Re: Machine-readable data licenses

Hi Deidre,

On Mon, Mar 24, 2014 at 6:50 PM, Lee, Deirdre <Deirdre.Lee@deri.org> wrote:
...

> What are your thoughts on how ODRS does/could address:
>
> a)      interoperability issues of data licenses
>
Licence compatibility is a difficult issue, I've done some fairly extensive
research as part of the ODRS work, and have collected together pointers
here:

https://github.com/theodi/open-data-licensing/blob/master/guides/licence-compatibility.md

Compatibility is a very difficult thing to capture, mainly because many of
the significant details are in licensing terms that aren't easy to capture
in a machine-readable way. Licensing tools typically focus on the key
provisions of a licence, e.g. share-a-like but can't give a definitive
answer (you need a lawyer to run the legal code on ;)

My preferred approach is to:

* ensure licenses and key pieces of information are discoverable, including
attribution statements
* that licences are described with some basic metadata (the Creative
Commons ccRel vocabulary covers this, as do several others)

I believe that's enough to provide some clarity for data consumers. Coupled
with general encouragement to use standard rather than custom licences, I
think this will move things in the right direction.

>  b)      Machine-readable SLA terms, with a focus on industry reuse
> https://www.w3.org/2013/dwbp/wiki/Use_Cases#Machine-readability_of_SLAs
>

I'd take the same approach here, capture essential details, e.g. support
contact details, support windows, rate limiting and usage quotas, but also
ensure that there are clear pointers to human-readable documentation to
cover the rest.

Cheers,

L.

-- 
Leigh Dodds
Freelance Technologist
Open Data, Linked Data Geek
t: @ldodds
w: ldodds.com
e: leigh@ldodds.com