Begin forwarded message:
> From: Les A Carr <lac -- ecs.soton.ac.uk>
> Date: January 3, 2012 11:24:40 AM EST
> Subject: Re: Mendeley users
>
> ...I did do a couple of analyses of Mendeley effectiveness for OA about 12
> months ago.
> You can see my writeups here:
>
> http://repositoryman.blogspot.com/2011/06/mendeley-download-vs-upload-growth.html
> and
> http://repositoryman.blogspot.com/2011/06/mendeley-measuring-oa-rates.html
>
> ...Mendley is not any more successful in providing OA than repositories
> are and in fact it is very disappointing in the number of open PDFs that it
> has created...
>
> I have detailed spreadsheets and can redo the analyses if they are
> interesting/useful.
> --
> les
>
>
>
>
> On 24 Dec 2011, at 04:06, Stevan Harnad wrote:
>
> Arthur,
>
> Thanks for the data about Mendeley. I have a few questions about things I
> don't understand;
>
> 1.4 million people have downloaded and installed Mendeley, from 32,000
> institutions. This means little, I think, though giving us another estimate
> of the number of research institutions. That there are 122,000 groups totally
> dwarfs the number of institutional repositories
>
> As far as I can tell, groups are subsets of people who have agreed to share
> their reference lists. This is the power set of the number of institutions,
> indeed of the number of individuals, so it is bound to be huge, given the
> number of k/N subgroupings are combinatorially possible!
>
> and 143 million articles is not to be sneezed at, though there is doubtless a
> lot of replicates.
>
> But what are they. Assuming they are indeed full-texts and not just
> bibliographic citations (of writings by Aristotle, for example) what we need
> to know is what percentage of total papers published in (say) 2010 they
> represent. (About 20% is the figure to beat. That's the spontaneous
> UNMANDATED self-archiving rate -- including both websites and repositories.
> Seventy percent would be the figure to match or beat for MANDATED
> self-archiving.
>
> To pursue the analysis a bit further, if the eight ID/OA mandated
> institutions have about 1000 academics each on overage, that?s 8,000 authors.
> The 200 dubious mandates contribute 1000 x 200 x 10% = 20,000 people, making
> 28,000 people contributing.
>
> I'm not quite sure why we are counting these authors from mandating
> institutions. The percentage of Mendel-OA papers per year is one benchmark.
> Another is the percentage of Mendel-OA papers at a given UNMANDATED
> institution.
>
> Do you see why I am now interested in the ?social media? pull rather than the
> IR push?
>
> Not yet, I'm afraid. What are needed in order to judge how well Mendel-OA is
> doing relative to gold OA, unmandated green OA ad mandated green OA is the
> comparative percentages I mention above.
>
> This perhaps begins to answer whether I can I justify my excitement?
>
> Mendeley is an exciting, useful tool -- but whether it is accelerating OA is
> not at all clear from the data you cite. (And I'm not clear on how how
> Mendeley stocks up: Only through authors importing, or does it also harvest
> from what's already on the web (i.e., unmandated green OA>?
>
> Chrs, S
>
>
>
