[Andy Turner]

I agree. If the research generates metadata that indicates what data were used in the conduct of the research and what data are used in the production of a specific result based on a program then all is good from a scientific standpoint. More than details of data are wanted in the metadata, to detail computational requirements and parameterisation. In parameterising, a large amount of data could be used, and this is where the line between software and input data gets a bit fuzzy.
If researchers can create DOIs that link to the two key sets of data used (lists of DOIs) then what you can have is something like two or three data references (that proxy to the detailed data references) that can be used in journal publications without the references, acknowledgements or indeed author lists becoming too big.
But yes, back to the point, I agree. Trying to regulate and make obligations to cite in specific ways can be restrictive and so I think this is best avoided. So my advice to researchers would be to use only the most liberal licenses and that hopefully other researchers will do the right thing so that they can get due credit for platforming further research.
Only the best research cites all sources (and resources used) and has all the appropriate metadata so that others can readily interrogate the results and process of producing these by reproduce them step by step so that the details can be inspected. It is up to the research community to their best and only the best should get a commensurate 5* status or whatever.
HTH
– Show quoted text -From: chris.morris=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of chrishmorris
Sent: 11 March 2015 10:00
To: ***@***.***-groups.org
Subject: Re: [rda-legalinterop-ig] Implementation Guidelines outline for the Legal Interoperabil…
HI,
I agree that “Attribution and citation are an essential part of the research process for providing appropriate credit…” and also that “ They should be encouraged
through community norms rather than through legal requirements.”
I think this point is important, because when later studies integrate data from many sources, the requirement to give attribution becomes onerous. Social norms can ensure that behaviour is usually appropriate. I doubt whether any legal wording could properly express the appropriate obligations if a researcher for example:
– Downloads the whole Protein Data Bank
– Develops and runs some analysis code, and publishes some generalisations about the evolution of protein folds
– Illustrates the general conclusions by comments on some specific proteins
– Does not mention in the paper other specific database entries which were edge cases during the code development
Being able to do research like this is a great benefit of open data. But the obligations to cite that arise are subtle.
Regards,
Chris Morris
STFC
From: pfuhlir=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of puhlir
Sent: 09 March 2015 22:29
To: ***@***.***-groups.org
Subject: [rda-legalinterop-ig] Implementation Guidelines outline for the Legal Interoperability Principles and updated agenda for breakout session
Dear colleagues,
Attached (see below) is an initial outline of the Implementation Guidelines for the draft Principles on Legal Interoperability of Research Data. Also included (again) is a draft of the Principles themselves and final version of the discussion agenda for our breakout session tomorrow afternoon at 4:00 p.m. Finally, there is an updated set of instructions for the case studies, which have gone from the initial product to the last one. We will have these documents available as handouts at the session.
Please provide your written comments on the draft Principles and the Implementation Guidelines outline by Monday, March 23. We will resume our regularly scheduled conference calls later that week
Thanks,
Paul, Enrique, and Bob
I agree. If the research generates metadata that indicates what data were used in the conduct of the research and what data are used in the production of a specific result based on a program then all is good from a scientific standpoint. More than details of data are wanted in the metadata, to detail computational requirements and parameterisation. In parameterising, a large amount of data could be used, and this is where the line between software and input data gets a bit fuzzy.
If researchers can create DOIs that link to the two key sets of data used (lists of DOIs) then what you can have is something like two or three data references (that proxy to the detailed data references) that can be used in journal publications without the references, acknowledgements or indeed author lists becoming too big.
But yes, back to the point, I agree. Trying to regulate and make obligations to cite in specific ways can be restrictive and so I think this is best avoided. So my advice to researchers would be to use only the most liberal licenses and that hopefully other researchers will do the right thing so that they can get due credit for platforming further research.
Only the best research cites all sources (and resources used) and has all the appropriate metadata so that others can readily interrogate the results and process of producing these by reproduce them step by step so that the details can be inspected. It is up to the research community to their best and only the best should get a commensurate 5* status or whatever.
HTH
Andy
From: chris.morris=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of chrishmorris
Sent: 11 March 2015 10:00
To: ***@***.***-groups.org
Subject: Re: [rda-legalinterop-ig] Implementation Guidelines outline for the Legal Interoperabil…
HI,
I agree that “Attribution and citation are an essential part of the research process for providing appropriate credit…” and also that “ They should be encouraged
through community norms rather than through legal requirements.”
I think this point is important, because when later studies integrate data from many sources, the requirement to give attribution becomes onerous. Social norms can ensure that behaviour is usually appropriate. I doubt whether any legal wording could properly express the appropriate obligations if a researcher for example:
– Downloads the whole Protein Data Bank
– Develops and runs some analysis code, and publishes some generalisations about the evolution of protein folds
– Illustrates the general conclusions by comments on some specific proteins
– Does not mention in the paper other specific database entries which were edge cases during the code development
Being able to do research like this is a great benefit of open data. But the obligations to cite that arise are subtle.
Regards,
Chris Morris
STFC
– Show quoted text -From: pfuhlir=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of puhlir
Sent: 09 March 2015 22:29
To: ***@***.***-groups.org
Subject: [rda-legalinterop-ig] Implementation Guidelines outline for the Legal Interoperability Principles and updated agenda for breakout session
Dear colleagues,
Attached (see below) is an initial outline of the Implementation Guidelines for the draft Principles on Legal Interoperability of Research Data. Also included (again) is a draft of the Principles themselves and final version of the discussion agenda for our breakout session tomorrow afternoon at 4:00 p.m. Finally, there is an updated set of instructions for the case studies, which have gone from the initial product to the last one. We will have these documents available as handouts at the session.
Please provide your written comments on the draft Principles and the Implementation Guidelines outline by Monday, March 23. We will resume our regularly scheduled conference calls later that week
Thanks,
Paul, Enrique, and Bob

Discussion