data sharing and the polar information commons

data sharing and the polar information commons

kaitlin thaneyprogram manager, science

creative commons

This presentation is licensed under the CreativeCommons-Attribution-3.0 license.

access is step one

content needs to be legally and technically accessible

knowledge?

journal articlesdata

ontologiesannotations

plasmids and cell lines

knowledge?

journal articlesdata

ontologiesannotations

plasmids and cell lines

... how to treat? like content? software?

as a means to achieve Open Access

data? not necessarily.

(it’s complicated)

copyright and databases

what’s protected? is it legal?

facts are free

to what extent is there creative expression?

the data “rights” conundrum...

©“creative expression”

is it creative?

is it creative?

is it creative?

category errors

Non-Commercial

the problem of...

for data

Non-Commercial

what’s a commercial useof the data web?

Share Alike

the problem of...

for data

1854

issue of license proliferation

whatever you do to the least of the databases, you do to the integrated system

(the most restrictive wins)

risk for unintended consequences

Attribution

the problem of...

for data

the problem of...

for data

any license

national law / jurisdiction-based hurdles

sui generis, “sweat of the brow”

Crown copyright “level of skill”

how internat’l data sharing efforts are affected?

attribution vs. citation

which one applies? which is best fit?what’s the difference?

“credit where credit is due”

attribution:(legal entity)

“triggered by making of a copy”does it apply to facts?

how to attribute? (papers, ontologies, data)

“in a manner specified by ...”attribution stacking

citation:(gentle(wo)man’s club)

legal requirement? interoperability?

credit where credit is dueentrenched scientific norm

we shouldn’t use the law to make it hard to do the wrong thing ...

need for a legally accurate and simple solution

reducing or eliminating the need to make the distinction of what’s protected

requires modular, standards based approach to licensing

... must promote legal predictability and certainty.

... must be easy to use and understand.

... must impose the lowest possible transaction costs on users.

full text: http://sciencecommons.org/projects/publishing/open-access-data-protocol/

norms approach

set of principles (not license)

open, accessible, interoperable

create legal zones of certainty

calls for data providers to waive all rights necessary for data extraction and re-use

requires provider place no additional obligations (like share-alike) to limit

downstream use

request behavior (like attribution) through norms and terms of use

Creating norms for polar data

1. How to preserve the source information? How should the user or copier preserve the provenance of the data set. What can be required by PIC that is locally relevant and acceptable? DOIs? Something like a notice inside the data? Ping to a URL at PIC? RDFa inside a section of every database that is provided by PIC?

2. How to cite the data set? Many examples out there including http://ipydis.org/data/citations.html

3. How to preserve quality standards? Perhaps we leave it up to the users?

4. How to note and release user contributions, mashups, repurposing? Do we need release guidelines of contributions, annotations, etc. to data sets. How to reward and track individual contributions to a collective - trackback, user accounts, etc.? A simple “share alike” request?

Some draft norms of appropriate scientific behavior when using PIC data

• Acknowledge the source of the data in accordance with the wishes of the provider, and explicitly cite the data when they are used in formal scientific publication (http://ipydis.org/data/citations.html).

• Maintain a link to the original information in any derived products, ideally through a persistent identifier, such as a Digital Object Identifier.

• Understanding that the data are made available “as is” and the accuracy of the data or documentation are not guaranteed. The provider assumes no responsibility for misuse or misinterpretation.

• Notify the data provider in the manner they describe on how you plan to use the data. For projects integrally dependent on the data consider requesting collaboration and/or co-authorship from the provider.

• Share any derived products in the PIC.

• Agree to IPY Data Policy

37

others?

5.4 million bibliographic records

at worst, we’re really wrong.

at best, we’re partially right.

data without structure and annotation is a lost opportunity.

data should flow in an open, public, and extensible infrastructure

support recombination and reconfiguration into computer models, queryable by search

engine

treated as public good

resist the temptation to treatas property

embrace the potential to treat instead as a network resource