[cc-community] wrong license info / logo
Mike Linksvayer
ml at creativecommons.org
Fri Feb 16 18:59:09 EST 2007
On Fri, 2007-02-16 at 12:12 +0100, Alek Tarkowski wrote:
> I understand there is little that you can do about this on the
> infrastructure side, but then a vision of myself trawling through the
> internet and checking licenses sounds extremely unpleasant and tedious.
This is probably the only viable approach, but perhaps a little bit of
code can help, see below.
> Hackers to the rescue? And CC as well? Would CC be intersted in first
> estimating the scale of the problem - this would require taking a sample
> of pages that use some common licensing phrase and checking (with some
> smart script) whether they a. link properly b. provide metadata. Cause
> maybe this isn't a big problem, only one visible to us...
> If it is a problem, a similar tool could in general crawl the net, find
> pages with problems - and then someone would just have to find an email
> address and mail some standard request.
Unfortunately I don't know of a good approach to doing a directed crawl
looking for sites that claim to be under a CC license but don't provide
a link or metadata. If Google facilitated searching only the non-CC web
(which wouldn't be a great thing, and they don't seem to have done so
inadvertently; I've played with the URLs a fair bit), then searching for
(made up syntax, and substitute your language's CC notice)
"This work is licensed under a Creative Commons license" AND -cc
would be a start. But not a great one, as people using a CC image or
variation of one and no text, and people using the default language from
the license chooser ... probably got it from the license chooser. :)
So the most likely way for a bad CC implementation to be noticed is by a
human. A bookmarklet could be provided (or equivalent built into MozCC)
that you could invoke when you happen upon a bad implementation. The
code could do some basic sanity checks, attempt to detect language, and
add page to a queue to be processed when volunteers who read the
relevant language want to try to contact webmasters to encourage them to
improve their CC usage.
> But I also think that this problem shouldn't just be left alone or fixed
> - if it really is a problem, then it means that we maybe need to rethink
> our 'education' strategy, since our key tool, the licensing engine, is
> somehow being ignored by licensors.
>
> best,
>
> alek
>
> ps. the issue with sites like flickr, revver, archive.org, you name it,
> doing weird things with licensing is a different but big issue. I'm not
> sure there's one site that conforms fully with the CC model. What seems
> most annoying to me as a national lead is the inability to use localised
> licensing. Let's face it, the big licensing work occurs through sites
> like flickr. And what's the point of having localised licenses if people
> there use the generic one? I think CC HQ should maybe push harder for
> its partners to try to comply - it's a pity that even a flagship project
> like archive.org doesn't do this.
Hmmm, I'm pretty sure the archive.org does allow selection of
jurisdiction licenses, though I haven't tested in ages.
Commercial sites nearly always want to minimize potential for user
confusion and minimize engineering work -- they almost always opt for
doing the simplest possible thing, i.e., offering the six unported
licenses in a drop-down, although I always encourage them to look at the
APIs we have available that facilitate choosing any jurisdiction
license.
--
http://wiki.creativecommons.org/User:Mike_Linksvayer
More information about the cc-community
mailing list