule, not to mention the bylaws of the APA.


Robert Paul
The Reed Institute

------------------------------------------------------------------
To change your Lit-Ideas settings (subscribe/unsub, vacation on/off,
digest on/off), visit www.andreas.com/faq-lit-ideas.html

Other related posts: