Why "green" open access self-archiving mandates must     come before "gold" open access publishing                        ...
What is OA?•   Free•   Immediate•   Permanent•   Online•   Access                 FNRS Bruxelles 28 septembre 2011
Open access to what?•   2.5 million articles•   Published yearly•   In 25,000 peer reviewed journals•   Across all scienti...
Why provide OA?• Because it maximizes research uses and  impact• Making it accessible to all users• Not just those who can...
How to provide OA?1. By publishing in any journal at all, but self-   archiving a copy free for all on the web2. By publis...
What are "green" and "                              " OA?• OA self-archiving is green• OA publishing is gold OA           ...
Why does green OA need to come          before gold OA?1. Because green OA can be provided by the   research community and...
"Gratis" OA vs "Libre" OA• Gratis OA means free online access• Libre OA means free online access plus certain  further re-...
Institutional repositories vs central repositories           Institutional vs funder mandates• Deposit institutionally, ha...
Parasitism? Catastrophe?• Subscriptions pay for publication• Authors contribute work for free• Peers review for free      ...
Copyright, Embargoes• Desirable but not necessary to retain  copyright• Immediategreen OA self-archiving can be  mandated ...
The optimalgreen OA self-archiving              mandate• Immediate Deposit/”Optional OA”• Deposit is sole means of submitt...
Leveraged transition1. Institutions and funders mandate ID/OA2. Universal green OA3. Subscriptions become unsustainable4. ...
No-fault peer review• Instead of charging for publication• Journals charge for each round go peer review• “No fault”• Reje...
Versions• Author’s peer-reviewed final draft has the  fewest publisher restrictions on it• Don’t over-reach for publisher’...
Why does green OA need to come          before gold OA?1. Because green OA can be provided by the   research community and...
EnablingOpenScholarship (EOS)          FNRS Bruxelles 28 septembre 2011
Mandated green OA levels (70%) vs self-selective (unmandated) green OA                             levels (20%)90%80%70%60...
Logistic regressionSet of 13 variables (plus one interaction) potentially influencing citation counts.                    ...
Exp(B)-1 values of logistic regression (N = 63,518)3                                                                      ...
Cumulative number of citations per article for Mandated and Self-     SelectedOAvsNOA articles (articles published in 2002...
Separate proportions for OAvsNOA articles at each citation count50%45%                          N = 63,518            N(OA...
Average cumulative number of citations per article as a function of article                age for articles published in 1...
Average cumulative number of citations per article as a function of article              age by field for articles publish...
Separate proportions for OA vsNOA articles at each citation count                      FNRS Bruxelles 28 septembre 2011
FNRS Bruxelles 28 septembre 2011
OA citation advantage for mandated vsunmandatedOA                                              4 institutions (N = 63,518)...
ROAR &ROARMAP                          Registry of Open Access Repositories                  Registry of Open Access Repos...
Scatter plot matrix :pairwise scatter plots of all variables           FNRS Bruxelles 28 septembre 2011
Leveraged transition1. Institutions and funders mandate ID/OA2. Universal green OA3. Subscriptions become unsustainable4. ...
Grasp what is within immediate reachDon’t keep over-reaching and getting next to nothing for yet another decade…      90% ...
EnablingOpenScholarship (EOS)          FNRS Bruxelles 28 septembre 2011
Thank youhttp://www.openscholarship.org    FNRS Bruxelles 28 septembre 2011
Upcoming SlideShare
Loading in …5
×

Stevan Harnad: On Designing Green Open Access Self-Archiving Mandates for Universities and Research Funders

547 views

Published on

Brussels Conference on University Open Access Policy

Published in: Technology, Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
547
On SlideShare
0
From Embeds
0
Number of Embeds
6
Actions
Shares
0
Downloads
7
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • - In order to display the OA advantage in a different way and show more graphically how it is distributed across the citation ranges, we calculated the proportions of OA articles at each citation count and compared them to those for NOA articles.- This Figure shows that as the citation count increases, the fraction of the total OA articles at each citation count overtakes the corresponding fraction of the total NOA articles.
  • In order to explore the evolution of OA advantage over time and by field, we plot the average cumulative number of citations per article as a function of article age, by field, for articles published in 2000.These figures show for each field, the average cumulative number of citations per article for OA and non-OA articles. For fields, as an article gets older, and its cumulative citations grow, its OA advantage grows too. However, this advantage differ in magnitude depending on the field; it is relatively high for Chemistery, ClininicalMedecine, Physics and Biomedical Research and is relatively weak in Mathematics, Biology, Social Science and Health.
  • These figures show that as the citation count increases, the fraction of the total OA articles at each citation count overtakes the corresponding fraction of the total NOA articles. The overtaking starts from 2 or 3 citations for Engineering, Mathematics and Social Science; from 4 or 5 citations and more for Biology; from 8 citations and more for Clinical Medecine; from 9 or 10 citations and more for Physics, Psychology, Chemistery and Earth Science; and from 11 or 12 citations and more for Biomedical research. However, for Health, the overtake is irregular for different levels of citation counts
  • Stevan Harnad: On Designing Green Open Access Self-Archiving Mandates for Universities and Research Funders

    1. 1. Why "green" open access self-archiving mandates must come before "gold" open access publishing or “Don’t over-reach: grasp what’s within immediate reach StevanHarnad Chaire de recherche du Canada en sciences cognitives Université de Québec à Montreal & Electronics and Computer Science University of Southampton FNRS Bruxelles 28 septembre 2011
    2. 2. What is OA?• Free• Immediate• Permanent• Online• Access FNRS Bruxelles 28 septembre 2011
    3. 3. Open access to what?• 2.5 million articles• Published yearly• In 25,000 peer reviewed journals• Across all scientific and scholarly disciplines• In all countries and languages• Don’t over-reach for open books, “open knowledge,” “open information” FNRS Bruxelles 28 septembre 2011
    4. 4. Why provide OA?• Because it maximizes research uses and impact• Making it accessible to all users• Not just those who can afford to subscribe to the journal in which the article was published FNRS Bruxelles 28 septembre 2011
    5. 5. How to provide OA?1. By publishing in any journal at all, but self- archiving a copy free for all on the web2. By publishing in a journal that makes its articles free for all on the web FNRS Bruxelles 28 septembre 2011
    6. 6. What are "green" and " " OA?• OA self-archiving is green• OA publishing is gold OA FNRS Bruxelles 28 septembre 2011
    7. 7. Why does green OA need to come before gold OA?1. Because green OA can be provided by the research community and gold OA can only be provided by the publisher community2. Because green OA can be mandated by research institutions and funders and gold OA cannot3. Because the money to pay for gold OA is still locked up in journal subscriptions4. Don’t over-reach for immediate gold OA FNRS Bruxelles 28 septembre 2011
    8. 8. "Gratis" OA vs "Libre" OA• Gratis OA means free online access• Libre OA means free online access plus certain further re-use rights• Don’t over-reach for libre OA FNRS Bruxelles 28 septembre 2011
    9. 9. Institutional repositories vs central repositories Institutional vs funder mandates• Deposit institutionally, harvest centrally• Don’t over-reach for direct central deposit FNRS Bruxelles 28 septembre 2011
    10. 10. Parasitism? Catastrophe?• Subscriptions pay for publication• Authors contribute work for free• Peers review for free FNRS Bruxelles 28 septembre 2011
    11. 11. Copyright, Embargoes• Desirable but not necessary to retain copyright• Immediategreen OA self-archiving can be mandated even if authors wish to comply with publisher OA embargoes• Don’t over-reach for copyright retention FNRS Bruxelles 28 septembre 2011
    12. 12. The optimalgreen OA self-archiving mandate• Immediate Deposit/”Optional OA”• Deposit is sole means of submitting research for institutional performance review (“Liège model”)• “email eprint request” button for embargoed deposits• Don’t over-reach for immediate-OA mandates FNRS Bruxelles 28 septembre 2011
    13. 13. Leveraged transition1. Institutions and funders mandate ID/OA2. Universal green OA3. Subscriptions become unsustainable4. Publishers downsize to peer review alone5. Offload access-provision and archiving on institutional repositories6. Journals convert to gold OA7. Gold (much lower) OA fees paid out of institutional windfall subscription cancelation savings FNRS Bruxelles 28 septembre 2011
    14. 14. No-fault peer review• Instead of charging for publication• Journals charge for each round go peer review• “No fault”• Rejected paper charges no longer wrapped into accepted paper fees• Journals not tempted to lower standards for more revenue FNRS Bruxelles 28 septembre 2011
    15. 15. Versions• Author’s peer-reviewed final draft has the fewest publisher restrictions on it• Don’t over-reach for publisher’s version-of- record FNRS Bruxelles 28 septembre 2011
    16. 16. Why does green OA need to come before gold OA?1. Because green OA can be provided by the research community and gold OA can only be provided by the publisher community2. Because green OA can be mandated by research institutions and funders and gold OA cannot3. Because the money to pay for gold OA is still locked up in journal subscriptions4. Don’t over-reach for immediate gold OA FNRS Bruxelles 28 septembre 2011
    17. 17. EnablingOpenScholarship (EOS) FNRS Bruxelles 28 septembre 2011
    18. 18. Mandated green OA levels (70%) vs self-selective (unmandated) green OA levels (20%)90%80%70%60% 200250% 2003 200440% 2005 200630% 200720% 2008 200910%0% N= 63,518 FNRS Bruxelles 28 septembre 2011
    19. 19. Logistic regressionSet of 13 variables (plus one interaction) potentially influencing citation counts. FNRS Bruxelles 28 septembre 2011
    20. 20. Exp(B)-1 values of logistic regression (N = 63,518)3 Age JIF Auth_N Ref_N Page_N2 Sci USA OA Age*OA1 M CERN South Minho Queens0 N=24,893 & 22,281 22,281& 5,796 22,281 & 4,798 22,281 & 2,834 zero/lo lo/med-lo lo/med-hi lo/hi 0 vs 1-4 cites 1-4 vs 5-9 cites 1-4 vs 10-19 cites 1-4 vs 20+ cites-1 FNRS Bruxelles 28 septembre 2011
    21. 21. Cumulative number of citations per article for Mandated and Self- SelectedOAvsNOA articles (articles published in 2002) 25 20 Cumulative count of citations 15 OM ØM 10 OS ØS 5 0 2002 2003 2004 2005 2006 2007 2008 2009 OM N = 897 ØM N = 493 OS N = 1,098 ØS N = 3,269 FNRS Bruxelles 28 septembre 2011
    22. 22. Separate proportions for OAvsNOA articles at each citation count50%45% N = 63,518 N(OA) = 44,497 N(NOA)= 19,02140%35%30%25% NOA20% OA15%10%5%0% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 >30 FNRS Bruxelles 28 septembre 2011
    23. 23. Average cumulative number of citations per article as a function of article age for articles published in 1998-2009 FNRS Bruxelles 28 septembre 2011
    24. 24. Average cumulative number of citations per article as a function of article age by field for articles published in 2000 FNRS Bruxelles 28 septembre 2011
    25. 25. Separate proportions for OA vsNOA articles at each citation count FNRS Bruxelles 28 septembre 2011
    26. 26. FNRS Bruxelles 28 septembre 2011
    27. 27. OA citation advantage for mandated vsunmandatedOA 4 institutions (N = 63,518) 0.5 2002 0.4 2003 2004 0.3 2005Citation ratios 2006 0.2 2007 2008 0.1 2009 0.0 O/Ø O S/Ø O M/Ø O S/Ø S O M/Ø S O M/Ø M O M/O S FNRS Bruxelles 28 septembre 2011
    28. 28. ROAR &ROARMAP Registry of Open Access Repositories Registry of Open Access Repository Archiving Mandates• AGE : age of institution’s deposit mandate (in months)• STRENGTH: strength of institution’s deposit mandate• deposits : total number of deposits in institutional repository• rate : rate of deposit (number of days per year with 10-99 deposits)• RANK : Webometrics rank of institution FNRS Bruxelles 28 septembre 2011
    29. 29. Scatter plot matrix :pairwise scatter plots of all variables FNRS Bruxelles 28 septembre 2011
    30. 30. Leveraged transition1. Institutions and funders mandate ID/OA2. Universal green OA3. Subscriptions become unsustainable4. Publishers downsize to peer review alone5. Offload access-provision and archiving on institutional repositories6. Journals convert to gold OA7. Gold (much lower) OA fees paid out of institutional windfall subscription cancelation savings FNRS Bruxelles 28 septembre 2011
    31. 31. Grasp what is within immediate reachDon’t keep over-reaching and getting next to nothing for yet another decade… 90% 80% 70% 60% 2002 50% 2003 2004 40% 2005 30% 2006 20% 2007 10% 2008 0% 2009 N= 63,518 FNRS Bruxelles 28 septembre 2011
    32. 32. EnablingOpenScholarship (EOS) FNRS Bruxelles 28 septembre 2011
    33. 33. Thank youhttp://www.openscholarship.org FNRS Bruxelles 28 septembre 2011

    ×