[Esip-preserve] DataCite to require "landing pages"

[Bruce Barkstrom]

What is a "full citation of the data"?

Are they expecting a single data file?
A collection of data files?
A collection of data files, together with associated metadata files
and documentation files?
A collection (or list) of references to individual data values (and ...)?

A concrete example of the last kind of citation comes from
the classic example of the solar constant at
http://www.pmodwrc.ch/pmod.php?topic=tsi/composite/SolarConstant
If the primary concern is giving a summary reference, this URL would
be fine.  If, on the other hand, you wanted to check the work and do
something different than what the authors had done, you need to be
able to cite which measurement entered a solar constant record for
a particular day (note the color coding on the three different time series
in the first figure).  The total period covers about 1978 to 2011, or about
30 years.  Since one year has about 1/3 of 1000 days, 30 years should
have about 10,000 days.  In other words, to provide a citation of the
exact data value that was used in a pretty important time series of
the "solar constant", the citation will need to identify 30,000 source
data values.

Bruce B.

On Wed, Feb 22, 2012 at 12:55 PM, Greg Janée <gjanee at eri.ucsb.edu> wrote:
> There are some changes brewing at DataCite regarding DOI requirements.  Per
> Joan Starr:
>
> "There is a very strong push within the group for a requirement that all
> DOIs point to landing pages and not to the object directly.  [For submitters
> that don't provide their own landing pages] DataCite is going to create
> canned landing pages based on the metadata that submitters provide."
>
> What constitutes a "landing page"?  Again, per Joan:
>
> "The landing page typically contains one or more of the following:
>  +full citation of the data
>  +statement on access to data (such as a link to the data or usage
> restriction information)
>  +associated metadata
>  +information, software, or context required for unpackaging, reading and
> interpreting the data"
>
> I'm curious what this group thinks about this requirement.  It certainly
> narrows the scope of applicability for DOIs.
>
> -Greg
>
> _______________________________________________
> Esip-preserve mailing list
> Esip-preserve at lists.esipfed.org
> http://www.lists.esipfed.org/mailman/listinfo/esip-preserve