[Userinvolvement] Supporting citation

[Tomaž Erjavec]

Dear Thorsten, all,

Thorsten Trippel je 01/10/2015 ob 17:39 napisal:
> Dear all,
>
> the Lindat example may contain a handle but this is not the classic 
> way to cite a handle which whould be hdl:11372/LRT-386 or as an 
> actionable PID it would be http://hdl.handle.net/11372/LRT-386. Though 
> the path in the URL says "handle" it does not become a handle just by 
> saying so in the path, but by being resolvable via the handle-resolver.
But surely this is just because Martin has cited it using the direct 
LINDAT URL?
But if you go to this URL it does clearly say that the way to cite this 
item is via http://hdl.handle.net/11372/LRT-386, just as you suggest above.

> Neither of them shows an obvious link to lindat. 
Well, if you use the handle that is where you wind up, and there it does 
say LINDAT (more than once:).

> So I guess we should prefer something like Footnote "The data is 
> available at http://hdl.handle.net/11372/LRT-386, provided via the 
> CLARIN Research infrastructure" or in the reference section as Branco, 
> António (2014). Nexing Corpus. http://hdl.handle.net/11372/LRT-386 
> provided via the CLARIN Research infrastructure"
>
> Zotero would not be more helpful as I do not see that Zotero provides 
> PIDs that are sustainable (i.e. independent of the hosting 
> institution). Maybe some day Zotero will be certified and provide real 
> PIDs ;-)
>
Maybe https://zenodo.org/ is a better exemplar, as it is also used for 
datasets and does provide DOIs.
Which brings me to a point which I already made with LINDAT folks, but 
was not met with much enthusiasm:
in spite of the short time our repository has been working, I've already 
had several experiences where potential depositors were put off by the 
fact that we don't use DOIs - it's a system they know and trust, whereas 
handles are to them some strange beast. I know that it is A Bad Thing to 
multiply identifiers, but maybe for the DOIs we could make an exception 
(as an option, not obligatory of course), as it does seems to be what 
 >90% of people use.
This might be getting off-topic though and something to better discuss 
in Poland.
Best,
Tomaž

> Greetings
>
> Thorsten
>
> Am 01.10.15 um 16:56 schrieb Martin Wynne:
>> Dear all,
>>
>> One of the suggestions at the User Involvement meeting last month was
>> that repositories should make it easy for users to cite CLARIN resources
>> that they have accessed and used. It was thought that this would be
>> encourage use of CLARIN resources, and and help measure usage of 
>> resources.
>>
>> I promised to make a recommendation to the Centres Committee, which I
>> have done in an email, but Dieter would like some more details of the
>> requirements, and some examples of good practice in this area.
>>
>> I think that the most important requirements are:
>> - to provide text on the landing page for a resource giving guidance to
>> users on how to cite resources (e.g. fragments of text for them to copy
>> and paste);
>> - make sure that there are persistent identifiers for online resources,
>> that can be cited in publications and which will remain valid;
>> - to inform users that citing the resources that they use is important
>> for the ongoing sustainability of the resources.
>>
>> For examples of good practice I have found the following, which provide
>> a fragment of text that can be pasted into a document (although the
>> British History one is only a URL not a handle or DOI):
>>
>> - https://lindat.mff.cuni.cz/repository/xmlui/handle/11372/LRT-386
>>
>> -
>> http://www.british-history.ac.uk/cal-state-papers/domestic/edw-eliz/addenda/1547-65/p554 
>>
>>
>>
>> Are there other requirements, and more examples? For example, would
>> links to index resources on Zotero be useful?
>>
>> Please send any comments of suggestions to the list, or to me and I will
>> summarize.
>>
>> Best wishes,
>> Martin
>>
>
>