[Wikimedia-l] Copyright infringement - The real elephant in the room

[Samuel Klein]

Aside @Fae: the tineye crew are curious & quite pro-freeculture, I bet they
would be glad to help design a bot that uses their API to check image
copyvios.
On Nov 13, 2013 6:48 AM, "Fæ" <faewik at gmail.com> wrote:

> On 13 November 2013 07:40, James Heilman <jmh649 at gmail.com> wrote:
> ...
> > Our biggest issue is copyright infringement.
> ...
>
> Thanks for raising this James.
>
> Yes, this is an issue but if you are gunning for elephants this month,
> I really don't think the copyright elephant is the biggest one in the
> herd.
>
> As a practical example of the tools we already have in place,
> yesterday I was facilitating an edit-a-thon for women in science with
> King's College London and we had one of the example stubs we had
> created on the English Wikipedia up on a projector. Within literally
> *minutes* of creation it had been (correctly) flagged by a bot as a
> possible copyright violation as some of the text had been cut & past
> from King's own website; one of the participants quickly re-wrote it
> using their own words. As the communications manager was sitting next
> to me at the time, no doubt she found this rather reassuring, even
> though in parallel she was asking about how best to "officially"
> release text. :-)
>
> We have a more complex problem with how images uploaded to Wikimedia
> Commons can be flagged where they match images found elsewhere on the
> internet, this is something that may be done by a future bot but we
> might need to partner with someone like Google Images or Tineye to
> make this truly effective. Having run my own experimental bots on this
> area, I would love to see this become a funded project.
>
> PS with regard to OTRS verification, we could do with better standards
> for verification, at the moment volunteers like myself are left to use
> our own judgement about what checks to make. I tend to double check
> text or images being released with Google, just in case, as well as
> doing "whois" checks on email domains. These sorts of checks could
> become part of OTRS guidelines and would make the reliability of OTRS
> tickets a notch higher.
>
> Cheers,
> Fae
> --
> faewik at gmail.com http://j.mp/faewm
>
> _______________________________________________
> Wikimedia-l mailing list
> Wikimedia-l at lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request at lists.wikimedia.org?subject=unsubscribe>