🚀 go-pugleaf

mainline@omni.voicenet.com (William Smargiassi) writes:
>>Couldn't happen. Mathematical formulas are not patentable.

warwick@cs.uq.oz.au (Warwick Allison) writes:
>So any "program" written in a purely functional language is not patentable?

Maybe not.  See below.

>Copyrightable?

The *expression* of the program (the organization as written on the page,
with comments, etc) would certainly be copyrightable.

Pure functional languages don't specify how a result is to be computed.
They only specify relationships between inputs and outputs.  They describe
what the result should be, not how to get that result.  In most cases,
there are many methods for computing the outputs of a functional program,
some much more efficient than others.  Since the program doesn't specify
a method, it would be hard to get a method patent for it.  Right?

Unfortunately, it's still possible to get software patents for various
aspects of the interpreter or compiler used to compute functional programs
efficiently.

Also unfortunately, though I could write a functional program (say, in a
stream-oriented functional language like Sisal) to describe the
relationship between an input stream and an LZW-compressed output stream,
while that might not violate the patent, as soon as I attach the input
stream to a data source and the output to a data sink, and the application
is communications or storage, I've then violated the patent.  So it
doesn't look like it works.

--
-- Joe Buck 	<jbuck@synopsys.com>	(not speaking for Synopsys, Inc)
Phone: +1 415 694 1729