Re: [edict-jmdict] SKIP license

[Jim Breen <jimbreen@*********>]

2010/1/20 Glenn Maynard <glenn@zewt.org>

> A thought: any copyrighted kanji indexing system must be severely
> flawed.

なるほど. I'd say it's meaningless.

> ... The purpose of an indexing system like SKIP is that anyone
> can take a kanji, follow a few mechanical steps, and come up with a
> specific number (2-3-5). With a good indexing system, you should
> always come up with the right number. For something to be
> copyrightable, it can't be a mechanical process where everyone comes
> up with the same results; it needs to be a creative process. SKIP's
> creation is probably a creative process, but that's its biggest flaw.
> Claiming copyright on an index is admitting to having a badly flawed
> system.

I've said all that to Jack. He should have patented it. We all know it
wouldn't hold up, but we are all friends so we go through the motions.
Similarly the quite historical four-corner codes can't really be copyrighted
either.

Martin Duerst generated the SKIPs for the JIS208 kanji not in Jack's book,
and I did them for the JIS212 kanji. If they were copyrightable, they'd be
ours; not Jack's. But as I say, we go along with it.

Cheers

Jim

--
Jim Breen
Adjunct Snr Research Fellow, Clayton School of IT, Monash University
Treasurer: Hawthorn Rowing Club, VCA Secondary School, Japanese Studies Centre
Graduate student: Language Technology Group, University of Melbourne