[okfn-discuss] Alpha ODI Open Data Certificates: comments please!

Mon Apr 8 10:09:50 UTC 2013

On Mon, Apr 8, 2013 at 10:29 AM, Jeni Tennison <jeni at theodi.org> wrote:

> Hi Peter,
>
> On 8 Apr 2013, at 08:42, Peter Murray-Rust <pm286 at CAM.AC.UK> wrote:
> > It is critical that data is labelled by an independent organization like
> ODI. because there are initiatives to reduce the scope of openness
> (certainly in "open Access"). For example some publishers are setting up
> (data) publication systems where there can be a  price differential between
> CC-BY and CC-NC - the latter would not earn any TBL stars, let alone 5.
>
> Agreed. But how would that work? A group could get together to pay for
> CC-BY and then republish under that licence, couldn't they?
>
> The problem in Open Access and now starting to emerge in Open Data is that
(a) many "OA" proponents think that licences don't matter and that the only
thing that matters is human visibility. Thus "free to read" but "forbidden
to copy" or "only reusbale for NC" is now commonly labelled as "Open
Access". Much of this is being promoted by schol publishers who are only
offering NC product but aclling them "fully open Access". Indeed many
publishers aren't even offering licences but still using this label. Not
enough people care about this. OASPA require BOAI compliance but the big
publishers ignore this.

This practice has rendered "Open Access" and "libre" useless labels. It is
starting to spread to data and I'd really value efforts to isolate it
before it gets a hold there as well.

> > Universities have almost completely failed to show any interest in
> labelling their published data but I think research funders would welcome
> an ODI stamp.
>
> Excellent. What I think we should aim towards is a statement like "data
> must be published to 'standard' or 'exemplar' level against the ODI Open
> Data Certificate". This is why we need to make sure we get the levels
> pitched right to enable these statements to be made and to be achievable.
>

Yes. The problem is that achieving even Standard is a significant effort
for academia. The current best we can hope for in long-tail science is
static data labelled with an ODI-compliant licence and dumpable where
possible.

Although it's current only marginally related the seriousness of the
problem is shown by the vested interests lobbying against content-mining
and the for the development of a new set of "Licences for Europe". That
means we are likely to get a series of non-ODI-compatible "Open Data"
licences issued by Europe unless we can change direction. see
http://en.wikipedia.org/wiki/David_Hammerstein_Mintz on
http://tacd-ip.org/archives/907

>
> > As a small provider of Open Data I understand the technical difficulties
> of making a dynamic data set completely available ("dump") and in some
> cases we may have to have a best-endeavour. The biosciences do very well in
> making their dynamic data available but they also get considerable funding
> to do it. I think stamps in that area would be very valuable
>
> It's tricky, and we need to balance ease of implementation and utility of
> dumps. If it's really hard to achieve the 'standard' level it will devalue
> the certificate as a whole, so I'd really welcome feedback about the
> feasibility of the requirements.
>

I have copied Ross Mounce and Sophie Kershaw - our Panton Fellows - who
have been working a lot in this area. Maybe we can visit?

There's a big difference between "big-science" - such as climate and
long-tail-science such as chemistry and bench bioscience. In the latter
data are fragmented, there is no training (Sophie is changing that), no
explicit money, no desire to publish data, etc. The practices are
increasingly set by scholpub some of whom see this as potentially a new
digital land grab. There is almost no way the average long-tail scientist
can achieve "standard" but it critical the data gets out and is labelled
ODI-compliant.

BTW as you can see I favour the OSI/BOAI/ODI-compliant language. "Open"
will inevitably be devalued

P.

-- 
Peter Murray-Rust
Reader in Molecular Informatics
Unilever Centre, Dep. Of Chemistry
University of Cambridge
CB2 1EW, UK
+44-1223-763069
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/okfn-discuss/attachments/20130408/cb357b8a/attachment.html>