[open-science] clarification on effectiveness of mandates

[Peter Murray-Rust]

On Mon, Apr 2, 2012 at 7:19 PM, Heather Piwowar <hpiwowar at gmail.com> wrote:

> [whoops, to the list this time]
>
> Peter, thanks for this discussion.
>
> I'd say it is important to keep the following three distinctions really
> clear when talking about open data mandates and their effectiveness:
>

Absolutely. And this is about data.

>
> *what is mandated*:  OA to literature and open data are very different,
> with different motivations and hurdles for the researcher.  Making
> parallels to mandates for OA literature risks confusing people on data, and
> should perhaps be done only when making these differences very clear.
>

I assume people on this list are more clued up than the average. My
argument - which was sketchy - was that funder mandates for OA papers are
only 55% successful by Wellcome. That mandates for data (which requires a
lot more work) is likely to be much worse. A PDF is relatively easy to
deposit, a dataset is 10 times harder at least


>
> *who does the mandating*:  journals or funders.  There is evidence that *
> journal* mandates are quite effective, and evidence they are one of the
> most effective tools for encouraging public data archiving.
>

Yes - and they come about by editors or societes insisting on good science.


>  I'm not familiar with any evidence on *funder* mandates.  There may be
> some, say with GWAS data funded by the NIH?
>

It's probably early days. But broad policy can only come from funders, not
from journals.

>
> *how mandates are monitored*:  Clearly mandates to "share on request" are
> likely much less effective than mandates to "provide an accession
> url/number before we'll consider your contract complete."
>
> Agreed. Unfortunately a URL can point at something far less than we want.

Was talking to a senior scientist today - s/he said "the dept will do as
little as possible to comply with the res council - just to get them off
our backs". So data mandates will not de facto deliver.


>
>
-- 
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/open-science/attachments/20120402/9856c4c5/attachment-0001.html>