Björn Brembs, Freie Universität Berlin

How much does it cost
to subscribe to 24,000 journals?

Serial & Monograph Costs, 1986-2002
North American research libraries ARL Statistics Ray English

<ul><li>Chemistry $3,254 </li></ul><ul><li>Physics 1,756 </li></ul><ul><li>Astronomy
2,850 </li></ul><ul><li>Engineering 1,548 </li></ul><ul><li>Geology 1,724 </li></ul><ul><li>Biology 1,323 </li></ul><ul><li>Math & Computer Sci 1,278 </li></ul><ul><li>Zoology 1,259 </li></ul><ul><li>Botany 1,238 </li></ul><ul><li>Health Sciences 1,132 </li></ul><ul><li>Library Journal Periodical Price Survey, April 2006 </li></ul>Ray English

<ul><li>People produce your product for
you </li></ul><ul><li>They check it for quality </li></ul><ul><li>They’re even kind enough to give you their intellectual property </li></ul><ul><li>You polish it up and distribute it </li></ul><ul><li>And you charge those same people handsomely to make their product available back to them </li></ul><ul><li>They think they must have your product, even though they created it, so you’re free to raise prices </li></ul>

What a magnificent ship! What
makes it go? Cartoon by Rowland B. Wilson

<ul><li>Request increased budgets </li></ul><ul><li>Cut subscriptions
</li></ul><ul><li>Reduce monograph purchases </li></ul><ul><li>Cut subscriptions and reduce monographs </li></ul><ul><li>Collective purchase of electronic journals </li></ul><ul><li>Rely on document delivery or ILL </li></ul>Ray English

Ray English

David Nicholas

<ul><li>Substantial portion is </li></ul><ul><ul><li>funded by
taxpayers </li></ul></ul><ul><ul><li>supported publicly </li></ul></ul><ul><ul><li>created in non-profit sector </li></ul></ul><ul><li>Journal literature is freely given away by authors </li></ul><ul><li>But journal publishing is largely under corporate control </li></ul><ul><li>A public good in private hands </li></ul>Ray English

<ul><li>“ By ‘open access’ to
this literature [i.e. peer-reviewed journal articles], we mean its free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself.” (Budapest Open Access Initiative, www.soros.org/openaccess/ .) </li></ul>

<ul><li>First route : authors deposit
copy of pre-print or post-print in an “institutional repository” or other open web-site </li></ul><ul><ul><li>Over 600 open repositories already established world-wide </li></ul></ul><ul><li>Second route : authors publish in peer-reviewed journals funded by publication charges rather than by library subscriptions </li></ul><ul><ul><li>Over 1400 peer-reviewed open access journals now listed in the Lund Directory of Open Access Journals www.doaj.org </li></ul></ul>ROAR DOAJ RoMEO

<ul><li>Open access enables more people
to read research reports </li></ul><ul><li>More readers lead to greater use and exploitation of research results (including higher numbers of citations), facilitating the funding of further research </li></ul><ul><li>Greater use of research results leads to more public awareness of the value of scientific research </li></ul><ul><li>More public awareness leads to a higher political profile for academic research </li></ul><ul><li>Repositories help university administrators to keep a record of university research reports </li></ul>Frederick Friend

(“Online or invisible?”, Steve Lawrence
in Nature 411, Number 6837, p. 521, 2001.)

<ul><li>Bits and bytes 1010100101000001101010 (not
paper) </li></ul><ul><li>In pervasive cyberspace (not physical space) </li></ul><ul><li>Databases and/or Web identified by URIs: (not buildings) </li></ul><ul><li>Cost of distribution fallen by orders of magnitude </li></ul><ul><li>Read and indexed by machines like Googlebot et al (not just humans) </li></ul><ul><li>Increasingly public, available to everyone via Open-Access publishing (less private, less restrictive copyright) </li></ul><ul><li>Everything is great? </li></ul>Alexander Griekspoor www.mekentosj.com Duncan Hull

www.scopus.com www.pubmed.gov http://ukpmc.ac.uk isiknowledge.com scholar.google.com
Duncan Hull

<ul><li>Isolation </li></ul><ul><ul><li>each discipline has its
own data silo </li></ul></ul><ul><li>Impersonal and unsociable </li></ul><ul><ul><li>“ who the hell are you”? </li></ul></ul><ul><ul><li>Where are “my” papers? (authored by me, or of interest to me) </li></ul></ul><ul><ul><li>What are my friends and colleagues reading? </li></ul></ul><ul><ul><li>What are the experts reading? What is popular this week / month / year ? </li></ul></ul><ul><li>“ Cold”: Identity of publications and authors is inadequate </li></ul><ul><li>Obsolete models of publication, not everything fits publication-sized holes </li></ul><ul><ul><li>Micro-attribution </li></ul></ul><ul><ul><li>Mega-attribution </li></ul></ul><ul><ul><li>Digital contributions (databases, software, wikis/blogs?) </li></ul></ul>Duncan Hull

How can I find anything?

<ul><li>http://pubmed.gov/18974831 </li></ul><ul><li>http://www.ncbi.nlm.nih.gov/pubmed/18974831 </li></ul><ul><li>http://ukpmc.ac.uk/articlerender.cgi?accid=pmcA2568856 </li></ul><ul><li>http://ukpmc.ac.uk/picrender.cgi?artid=1687256&blobtype=pdf </li></ul><ul><li>http://www.ploscompbiol.org/article/info%3Adoi%2F10.1371%2Fjournal.pcbi.1000204
</li></ul><ul><li>http://www.dbkgroup.org/Papers/hull_defrost_ploscb08.pdf </li></ul><ul><li>http://dx.doi.org/10.1371/journal.pcbi.1000204 </li></ul><ul><li>One paper, many URIs. Disambiguation algorithms rely on getting metadata for each </li></ul><ul><ul><li>Big problem for libraries is these redundant duplicates </li></ul></ul><ul><li>Matching can be done by Digital Object Identifier (DOI) and PubMed ID (PMID); </li></ul><ul><ul><li>these are frequently absent < 5% (Kevin Emamy, citeulike) </li></ul></ul>Duncan Hull

Or filter failure?

<ul><li>Publish or perish: number of
publications </li></ul><ul><li>Where are you published? </li></ul><ul><ul><li>~24,000 scholarly journals (~6,000 with IF) </li></ul></ul><ul><ul><li>~2.5 million publications/year </li></ul></ul><ul><ul><li>60-300 applicants per tenure-track position </li></ul></ul><ul><li>Reading enough publications is impossible! </li></ul>

<ul><li>Thomson Reuters: Impact Factor </li></ul><ul><li>ScImago
JournalRank </li></ul><ul><li>Eigenfactor </li></ul>

Publikationstätigkeit (vollständige Publikationsliste, darunter Originalarbeiten
als Erstautor/in, Seniorautor/in, Impact-Punkte insgesamt und in den letzten 5 Jahren, darunter jeweils gesondert ausgewiesen als Erst- und Seniorautor/in, persönlicher Scientific Citations Index (SCI, h-Index nach Web of Science) über alle Arbeiten) Publications: Complete list of publications, including original research papers as first author, senior author, impact points total and in the last 5 years, with marked first and last-authorships, personal Scientific Citations Index (SCI, h-Index according to web of science) for all publications.

As much as some may
want metrics to go away entirely, that Genie is already out of the bottle and won‘t go back in.

Albert Einstein

<ul><li>Who knows what the IF
is? </li></ul><ul><li>Who uses the IF to pick a journal (rate a candidate, etc.)? </li></ul><ul><li>Who knows how the IF is calculated and from what data? </li></ul>

<ul><li>50,000 employees </li></ul><ul><li>US$600million profit/quarter </li></ul><ul><li>Thomson
family owns 53% </li></ul><ul><li>$30,000-120,000/year subscription rates </li></ul>

<ul><li>Introduced in 1960’s by Eugene
Garfield: ISI </li></ul>2006 and 2007 2008 IF=5 Articles published in 06/07 were cited an average of 5 times in 08. citations articles

Journal X IF 2008= All
citations from Thomsons Reuters journals in 2008 to papers in journal X Number of citable articles published in journal X in 2006/7

<ul><li>Negotiable </li></ul><ul><li>Irreproducible </li></ul><ul><li>Not mathematically sound
</li></ul>

<ul><li>PLoS Medicine, IF 2-11 (8.4)
</li></ul><ul><li>Current Biology IF from 7 to 11 in 2003 </li></ul><ul><ul><li>Bought by Cell press in 2001 </li></ul></ul>

<ul><li>Rockefeller University Press buys their
data from TR </li></ul><ul><li>Up to 19% deviation from published records </li></ul><ul><li>Second dataset still not correct </li></ul>

<ul><li>Left-skewed distributions </li></ul><ul><li>Weak correlation of
individual article citation rate with journal IF </li></ul>

www.mendeley.com www.zotero.org connotea.org www.mekentosj.com hubmed.org
Re-couple metadata that has be de-coupled from data www.2collab.com www.refworks.com “ iTunes for PDF files”

<ul><li>Where the work is published
</li></ul><ul><ul><li>JournalRank </li></ul></ul><ul><li>Citations </li></ul><ul><ul><li>scholarly, hyperlinks, social bookmarks </li></ul></ul><ul><li>Web usage </li></ul><ul><ul><li>Publisher platform; 3rd party locations </li></ul></ul><ul><li>Expert ratings </li></ul><ul><ul><li>F1000; Peer Reviewers; Ed Boards etc </li></ul></ul><ul><li>Community rating & commenting </li></ul><ul><ul><li>Digging; Commenting; Rating etc </li></ul></ul>Peter Binfield

<ul><li>Media/blog coverage </li></ul><ul><ul><li>Which sources are
considered the most important? </li></ul></ul><ul><li>Policy development? </li></ul><ul><li>Who published it? </li></ul><ul><ul><li>And where do they work? What did they publish before? How ‘impactful’ are they? </li></ul></ul><ul><li>Who is talking about it? </li></ul><ul><ul><li>And what authority do they have? </li></ul></ul><ul><li>Who is citing it ? </li></ul><ul><ul><li>And what authority do they have? </li></ul></ul>Peter Binfield

“ Your article was published
in a journal with an Impact Factor of X” Peter Binfield

<ul><li>Your article: </li></ul><ul><li>Received X citations
(de-duped from Google Scholar, Scopus, and Web of Science) </li></ul><ul><li>It was viewed X times, placing it in the top Y% of all articles in this journal/community </li></ul><ul><li>It received X Comments </li></ul><ul><li>It was bookmarked X times in Social Bookmarking sites. </li></ul><ul><li>Experts in your community rated it as X, Y, Z. </li></ul><ul><li>It was discussed on X ‘respected’ blogs </li></ul><ul><li>It appeared in X, Y, Z International News media </li></ul>Peter Binfield

<ul><li>Prestige of the publisher (if
any). </li></ul><ul><li>Prestige of commenters/users </li></ul><ul><li>Percentage of a document quoted in other documents. </li></ul><ul><li>Raw links to the document. </li></ul><ul><li>Valued links </li></ul><ul><li>Obvious attention: discussions in blogspace, comments etc </li></ul><ul><li>Language in comments: positive, negative, clarified, reinterpreted. </li></ul><ul><li>Quality of author's institutional affiliation(s). </li></ul><ul><li>Significance of author's other work. </li></ul><ul><li>Amount of author's participation in other valued projects. </li></ul><ul><li>Reference network: the significance of all the texts cited. </li></ul><ul><li>Length of time a document has existed. </li></ul><ul><li>Inclusion of a document in lists of &quot;best of,&quot; in syllabi, indexes, etc </li></ul><ul><li>Types of tags assigned to it </li></ul><ul><li>Authority of the taggers, the authority of the tagging system. </li></ul>Peter Binfield

Discuss…

<ul><li>No more publishers – libraries
archive everything </li></ul><ul><li>Single semantic, decentralized database of literature and data </li></ul><ul><li>Peer-review done by independent body </li></ul><ul><li>Link typology for text/text, data/data and text/data links („citations“) </li></ul><ul><li>Semantic Text/Datamining </li></ul><ul><li>All the metrics you (don‘t) want (but need) </li></ul><ul><li>Tagging, bookmarking, etc. </li></ul><ul><li>Unique contributor IDs with attribution/reputation system (teaching, reviewing, curating, blogging, etc.) </li></ul><ul><li>Technically feasible today (almost) </li></ul>