[Open-access] Letter to publishers - URGENT

Mon Mar 5 14:14:27 UTC 2012

I humbly offer the following text as a possible revision in the light
of this morning's discussion. I hope it is a helpful contribution to
the drafting process.

Briefly, I have substituted "fundamental" for absolute,  replaced
"arbitrarily large amounts" with "without limitation", and "whenever
they wish" by "entire current" (which means the old footnote 7 becomes
the new note 4), and inserted a footnote [2] on being nice with the
server (thus renumbering successive footnotes).

I am somewhat aghast at my own temerity in so-doing, and will not be
offended in the slightest should anyone decide to hit the revert
button. The new introductory paragraph is entirely my invention and
could and should be improved by others more knowledgeable than I.

Regards to all,

D.

***
Dear [NamedPersonAtOrganisationX],

As you know, progress in networked computing currently offers
unparalleled opportunities for progress in many areas through the use
of textmining algorithms on the existing scholarly literature.
[[Required: 1 sentence on potential benefit for humanity. . . [Such
techniques could [???speed drug discovery, save scholarly time. . .
thus saving lives and generating economic benefits . . . add here] ]]
We are writing to several major scholarly publishers to ascertain
their current practice in this domain, and plan to collate the replies
we receive as part of our submission to the Hargreaves enquiry.

We assert that subscribers [1] have a fundamental right to use
machines [2] to extract facts [3] from the entire current scholarly
literature [4] in all forms [5] without limitation [6] and publish the
output as CC0 [7]. This right extends to creating CC0 indexes [8] and
summaries [9]. Publishers have a responsibility to make this process
simple and reliable for all subscribers [10].

We would appreciate a clear statement of your support for this
important scholarly principle, and for specific details of how
[OrganisationName] is furthering scientific progress in this
endeavour.

You are also cordially invited to join in the discussion of these
issues on the mailing lists and blogs of the okfn. [[provide URLs
here]]

Thank-you for your response.

Yours sincerely,
[PMR on behalf of . . . ?]

Notes:
1. A subscription to a paper artefact allows the reader to extract
facts, create indexes and makes summaries without requiring
permission. We are asking for this right for legitimate subscribers to
electronic journals and other scholarly artifacts.
2. Subject to best network practice in the prevention of server
overload by intensive automated requests, i.e. in conformance to any
reasonable API service limit.
3. Facts are not copyright.
4. Scholars should be able to extract information as soon as it
appears and to delve backwards as far as they wish. They should not be
dependent on publishers providing dumps, though this may be a useful
additional option.
5. Scholarship expresses facts [at least] as text, numbers, tables,
diagrams, images, spoken discourse and video. Machines can reliably
extract facts from all of these
6. Limitations on volume are unacceptable, just as they are for human
extracters.
7. CC0 (Creative Commons Public Domain Dedication) means that the
information can be freely used with no restrictions (See
http://creativecommons.org/publicdomain/zero/1.0/).
8. There is a traditional right to index the literature. Many indexes
are factual, others involve judgment/classification which can now be
provided by machines.
9. There is a traditional right to create summaries of works and publish them.
10. No publisher should install robots to block legitimate use by
subscribers. No publisher should insert clauses in contracts which
militate against the subscribers' rights. No publisher should require
individuals to ask for permissions or justify their actions.

-- 
Douglas Carnall
dougie.carnall at gmail.com

http://cabinetbeezer.info

Traduction vers l'anglais
Rédaction de textes en anglais
Coaching pour présentations en anglais