[open-science] [SCHOLCOMM] Libre open access, copyright, patent law, and, other intellectual property matters
jtw at del-fi.org
Wed Mar 21 23:58:17 GMT 2012
I'm going by the BBB declarations.
Budapest: only role for copyright is attribution (not commercial
restriction as in CC NC licenses, and one might even argue, not
share-alike to compel behavior in derived works either, though that's
Bethesda and Berlin: The author(s) and copyright holder(s) grant(s) to
all users a free, irrevocable, worldwide, perpetual right of access to,
and a license to copy, use, distribute, transmit and display the work
publicly and to make and distribute derivative works, in any digital
medium for any responsible purpose, subject to proper attribution of
authorship, as well as the right to make small numbers of printed
copies for their personal use.
Note: "all users" doesn't include permission to say "no commercial users".
I have personally bought text mining services from multiple companies,
both in my commercial startup days (would have been a violation of the
NC clause) and while building the initial corpus behind the Neurocommons
(see corpus at http://neurocommons.org/page/Text_processing_pilot built
using Temis commercial software and services, of which the latter would
definitely violate NC clauses).
I also know that for-profit companies like pharmaceutical organizations
fall under assertions of commercial use under BY NC all the time, and
based on personal conversations I have had around the adoption of
scientific reports, I know that the restriction of text mining is one of
the specific reasons for the adoption of the NC clause.
Scientific reports is free of cost, but it is far from an open access
journal. Copies of the papers may be deposited in repositories, and
those copies qualify under Berlin and Bethesda as green / gratis OA, but
that doesn't make the journal itself OA any more than being allowed to
archive out of a toll based journal makes that journal OA itself.
This is why it's important to keep the definitions clear. Gratis OA is
about archives, not journals. Journals are adopting the rhetoric of
gratis OA and co-opting the libre OA in the definitions. I personally
think that's bad. Your mileage may vary.
On 3/21/2012 4:07 PM, Heather Morrison wrote:
> hi John,
> With all due respect ~ it appears that your definition of open access and mine do not coincide. Details of my perspectives on OA can be found in my draft dissertation chapter on the topic, much of which is definitional in nature:
> To get back to my question - I do not see that the Scientific Reports' CC licenses would stop text or data mining. Would you, or someone else, care to answer that question?
> As for Scientific Reports' numbers - the journal just launched mid-2011. PLos ONE achieved the number of 14,000+ per year in its fifth year. Nature is entering a field with competitors (notably PLoS ONE, but there are others), while PLoS ONE was breaking new ground.
> Heather Morrison, MLIS
> Doctoral Candidate, Simon Fraser University School of Communication
> The Imaginary Journal of Poetic Economics
> On 2012-03-21, at 9:47 AM, john wilbanks wrote:
>> Nope, Scientific Reports fails every single definition of OA. Non commercial restrictions = FauxA, not OA. Just because Nature calls it OA doesn't make it so, and when an author retains copyrights but not actual rights to do stuff, the copyright status is not relevant. Usage is what matters, not who the owner is...when the owner has signed away rights to allow usage.
>> The community knows this, which is why SR has just over 300 papers after a year and plos one has 14,000+ per year.
>> On 3/21/2012 8:10 AM, Heather Morrison wrote:
>>> Would you agree with me that Nature Publishing Group's Scientific Reports provides a good role model? Scientific Reports is a fully open access journal. Authors are offered two creative commons license choices, Attribution-Noncommercial-Sharealike or Attribution-Noncommercial-Noderivatives.
>>> In either case, it is the author, not the publisher, who retains copyright. Here are some of the details (from http://www.nature.com/srep/faqs/openaccess-faqs.html):
>>> Who retains copyright of the open access articles?
>>> Authors retain copyright for their article, with content licensed under one of two Creative Commons licences. The author can choose to opt for:
>>> Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. The author will thereby permit dissemination and reuse of the article, and so will enable the sharing and reuse of scientific material. It does not, however, permit commercial exploitation or the creation of derivative works without specific permission. To view a copy of this licence visit http://creativecommons.org/licenses/by-nc-nd/3.0
>>> The other choice is the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License, which allows readers to alter, transform, or build upon the article and then distribute the resulting work under the same or similar licence to this one. The work must be attributed back to the original author and commercial use is not permitted. To view a copy of this licence visit http://creativecommons.org/licenses/by-nc-sa/3.0
>>> Comment: my interpretation is that data or text mining would be allowed with either option. What is not permitted is resale, which I consider a sensible option for authors and publishers alike. NPG has also committed to contributing $20 to Creative Commons for every article CC licensed, which strikes me as a very good role model. So, too, is offering authors a choice of license. Much as I like PLoS, as an author when I get around to publishing I would likely consider Scientific Reports over PLoS ONE because of the license choices.
>>> Heather Morrison
>>> The Imaginary Journal of Poetic Economics
>>> On 2012-03-21, at 7:48 AM, john wilbanks wrote:
>>>> We must think *strategically* and not *academically*.
>>>> It isn't helpful to frame the text mining debate in copyright terms. Text mining is a contractual problem, not a copyright problem, but because publishers insist on layering text mining prohibitions on top of copyright licenses, they have conflated the two in a way that libre open access can untangle.
>>>> Here's an example of how elsevier does this, but they're hardly alone.
>>>> Either they have to stop bundling prohibitions on text mining with licenses, stop using DRM to enforce prohibitions, and stop chilling bulk downloading by text miners without threatening lawsuits (which then we can use your arguments to defend, in court, against their army of lawyers, using our massive financial advantage as an open movement against a bunch of multinational corporations with arms-dealing resources), or we need libre OA to include text mining as a strategic goal of the overall OA movement.
>>>> For me the choice is pretty clear. Expecting publishers to unbundle text mining from their licensing agreements and eliminate DRM is magical thinking, and the primary legal use arguments for text mining are only good as defenses *after* one has been sued, not prospective rights one can claim to prevent injunction. See lessig: fair use is the right to call a lawyer and nothing more.
>>>> We need explicit rights to text mine that obviate the language publishers use to prevent it. Leveraging copyright law to prevent them from adding those preventions is better strategy than planning to beat back the copyright cartel in court, IMHO.
>>>> IANAL, YMMV, etc.
>>>>> 2012/3/20 Heather Morrison<hgmorris at sfu.ca>:
>>>>>> A post on libre open access, copyright, patent law, and other intellectual
>>>>>> property matters. In brief, I argue that text and data mining materials on
>>>>>> the open web does not require special permissions. This has implications for
>>>>>> understanding what needs to happen to make libre open access a reality.
>>>>>> Comments welcome.
>>>> open-science mailing list
>>>> open-science at lists.okfn.org
>> john wilbanks
More information about the open-science