[Patents] Counterexamples to John, Hartmut, Edward

[PILCH Hartmut phm@a2e.de]

> A more interesting example, between the infamous one-click and the

Actually the One-Click algorithm is less trivial than many softpat
algorithms, and the claimed scope is narrower. One-Click happens to be an
easy target of derision, but it is hardly one of those patents that stifle
innovation.

> intractable natural language problem, is given by MP3.  It was claimed
> that only a team of dozens of mathematicians working for months or
> years and financed by expected patent royalties could possibly have
> devised an audio compression algorithm of such high quality.

Maybe dozens of psycho-acoustics specialists had done some base work for a
few decades before.  The algorithms found later were hardly the difficult
part.

These algorithms are at least based on new empirical knowledge, but other
factors should still disqualify them from patentability in Europe:
(1) the empirical knowledge was not new at the date of patenting
(2) the problem to be solved was not a physical problem but a problem
    of information handling (data compression).

> I suspected this was false when I first saw it, and now it looks as if
> Ogg Vorbis is disproving it.

Who is Ogg Vorbis?

-phm