SUMMARY: Elsevier has since at least 2004 given every single author of every single article published in every single one of its c. 2000 journals the green light to self-archive it if they wish to make their article Open Access (OA). All Elsevier authors can, with Elsevier's blessing, deposit their peer-reviewed final drafts (postprints) immediately upon acceptance for publication (no embargo -- and their pre-refereeing preprints even earlier, if they wish) in their own OA Institutional Repositories (IRs).
No author desirous of OA can ask for more from their publisher; yet only about 15% of authors are as yet self-archiving spontaneously, even with their publisher's green light. And authors (and librarians, and even some OA advocates) do still keep asking for more:
Elsevier does not endorse the self-archiving of its proprietary XML/PDF, nor does it endorse self-archiving in a 3rd-party central repository rather than the author's own IR. For some reason, there are complaints about this (even though full-blooded OA does not at all require free access to the publisher's proprietary draft, and even though the [brief] era of central self-archiving is already over, replaced by distributed OAI-interoperable self-archiving in the IR of the research-provider, with central harvesting into "virtual" archives, if and when desired). And the complaints, I'll warrant, are not coming from the authors who are self-archiving, but from those who are not.
There is only one cure for this fruitless inertia: Authors' institutions and funders need to mandate self-archiving, so authors (and users) can stop complaining about needing or wanting OA, and start doing something about it, for their own (and their institutions') good, as well as the good of research and the public that funds it.
(We don't need more solemn signings of pious declarations of abstract principle; we don't even need mere sanguine signatures from provosts endorsing national legislative proposals to mandate concrete self-archiving: Charity begins at home, and provosts should also sign self-archiving mandates for their own institutions, rather than just signing endorsements and then waiting passively for their governments to act: with local mandates in place it will matter less whether the national legislation even manages to get passed!)
On LIS-ELIB, Mike McGrath [MM], British Library, wrote:
MM: "The posting from Stevan [Harnad] on self archiving below sounds good. I extract one element of it below and highlight the bit I am querying":
'There is a growing national and international movement for authors of peer-reviewed journal articles to self-archive their work in repositories that are openly accessible.'
This quote was actually from the
OhioLINK recommendation, not from me (though I of course fully concur!). Read on:
MM: "In Library Connect Vol 2 No 2 (2004) Elsevier state:'The posting cannot be for commercial purposes (such as systematic distribution or creating links for customers to articles) and it is not permitted to post to Web sites outside of their institution ....Similarly posting of the journal's PDF or HTML files is not permitted.'
MM: "I commented that 'Some have suggested that the implication of this is that one would need to search each institution's web site separately in order to locate relevant material - clearly not often a practical option. (Interlending and Document Supply Vol 33 No 1 page 44). It seems to be a murky but important area. I suspect that Elsevier would not be keen to push the issue as the whole area of copyright assignment is itself pretty murky."
Elsevier is perfectly right in every single one of its stipulations: each stipulation is reasonable, and perfectly compatible with providing 100% OA to peer-reviewed research output for all would-be users, webwide, including visibility and findability via OAI search engines or google, without the searcher's needing to know the deposit's locus in advance or to go directly to the site where it is self-archived to seek it.
("Interlending," by the way, is
completely obsolete for the Open Access [OA] corpus!)
Elsevier: The posting cannot be for commercial purposes
This is perfectly fine, and justified. Self-archiving is done by authors in their own
Institutional Repositories (IRs), which are not commercial, and do not sell access, but provide it, to their own research output only, free for all, webwide. That is the meaning and purpose of OA, nothing else.
Elsevier: such as systematic distribution
Again, a university's own IR is not a database selling any systematic redistribution of a journal's contents. The IR is making only the author's own final drafts -- of those individual papers that were written by the institution's own employees -- freely available to would-be users the world over who cannot afford access to the official published version.
The metadata are of course also harvested by google, OAIster and other search engines, as are all free contents on the web.
The full-texts are also harvested, inverted and cached by google, as are all free contents on the web. That all comes with the WWW territory.
Elsevier: or creating links for customers to articles
Neither the author nor the author's institution is creating any links to articles for "customers." (Indeed, the self-archived draft's metadata include, where available, links to the official version at the publisher's website.)
If there are commercial harvesters on the web, creating and selling links from X to Y, that has nothing whatsoever to do with the author or his institution. Those who may wish to go after such commercial services legally can go ahead and try to do so (but it would be rather difficult to pursue, because linking too comes with the WWW territory).
The point is irrelevant to author self-archiving in particular.
Elsevier: it is not permitted to post to Web sites outside of their institution
The optimal way to self-archive is to deposit ("post") the author's own final draft and metadata in the author's own Institutional Repository (IR). Then central "virtual" repositories need merely harvest the metadata (or full-text, if desired). This is one of the main reasons the
OAI metadata harvesting protocol was created and why the
IR software makes the IRs OAI-compliant.
The era of direct depositing in
central repositories such as
Arxiv or
PubMed Central is already obsolete, even though well-meaning people still don't understand or realise it (and suboptimal interim practices do persist in the few fields where they have become habitual in the past decade and a half). There is not only no need to deposit centrally any longer in the OAI-interoperable era of distributed OA IRs and central harvesting, but depositing centrally is in fact a needless retardant and complication in the process of systematically providing OA to all research output through self-archiving. It is a relic of the obsolete paper-era notion that documents all need to be "in one place" in order to be accessed, curated, classified and searched. The
optimal system in the OA/OAI age is for the research-provider to self-archive its own output, and for harvesting services to take care of distributed search. Distributed local institutional self-archiving, for all of its own research subjects, is also the system that will most simply and naturally cover all of research space worldwide. An overlapping (and arbitrary) system of central subject-based repositories to deposit into directly would only create needless confusion, compliance-monitoring problems, and gaps as well as redundancies.
Elsevier is hence actually helping OA progress by opposing central self-archiving: By self-archiving only in their own institution's IR, authors ensure that they are not redistributing the journal's contents to a potential free-rider or rival publisher. Reaching 100% OA worldwide is in no way compromised thereby. And universities and funders who adopt self-archiving policies are thereby encouraged to adopt rational policies, that will scale up to systematically to cover all of research output, by specifically mandating that the deposit be done in the author/fundee's
own institutional IR rather than a site outside of their institutions.
OAI harvesting will take care of all the rest.
Elsevier: posting of the journal's PDF or HTML files is not permitted.
This too is perfectly reasonable: The publisher's XML and PDF are his own proprietary product and OA by no means requires that that version be the one that is deposited in the author's IR: The author's own peer-reviewed final draft (postprint) is the one that can and should be deposited. A link should also point to the publisher's official version.
MM: "Some have suggested that the implication of this is that one would need to search each institution's web site separately in order to locate relevant material - clearly not often a practical option".
And clearly completely nonsense, being based on a lack of understanding of the nature of the web, of web search, or OAI-interoperability.
All of this will become clearer with time. At the moment, though, misunderstandings like this, freely propagated from one misunderstander to another, are among the remaining obstacles in the path of OA. Elsevier is certainly not the source or cause of these misconstruals of the online medium!
MM: "It seems to be a murky but important area. I suspect that Elsevier would not be keen to push the issue as the whole area of copyright assignment is itself pretty murky."
I am not sure what this comment means. It is not for Elsevier to push self-archiving: It is enough that they do not attempt to stand in its way. It is for researchers' institutions and funders to push self-archiving, by
mandating it.
The murkiness in all this is in the technical and practical grasp of what it means to self-archive an article in an OAI-compliant OA IR. But that murkiness will all clear up as more and more researchers do it, and more and more universities and funders mandate it.
And copyright assignment has next to nothing to do with it any more (if it ever did, which I doubt!): Ninety-four percent of journals -- including all of Elsevier's c. 2000 journals -- have already given their green light to author self-archiving. It's now up to the authors (or their designees) to go ahead and do it and their institutions and funders to ensure that they (or their designees) do it, for their own good, by mandating it. It is always good to retain copyright where desired and possible, but it is most definitely
not a prerequisite -- either for self-archiving, or for mandating self-archiving, or for reaching 100% OA -- to retain copyright.
Pertinent Prior American Scientist Open Access Forum Topic Threads;
Milestones Along the Road To OA Mandates:
1998:
"Elsevier Science Policy on Public Web Archiving Needs Re-Thinking"
1999:
"Central vs. Distributed Archives"
2002:
"Evolving Publisher Copyright Policies On Self-Archiving"
2003:
"Draft Policy for Self-Archiving University Research Output"
"A Keystroke Koan For Our Open Access Times"
"Central versus institutional self-archiving"
"What Provosts Need to Mandate"
"Written evidence for UK Select Committee's Inquiry into Scientific Publications"
"Recommendations for UK Open-Access Provision Policy"
2004:
"University policy mandating self-archiving of research output"
"Meeting: National Policies on Open Access Provision for University Research Output"
"Elsevier Gives Authors Green Light for Open Access Self-Archiving"
"Mandating OA around the corner?"
"Implementing the US/UK recommendation to mandate OA Self-Archiving"
"A Simple Way to Optimize the NIH Public Access Policy"
2005:
"Please Don't Copy-Cat Clone NIH-12 Non-OA Policy!"
"International meeting could create a worldwide policy for Open Access"
"How to Word Institutional Self-Archiving Policy"
"Comparing the Wellcome OA Policy and the RCUK (draft) Policy"
"New international study demonstrates worldwide readiness for Open Access mandate"
"Maximising the Return on UK's Public Investment in Research"
"DASER 2 IR Meeting and NIH Public Access Policy"
"Mandated OA for publicly-funded medical research in the US"
2006:
"Mandatory policy report" (2)
"The U.S. CURES Act would mandate OA"
"Generic Rationale and Model for University Open Access Mandate"
"Optimizing MIT's Open Access Policy"
"How to Counter All Opposition to the FRPAA Self-Archiving Mandate"
Stevan Harnad
American Scientist Open Access Forum
