The US Department of Energy (DOE) has developed a prototype public access system that is designed to go to publisher’s websites. The system is presently called “Public Access Gateway for Energy and Science” or PAGES. A linkage between PAGES and CHORUS is planned, assuming CHORUS develops as presently described. However DOE has not yet committed to using PAGES. As the OSTP memo process continues, everything is in flux at this time.
The PAGES system is being developed by the Office of Scientific and Technical Information (OSTI). Among the federal agencies, OSTI is the expert in building distributed search systems that federate large numbers of websites and databases, which is what PAGES will have to do. For example, OSTI’s E-Print Network product searches about 60 repository databases and 36,000 author and departmental websites, covering about 5 million articles.
They also built and operate Science.gov which federates all of the federal report databases plus about 2,000 agency websites, with an estimated 200 million pages of content. Note that Science.gov is a multi-agency collaboration. Then they helped weave Science.gov into WorldWideScience.org, an international collaboration presently including about 90 national and international document databases with an estimated 400 million pages of content.
PAGES has been in development for several years. The PAGES concept of using publisher websites rather than a federal repository was developed during the course of the OSTP interagency task force on open access, which OSTI director Walt Warnick co-chaired. Working with several publishers, PAGES achieved proof of concept some time ago and a prototype is working today.
PAGES uses what might be called a tiered approach. If the article in question is published openly within the specified agency timeframe, which OSTI calls the “administrative interval”, then PAGES sends the user to the publisher’s website for it. If not, then the author is asked if it is in a repository, and if so PAGES sends the user there for the article. In the last resort, where there is no public source for the article, the author is required to submit their accepted manuscript to an agency repository, which will be created.
While no decision has been made to use the PAGES system, it clearly has the potential to be a multi-agency portal along the lines of Science.gov or even as the public access component of Science.gov. This would solve the monster problem of multiple agency public access systems I flagged when the OSTP memo first came out.
There is also the possibility of linking published articles with agency research project reports, which often contain additional information about the project results being reported in the journal. For example DOE estimates that its average project report, which it already publishes, is about 60 pages long, far longer than the average journal article.
The PAGES and CHORUS design teams are aware of each other’s work. While integrating the two systems is logical, it depends in part on CHORUS meeting federal needs, which are still evolving. It remains to be seen where this is all going, but PAGES increases the possibility that the OSTP mandate can be met without creating a bunch of separate public access systems or redundant repositories. By the same token integration of PAGES with CHORUS creates obviously beneficial efficiencies.
Discussion
13 Thoughts on "Meet PAGES — DOE’s Prototype Public Access System"
Do you imagine an eventual connection between these developments and the desire of many to see the emergence of national digital public libraries that could be federated into an international digital library? For example David Rothman’s conception of the Digital Public Library of America (DPLA), see: http://chronicle.com/article/Its-Time-for-a-National/126489/.
There is a parallel of sorts but I know of no policy push in the direction you describe. We already have several national libraries in the executive branch of the Federal government, including Medicine (huge), Agriculture (medium) and Energy (small). They will probably be handling these public access systems, and Medicine already does in PubMed Central. They have a lot of digital content, as does the Library of Congress, but I have no idea how this fits into the vision you are referring to. My focus is the scientific literature, not books. I would like to see all the agency public access systems federated and PAGES is clearly a candidate for that role.
Calling repositories “redundant” is a bit harsh, after all in this model they are vital.
In this model repositories are only needed when the article is not publicly available on the publisher’s website. In principle this need could be zero, if all publishers agree to the federal embargo requirements. Redundant refers to creating repositories that duplicate the publisher’s accessible holdings.
It should probably be noted that CHORUS includes the use of LOCKSS (http://www.lockss.org/) and CLOCKSS (http://www.clockss.org/clockss/Home) to serve the archival purposes of repositories.
Yes but the contents therein will not be publicly accessible unless the publisher version becomes unavailable for some reason, so there is no redundancy of access. Also as I understand it PAGES will probably not use that feature of CHORUS, rather they will have their own dark archive or some such.
“In principle this need could be zero, if all publishers agree to the federal embargo requirements.”
Embargo periods still exclude readers (i.e. developing countries and researchers at smaller institutions). Your so called “redundant copies” could at lest provide timely access to (a version of the) research to everyone.
I’m not going to turn this into a fight about who should pay for what, I just want to point out that “embargoes” are not a one-size fits all solution.
You have missed my meaning. I am referring solely to the use of redundant repositories in Federal public access systems, along the lines of PubMed Central, which also use embargo periods. I am not referring to institutional repositories that publish articles before the publisher’s embargo period ends, sometimes immediately upon publication. That is a separate, complex issue. Here I am only interested in the design of the emerging Federal system, which will include embargo periods as specified by the OSTP public access memo.
It was not apparent what type of repository you meant. Your comment makes more sense now.
Redundant copies would not be necessary anywhere if repositories and any federal public access systems would accept publishers’ tollfree hyperlinks leading to the version of record. A number of publishers already offer tollfree hyperlinks (particularly many publishers on the HighWire Press platform, but also the Association of Computing Machinery), and these achieve the goal of free access, while preserving publisher business models and avoiding redundant copies. Publishers offering these links now frequently provide them to institutional repositories, and federal agencies could also choose to accept these links in lieu of author submission of an acdepted manuscript
This is exatly what PAGES and CHORUS propose to do. What people misunderstand about CHORUS is that it is far more than technology; rather it is publishers agreeing to meet Federal embargo deadlines.
Thanks! After reading more about CHORUS on your hyperlink, I see that it’s a great idea. Publishers agreeing to meet federal embargo deadlines is a key step. Now the federal government has to agree to accept the links, correct? I believe publishers previously offered them to NIH, which refused and insisted on deposit of a redundant copy.
That is correct and PAGES is being designed to use the CHORUS links, assuming CHORUS works as promised.
