[okfn-discuss] Annotating Open Images with licence and authorship to prevent copyfraud

[Jonas Öberg]

Hi Peter, and everyone who's chimed in to this discussion so far. I'm
a newcomer to this list, so allow me to introduce myself too in my
response.

Since March this year, I'm a Fellow of the Shuttleworth Foundation
where I work with metadata for attribution and licensing information,
and all the promises and problems related with it. So I'm very
interested in this thread!

> If the images were indelibly stamped with licences and authorship
> copyfraud would be very much harder. I suspect this would also be welcomed
> by authors themsleves as it promotes their authorship.

Full agreement on this point from Peter, and I'm happy to see others
have implemented or thought similarly. Now, stamping or watermarking
has its own problems: it gets distorted when scaling, for instance.
This doesn't mean that it shouldn't be used when possible, but that
there are also other mechanisms that could be employed, and embedding
metadata in the works is one such method.

Peter mentioned separately in the thread that metadata is very easily
stripped from a work. This is very true. The International Press and
Telecommunications Council of which we're a member have studied this
in social media and found out that metadata *in most cases* gets
stripped when distributing a work. There's been some changes recently
from Flickr to retain the metadata but otherwise, it's not very
dependable.

So for a short-term solution, it's not viable. But it's useful in
closed environments where you have an idea of the toolchain involved
and can implement relevant support it in. It might also be viable in
the long term, and the people I'm speaking to from the industry is
certainly pushing in the direction of increased retention of metadata.
The UK based Copyright Licensing Steering Group (CLSG) released a
draft voluntary Code of Practice for creating and retaining Metadata
in images just before the summer:
http://www.clsg.info/codeofpractice.html

The work that I do at the Shuttleworth Foundation is in Commons
Machinery, an initiative to delve deeper into the metadata field, to
push for (reasonable) retention of metadata, building prototypes to
show how this could work, and contributing to the standardisation
efforts in this area. You're welcome to look at our work on
http://commonsmachinery.se (we have an awesome introductory video
there too).

You can also watch a presentation that I gave at the RMLL earlier this
year which outlines our current work and thinking.
http://video.rmll.info/videos/embedded-metadata-for-licenses/

Oh, and I should also mention that we're looking to hire another
developer to join our team :-)
http://commonsmachinery.se/get-involved/job-openings/


Sincerely,
Jonas