JiscMail Logo
Email discussion lists for the UK Education and Research communities

Help for JISCMRD Archives


JISCMRD Archives

JISCMRD Archives


JISCMRD@JISCMAIL.AC.UK


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

JISCMRD Home

JISCMRD Home

JISCMRD  May 2013

JISCMRD May 2013

Options

Subscribe or Unsubscribe

Subscribe or Unsubscribe

Log In

Log In

Get Password

Get Password

Subject:

Re: Controlled access to data in EPrints

From:

Rachel Bruce <[log in to unmask]>

Reply-To:

Rachel Bruce <[log in to unmask]>

Date:

Sat, 18 May 2013 11:56:48 +0000

Content-Type:

text/plain

Parts/Attachments:

Parts/Attachments

text/plain (135 lines)

Since metadata has come up I'll share a couple of links. The Discovery initiative at Jisc ( working with RLUK ) was all about open metadata being at the heart of ensuring resources are finable and hopefully in most cases accessible. In order to push towards a change in the resource discovery environment we - with others such as the BL , NLS, NLW, TNA, Sconul etc back in 2009 - agreed that open metadata was the key to a more effective and efficient discovery infrastructure. At the time we restricted ourselves to thinking about archives, museums and library resources but always felt the approach should be more broadly relevant. Starting out then we were not sure whether open metadata would be easy, especially because of vendor interests and different business models in the space. Anyhow it turns out it has been ok and metadata has been able to be open, and most often using CC0.  The idea of course with Discovery is because it is important to aggregate metadata for different purposes, to allow the recombination of different metadata sets, to allow it to be part of different systems and workflows and to be able to perhaps cut down the multiple creation and effort that there must be in the system. So the initiative is different in the delivery model to Europeana and DPLA but they reach similar conclusions and in terms of metadata approach we're all in line I think. 

These documents are little old now but for reference:

http://discovery.ac.uk/businesscase/principles/

This one covers the licensing of metadata and other types of data - perhaps indicating a wider range of licences as "truly open" than strict open definitions would allow but the guide does cover the main issues around licensing and open data - it was written in 2011 so some things have changed since but largely it remains true. 

http://discovery.ac.uk/files/pdf/Licensing_Open_Data_A_Practical_Guide.pdf
cheers, Rachel
-----Original Message-----
From: JISC Managing Research Data Programme [mailto:[log in to unmask]] On Behalf Of Cameron Neylon
Sent: 18 May 2013 11:16
To: [log in to unmask]
Subject: Re: Controlled access to data in EPrints

Absolutely agree with Chris' point that there are cases where data should not be open and where registration may be appropriate. I don't think we do have a good name for this, which in part is why I try resist the (further) dilution of "open" where I can. We should seek to find some good descriptive terms so we can have good conversations about what is the appropriate stance to take on specific data resources.

Something that people might find interesting with respect to data licensing also came into my stream today which illustrates some of the issues around interoperability for the commercial use of data:

http://knoesis.wright.edu/faculty/pascal/pub/nomoneylod.pdf

Cheers

Cameron

On 18 May 2013, at 10:36, Chris Rusbridge <[log in to unmask]> wrote:

> I think Cameron is right: the CC licences with NC and SA clauses are not Open (in the BOAI sense), and data which is freely accessible after a registration is also not Open in that sense. CC0 is the better choice if you want to be strictly Open.
> 
> We should also all be clear though: there are MANY cases where research data should NOT be Open in that sense, and MANY cases where some form of registration and agreement of terms is entirely appropriate. There has been too little discussion about this important issue. I don't think the old UKDA argument about registration being important in order to know who's accessing the data and ensuring credit is sufficient in itself, but where there is confidential, private and/or sensitive data involved, or "anonymised" data that could be disclosive if combined with other data, then registration and agreement to terms is the very least that should apply.
> 
> Is there a good name for this class of data that is accessible without fee after agreement?
> 
> --
> Chris Rusbridge
> Mobile: +44 791 7423828
> Email: [log in to unmask]
> Adopt the email charter! http://emailcharter.org/
> 
> 
> 
> 
> On 16 May 2013, at 14:19, Cameron Neylon wrote:
> 
>> Dear All
>> 
>> It might be worth going back to some of the history of the term "open" in this context. It derives from usage in the free and open source software movement as well as the free culture movement and has been adopted into the data context.
>> 
>> When Open Access was defined in the BOAI it was defined essentially as "...freely available on the internet...for use for any purpose...with no limitations save those intrinsic to gaining access to the internet itself" [paraphrased].
>> 
>> So I would argue that putting login requirements does mean something 
>> is not "open" in the sense that it is intended by those who coined 
>> and promoted the terms Open Source, Open Data, or Open Access. That 
>> doesn't mean its necessarily the wrong thing to do, just that it is 
>> no longer really open, but "accessible" or "usable". (I happen to 
>> personally believe it *is* (usually) the wrong thing to do but that's 
>> a separate argument - and there are clear cases where access controls 
>> are necessary but in my view these are edge cases.)
>> 
>> The licensing argument is more complex. The reason CC0 was developed was because of a significant quantity of work that identified real risks of using copyright licenses for data. Creative Commons have sought to ensure that those using CC licenses for data don't get a nasty surprise in terms of what they cover in v 3.0 and the coming v 4.0 but there are real limitations in relying on those licenses to limit data usage and the unexpected side effects can be disastrous. I spent too long working through use cases for trying to combine incompatible licenses for data and other scholarly objects but the long and short of it is that it is very easy to create a series of objects which cannot be recombined and that this risk is at its very worst for data - which will frequently carry other obligations.
>> 
>> Consider the following example. A community health professional obtains mapping data which is CC BY-SA and wants to use this to create a map for their patient group to show each other where they live so they can stay in contact. However it would be inappropriate, and quite possibly illegal, for the CHP to give the patients the right to redistribute each others address information. But by creating a derivative data work which triggers the SA requirement the advocate is *required* to provide it under a CC BY-SA license, explicitly allowing redistribution. The problem is one of interoperability and splitting the commons. By using more restrictive terms you're basically re-enacting the enclosure act - and you lose the value of being able to recombine the whole set of resources available and interoperable.
>> 
>> It's these incompatibilities that lead those of us working in this space to recommend CC0 as the licensing/waiver mechanism for data complemented by the building up of our existing cultural norms around citation and attribution to support good social behaviour. Its particularly crucial for metadata which by its nature is most suited to re-aggregation, indexing, and repurposing which is why DPLA and Europeana are both adopting that as a standard - alongside other organisations from Merck to CrossRef releasing data under CC0.
>> 
>> Hope this is helpful - for more information I'd recommend some of the older Science Commons information on data sharing and licensing. That's where these issues were worked through in most detail from around 2005-2009.
>> 
>> Cheers
>> 
>> Cameron
>> 
>> 
>> On 16 May 2013, at 12:26, Dr Bill Worthington <[log in to unmask]> wrote:
>> 
>>> Hi all,
>>> 
>>> this is a useful thread. I have the following reflections on it. 
>>> 
>>> - i agree with Louise, in that I believe you can still say data is 
>>> open, even if it isn't available for immediate click-thru
>>> 
>>> - at UH we are planning to use the same back end archival storage as 
>>> Leeds/Southampton, but with dSpace, rather than ePrints
>>> 
>>> - the nature of this technology means that we have no choice but to 
>>> deploy a request-retrieve-deliver journey for the dataset, 
>>> corresponding to request-wait-pickup for the requester
>>> 
>>> - this is not a bad thing, it allows us to: have much better 
>>> evidence of actual use than simple click stats, with onward benefit 
>>> to inform a retention/disposal schedule; and, insert workflow 
>>> triggers to deal with any of Rachel's four scenarios, resulting in a 
>>> request-decide-retrieve(or decline)-deliver data journey
>>> 
>>> - Robin's post brings up the biggest challenge as far as I can see: requests that require a yes/no decision from their depositor. Datasets are likely to outlive their owners (at UH we use the term Data Steward because UH usually owns the data) so who makes the decision when the researcher moves on? We have a proposed workflow for depositors which includes 'criteria for restricted access', which could then be acted on by an administrator in the absence of the owner/PI/data steward, but I am quite skeptical about how practical this will be.
>>> 
>>> - finally, to go back to Louise's comment about CC0, we will probably use CC-BY-NC-SA. I would rather use CC-BY-NC because SA creates a license cascade which I think is impractical. Someone pointed out that the NC clause is against the open data concept too. Any thoughts on this? (maybe in another thread).
>>> 
>>> best wishes,  Bill
>>> 
>>> ------------------------------------------------
>>> 
>>> Research Data Management Projects Manager (JISC Managing Research 
>>> Data Programme) Projects website: 
>>> http://www.herts.ac.uk/research-data-toolkit
>>> 
>>> Project tags:  #rdtk_herts, #rdmtpa_herts
>>> 
>>> Dr. W J Worthington
>>> 
>>> University of Hertfordshire
>>> PO Box 109, College Lane
>>> Hatfield
>>> Hertfordshire
>>> AL10 9AB
>>> T: +44 (0)1707 284000  ext. 77883
>>> E: mailto:[log in to unmask]
>>> Twitter, Skype:  wjworthington
>>> 
>>> ####################################################################
>>> #### To unsubscribe from the JISCMRD list go to 
>>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=JISCMRD
>> 
>> #####################################################################
>> ### To unsubscribe from the JISCMRD list go to 
>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=JISCMRD
> 

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

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

Top of Message | Previous Page | Permalink

JiscMail Tools


RSS Feeds and Sharing


Advanced Options


Archives

October 2016
September 2016
August 2016
July 2016
June 2016
May 2016
April 2016
March 2016
February 2016
January 2016
December 2015
November 2015
October 2015
September 2015
August 2015
July 2015
June 2015
May 2015
April 2015
March 2015
February 2015
January 2015
December 2014
November 2014
October 2014
September 2014
August 2014
July 2014
June 2014
May 2014
April 2014
March 2014
February 2014
January 2014
December 2013
November 2013
October 2013
September 2013
August 2013
July 2013
June 2013
May 2013
April 2013
March 2013
February 2013
January 2013
December 2012
November 2012
October 2012
September 2012
August 2012
July 2012
June 2012
May 2012
April 2012
March 2012
February 2012
January 2012
December 2011
November 2011
October 2011
September 2011
August 2011
July 2011
June 2011
May 2011
April 2011
March 2011
February 2011
January 2011
December 2010
November 2010
October 2010
September 2010
August 2010
July 2010
June 2010
May 2010
April 2010


WWW.JISCMAIL.AC.UK

Secured by F-Secure Anti-Virus CataList Email List Search Powered by the LISTSERV Email List Manager