Richard, hello.
On 25 Nov 2015, at 18:42, Richard Rankin wrote:
> We gave a workshop today on Research Data Management and the question
> was asked - I can make the data available but it is meaningless
> without using the software I have been developing over my career -
> don't mind making data available - however no way am I making the
> source code of my software
Then I for one would simply disagree that the data has been made
available.
If the data had been 'made available', but in encrypted form, would that
be any different? I don't think so. This seems flatly contradictory to
the spirit of the requirements, and this isn't a question of data
preservation (which is the direction this thread has taken).
The point of the stipulation that data be made available is to allow
others in the community to check one's analysis. Now, one may question
how practical this reexamination is, or how useful, or how likely, and
the answer will probably depend on the discipline. But the underlying
goal is an extension of the centuries-old scientific principle of 'show,
don't tell': the data is released so that it can in principle be
criticised or reanalysed, and if it can't be so reanalysed, even in
principle, then this process can't work.
Analysis _software_ is different. I can see why someone would want to
keep their analysis software secret, and simply publish the method it
implements (I'm not convinced this is a wise approach, but it's at least
strongly defensible in some cases). But private software acting on
public data is different from refusing to release the data at all.
All the best,
Norman
--
Norman Gray : https://nxg.me.uk
SUPA School of Physics and Astronomy, University of Glasgow, UK
|