[Bug localedata/11213] localedata licencing issues

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

[Bug localedata/11213] localedata licencing issues

maiku.fabian at gmail dot com
http://sourceware.org/bugzilla/show_bug.cgi?id=11213

--- Comment #18 from Thorsten Glaser <tg at mirbsd dot de> 2012-07-16 13:13:31 UTC ---
Keld, while you may have had a lot of “sweat of brow”, in the end, the locale
data (in the one file I looked at) is mere fact, and there is, basically, only
one way to express such facts using the format glibc expects. This is, at
least, under the EU interoperability directive, not copyrightable (neither are
*.h files, unless they include insane amounts of inline functions, by the way).
As for the USA, you probably know that better than I do, but the work, while it
may have cost you a lot of creativity, is mostly “sweat of brow”, to express
the fact that way. You may have designed the interface, the character naming
convention, etc. but these are interfaces, not works in the sense of copyright.
I really do not want you to lose any attribution of that, but I don’t believe
the (one file I looked at with) locale information as shown is not
copyrightable.

“something that could have been patentetd”[sic] is of different scope than
copyright relevant work. I also fear you could probably have patented it, but
not put it under copyright protection. (Jonathan, mere effort is also not
enough in Germany, although USA’s “sweat of brow” doctrine is a tad stronger.
Still, we’re facing interoperability interfaces here, and mere fact; the
currency of the USA is the Dollar, no matter what.)

The presentation and combination of the data may, may, be affected by copyright
or even database law, true. I guess SFLC and, if possible, international
lawyers could have a look at that.

I laud your attempt to keep quality, but IMHO, using legal matters for that is
not the way to go. You could run a “locale data repository”, from which e.g.
glibc could then pull, for that, like tzdata.

--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
Reply | Threaded
Open this post in threaded view
|

Re: [Bug localedata/11213] localedata licencing issues

Keld Simonsen-2
On Mon, Jul 16, 2012 at 01:13:31PM +0000, tg at mirbsd dot de wrote:
> http://sourceware.org/bugzilla/show_bug.cgi?id=11213
>
> --- Comment #18 from Thorsten Glaser <tg at mirbsd dot de> 2012-07-16 13:13:31 UTC ---
> Keld, while you may have had a lot of ???sweat of brow???, in the end, the locale
> data (in the one file I looked at) is mere fact, and there is, basically, only
> one way to express such facts using the format glibc expects.

Yes, some of the locales are stright forward. But if you look at the whole
combination of a locale, with the LC_COLLATE, the LC_CTYPE, and the transliteration
specs, there is not only sweat in it but a lot of decisions.
Even the LC_MESSAGES data for yes and no contains a number of variations to consider.
And even the LC_TIME has some tricks. The choices is a common subject we talk about on the
localedata reflector, because there are different ways of doing a lot of
things in the locale data. And when there is choice, there is creativity
and work height.

> This is, at
> least, under the EU interoperability directive, not copyrightable (neither are
> *.h files, unless they include insane amounts of inline functions, by the way).
> As for the USA, you probably know that better than I do, but the work, while it
> may have cost you a lot of creativity, is mostly ???sweat of brow???, to express
> the fact that way. You may have designed the interface, the character naming
> convention, etc. but these are interfaces, not works in the sense of copyright.
> I really do not want you to lose any attribution of that, but I don???t believe
> the (one file I looked at with) locale information as shown is not
> copyrightable.

(maybe too many negations here - but I think I know what you mean.)
I think there is differences of what level of work height that is needed for it to be
copyrightable in different countries. but I would be astonished if say 1000 lines of
data with a number of serious design decisions on the solutions did not
enjoy copyright in any country (under the Berne convention)..


I do think locales per se are works. They are freestanding items, that are supposed then to
work toghether with other parts to make a functioning system.

> ???something that could have been patentetd???[sic] is of different scope than
> copyright relevant work. I also fear you could probably have patented it, but
> not put it under copyright protection. (Jonathan, mere effort is also not
> enough in Germany, although USA???s ???sweat of brow??? doctrine is a tad stronger.
> Still, we???re facing interoperability interfaces here, and mere fact; the
> currency of the USA is the Dollar, no matter what.)

:-)

best regards
keld