🚀 go-pugleaf

In article <1991Sep15.052225.9619@franklin.com> jesse@franklin.com (Jesse Perry) writes:
> That's because the LZW patent is *NOT* a patent on the
> compress(1) program; it is instead a patent on a method *used by*
> compress(1), and thus is a patent which covers compress(1), as
> well as many other possible programs.  This is why it is valuable.

This is ridiculous. Yes, we are in violent agreement that a ``patent on
the compress(1) program'' would be idiotic if it weren't phrased like
the LZW patent. That's why, when normal people talk about patenting a
program (like compress), they mean patenting any or all of the methods
used in the program (like LZW). (This is exactly the wording used in,
e.g., Anthony and Colwell's article on computer patents. As they say,
you can copyright a program, you can keep it secret, or you can patent
it.) Why you and Zippy persist in talking about a different meaning is
beyond me.

Zippy's original statement, namely that you can't (in practice) patent
an entire program, is pointless at best and misleading at worst.
Starting from a program (my yabba compressor, for instance), I could
apply for a patent covering Y coding, the file formats, huptries, and
everything else involved in that program. Zippy says such a patent is
useless because he can just avoid *one* of the patent methods (the file
formats, say) and he'll be safe. But that's wrong. If he uses *any* of
the methods used by the original program, he'll infringe the patent.

---Dan