Gold OA article growth NOT 5.7 times total article growth by 2020

[Stevan Harnad <amsciforum@gmail.com>]

Heather Morrison's AmSci posting has some errors and 
misconceptions, so in the interests of OA there needs to be some 
clarification:

On 2010-12-20, at 9:06 PM, Heather Morrison wrote (in the 
American Scientist Open Access Forum):

> Dear Am Sci readers - this is not new information, but 
repeating as I think that this is worth highlighting.

The source information is not new, but it is worth discussing; 
some of the interpretation. however, is not only new but 
arbitrary and misleading.

> According to this chart by Thompson-Reuters

The chart is not by Thompson-Reuters/ISI, it is by Springer. (The 
figures are based on Thompson-Reuters/ISI-indexed 
articles/journals.)

http://users.ecs.soton.ac.uk/harnad/Temp/springergoldgrowth.png 
http://www.berlin8.org/userfiles/file/Berlin8_OA_Conference_PH_v1.pdf

> gold OA article growth (at 20%) could be 5.7 times higher than 
total article growth (at 3.5%) by 2020.

(1) 20/3.5 = 5.7 now (and always), not particularly in 2020.

(2) The Gold OA article growth of 20% per year means 20% *of Gold 
OA articles * per year, not 20% of *all articles* per year.

(3) The overall article growth rate of 3.5% per year is based on 
*all* articles per year.

(4) Hence the tautology that the "20% Gold article growth rate is 
5.7 times the 3.5% total article growth rate" is just numerology 
-- it is virtually meaningless.

(5) The Springer graph shows that in 2010, the percentage of Gold 
OA articles is about 8% out of all (ISI-indexed) articles, and 
that at the current growth rate (consisting of a 20% Gold OA 
increase per year relative to the prior year's Gold OA articles) 
the percentage of Gold OA articles in 2020 will be about 27% out 
of all (ISI) articles. That "averages" to a yearly Gold OA 
increase of about 2% of all (ISI) articles, compounding across 
the ten years from 2010 to 2020.

(6) So the point of the Springer graph is simply that at the 
current Gold OA growth rate, in 10 years the overall percentage 
of Gold OA will have increased from its current 8% Gold OA out of 
all (ISI) articles published in 2010 to 27% Gold OA out of all 
(ISI) articles published in 2020.

(7) That rate is not at all heartening for those who are seeking 
100% OA, now (and have been seeking it for nearly two decades 
already).

(8) This is why portraying this rate as "5.7 times total article 
growth by 2020" is arbitrary numerology wrapped in premature (and 
groundless) triumphalism.

(My own interpretation of the Springer graph is that it provides 
further evidence -- if further evidence was needed -- that the 
fastest and surest road to 100% OA is for institutions and 
funders to mandate Green OA self-archiving, now, and not to sit 
around waiting for Gold OA (or applauding its growth rate) for 
yet another decade. Unmandated Green OA happens to be growing 
faster than Gold OA among ISI-indexed journal articles -- but 
that's nothing to take heart from either: 
http://openaccess.eprints.org/index.php?/archives/760-guid.html
What is urgently needed is institutional and funder Green OA 
mandates, worldwide. Mandate adoption, too, is growing, but its 
growth rate is likewise nothing to take heart from:
http://www.openscholarship.org/upload/docs/image/jpeg/2010-10/oct_22.jpg
The only difference is that *there is something we can do about 
that*: It's 100% in the research community's hands to mandate and 
provide Green OA, whereas providing Gold OA is in the hands of 
the publishing community -- and the potential money to pay for a 
100% transition to Gold OA is currently tied up in institutional 
journal subscriptions that are uncancellable -- until Green OA is 
100% mandated. )

We know from my own groups' work using a robot to trawl the Web 
looking for OA articles, and from Bo-Christer Bjork's periodic 
measurements of the number of OA articles, that the proportion of 
articles that are OA, expressed as a percentage of the total 
number of articles published, has grown very little over the last 
five years - a couple of percentage points at best.

There is much work to be done persuading people to take the right 
approaches that will increase this figure. That is where the 
effort should be going now.

The best source of reliable strategic and policy information and 
guidance for institutional policy-makers worldwide is 
EnablingOpenScholarship (EOS) http://www.openscholarship.org

Further practical information is available from OASIS 
http://www.openoasis.org . Advice on advocacy is also available 
from SPARC http://www.arl.org/sparc/advocacy/campus/

Stevan Harnad
American Scientist Open Access Forum
http://amsci-forum.amsci.org/archives/American-Scientist-Open-Access-Forum.html