Print

Print


Thanks Phil. It seems the chances of both a reader or a machine finding 
the data from an article DOI will improve if it also has a data DOI. 
Crossref and Datacite have details on 'cross cite' content negotiation 
at http://crosscite.org/cn/

Re the Amsterdam Manifesto statement that data should be deposited in 
"persistent public repositories",  the PREPARDE draft guidelines for 
publishers on recommending repositories for data deposit (at 
http://bit.ly/ZhYHZl ) are a useful guide to work out what that means. 
The guidelines include factors like whether the repository is well used, 
as an indicator of likely persistence, along with more rigorous evidence 
like certification on 'trusted repository' standards like Data Seal of 
Approval and ISO 16363.


Angus



On 30/04/2013 11:56, Phillip Lord wrote:
> Angus Whyte<[log in to unmask]>  writes:
>> Thanks for the reply. I think the Amsterdam Manifesto is concise and clear on
>> most points, but that one could be a little clearer. I wonder if it should be
>> read as  '...persistent and publicly funded' e.g. on the basis that public
>> institutions are more likely to persist, or as '..persistent and publicly
>> accessible" ?
> It's possibly a strange distinction to make. I mean, lots of code (and
> code is data) goes into github because it is good.

>> When I was referring to publisher's landing pages I only meant as a location a
>> reader might end up at if they were looking for data, not as the ideal place
>> to put it or find data. I'm assuming that readers will be more likely to find
>> within the article itself a crossref DOI to the article landing page than a
>> datacite ref to the data landing page, and on that basis the correct place for
>> a statement identifying how data may be accessed is in the paper itself. Of
>> course if they are already at the data landing page they don;t need to see any
>> statement on how to access the data.
> I believe at the moment that CrossRef does not release linked data which
> can be used to get DataCite DOIs; DataCite do this in some cases,
> however. So there is, at least, a chance given a CrossRef DOI of getting
> to a dataset.
>> Given that, would it in your view be appropriate to include, as standard, a
>> statement in the Acknowledgement list as well as (preferably) a citation to
>> the data in the Reference list?
>>
> I would look at use cases. I can see two broad ones:
>
> Given a paper, do I, the scientist, have a chance of working out where
> to get the data that was used, by a technique more rapid than emailing
> the authors of the paper.
>
> Given a paper, do I, the computer, have a change of working out....
>
> For the former, I think, putting stuff into the reference list would be
> most helpful. For the latter, putting things in anywhere, so long as it
> has a clearly defined syntax (like a DOI or URI to a standard
> repository) is most helpful.
>
> Phil
>
> ########################################################################
> To unsubscribe from the JISCMRD list go to
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=JISCMRD
>


-- 
Dr Angus Whyte
Senior Institutional Support Officer
Digital Curation Centre
University of Edinburgh
Crichton St, Edinburgh EH8 9LE
+44-131-650-9986



The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

########################################################################
To unsubscribe from the JISCMRD list go to
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=JISCMRD