Trends for the Digital Library
David Seaman
Executive Director, Digital Library Federation
Access 2004, Halifax, Nova Scotia
15 October 2004

Preamble
http://www.diglib.org/




Thirty-three members – major academic and national libraries, including The British
Library; four allies ( CNI; RLG; OCLC; LANL )
Created in 1996 by directors of US research libraries; fills a need not simply met by larger library organizations: focuses exclusively on
DL needs and strategies for large libraries
Nimble, agile, collaborative
Practical and strategic areas of activity








USER SERVICES
Dimensions and use of the scholarly information environment www.diglib.org/pubs/scholinfo
IMS –repository/courseware integration
Distributed single collection of our own material
METADATA STANDARDS
OAI (Open Archives Initiative)
METS (Metadata Transmission Standard)









RESOURCE MANAGEMENT
XML format for license content (ERMI)
Registry of Digital Masters
PRODUCTION
Production benchmarks and good practices
PRESERVATION
Journals preservation
Global Digital Format Registry

Decoding Buzz
 New library disciplines still solidifying; new skills sets and work habits




Tipping points -- when does a new item move from irrelevant to “surprisingly non-terrible” to indispensable?
Non-library arbiters of access to scholarship
Ambition, Ignorance, and Lack of Money
Seismic events are routine and continuing
(Mosaic; Google; eBay; Palm Pilot; wireless).

Decoding Buzz
 Wireless delivery of library content



Text messaging, IM, BLOGS, and WIKIS
Harvestable metadata critical (OAI)
Acronymic Promiscuity 1 : XML; XSLT; CSS;
FOP; OAI-PMH; METS; MODS; MIX; DOI;
DC; ONIX; SCORM; MARC; EAD; TEI;
YAP; OEB; ARK; et al
 Embrace the churn
1 term borrowed from OCLC’s Lorcan Dempsay





In all of this whirl of change and possibility, collaboration is raised from something that is good to do to an imperative survival skill
Even for libraries (a more collaborative discipline than most), fundamental collaboration is really hard
Means shifting focus on where we compete onto services and local customization, not access to content? The challenge of sharing our stuff.
Means being deeply reliant on others (OCLC model of shared cataloging extensible to data too?)

Beyond Buzz






Blackboard/WebCT/rol your own
OpenCourseWare at MIT
Flecker/McLean DLF report: Digital Library
Content and Course Management Systems: Issues of
Interoperation http://www.diglib.org/pubs/cmsdl0407/
Libraries often absent from VLE decisions
SAKAI – a collaboration imperative at work
Libraries have an opportunity to be in the classroom like never before

Beyond Buzz








National Digital Information Infrastructure
Preservation Program (NDIIPP)
Global Digital Format Registry (DLF)
Digital Curation Centre (JISC UK)
Digital Preservation Coalition (UK)
Preservation metadata/tools (New Zealand Natl Lib)
PADI (National Library of Australia)
OCLC; RLG; DLF; CLIR
Cornell (excellent online training guide) http://www.library.cornell.edu/iris/tutorial/dpm/






Beyond Buzz
“There is an growing interest in the more coordinated management and disclosure of digital assets of institutions — learning objects, data sets, e-prints, theses, dissertations and so on.”
OCLC Environmental Scan , 2003. www.oclc.org/membership/escan/research/default.htm
Not really demand-driven. Some evidence that different disciplines react differently to the service and opportunities that repositories offer.
Real resistance from faculty to being seen as asset workers for an institutional content management system.
Intellectual capitalism: hoary question of who owns faculty
(or student) output and who can exploit it and for how long?
Faculty loyal to discipline and department more than to institution?




British House of Commons Science and Technology Committee.
Scientific Publications: Free for all? July 2004. (HC 399-I). http://www.publications.parliament.uk/pa/cm200304/cmselect/cmsctech/3
99/39914.htm

“This Report recommends that all UK higher education institutions establish institutional repositories on which their published output can be stored and from which it can be read, free of charge, online. It also recommends that Research Councils and other Government funders mandate their funded researchers to deposit a copy of all of their articles in this way.”
The U.S. House Appropriations Committee draft report, concerning NIHfunded research – requirement to deposit in PubMed Central’s repository.
New Zealand : “NZD$24 million funding from the latest budget will allow New Zealand to ward off ‘digital amnesia’, and protect New
Zealand's documentary heritage.” www.natlib.govt.nz/bin/news/pr?item=1085888952

Beyond Buzz
 The key issues are social and not technical:
 how to populate it; who gate-keeps it; how to build trust in it; how to advertise its worth;
 how to stop in becoming last year’s thing and the first thing that gets cut in lean times? That is, how so you embed deep in the institution’s conscience a sense that the IR is a long-term pledge of persistence.

Real danger of broken promises between institution and faculty.

Core question: how is the arrival of the Institutional
Repository (and Open Access) tied in with changes in the faculty rewards system? How integrated into institution’s reflection of valuable and rewardable contributions?

Beyond Buzz



Exciting glimpse of a future where all scholarship is free, with rich metadata that allows virtual aggregations of content held in seamlessly integrated IRs across the globe, all controlled by libraries and leading to a productivity boom and a global public informed by the best data there is.
New roles for libraries – publishing from IRs; customer support. New roles for publishers – add value not control access.
Investment needed to ad value to public content favors largest publishing operations – threatens smaller and society publishing? May lead to fewer publishers?





Beyond Buzz
Most journals now allow self-archiving: http://www.elsevier.com/wps/find/authored_newsitem.cws_home/companynews0
5_00145
Only 5% of journals are gold (typically “author-pays” model), but over 90% are already green (i.e., they have given their authors blanket permission to self-archive); yet only about 10-20% of articles have been self-archived. http://www.ecs.soton.ac.uk/~harnad/Temp/impact.html
Now it is up to us – no legal barriers to having (most) scholarship published in traditional peer-reviewed journals also freely published from our IRs
So, how are our scholar-authors motivated to self-archive, learn metadata skills, publishing skills? Why should they do it? Should they be encouraged to do this for the general good? How to reward them?

Beyond Buzz

While any or all of these specific examples may fail to come to fruition, the trend is solid:

Digital Opportunity Investment Trust (DOIT): $20 billion
“digital gift to the nation.” Massive digitizing for the public good
<http://www.digitalpromise.org/>
 US Government Printing Office: government documents print/film conversion – 2.2 million items
 Carnegie Mellon’s Million Books Project
 Google’s Project Ocean. Massive library digitizing of out-of-copyright material.




Beyond buzz
 Integration of content, especially licensed journals.
Every publisher (and library production unit?) is an island; we produce silos of data that plays badly with others. A good silo is a lovely thing – but not sufficient always.
OAI – Open Archives Initiative (XML, DC)
OpenURL/CrossRef
The need to have content that encourages local reorganization and creation of services, and that permits “beyond browsing and searching” engagement by individual users

Beyond buzz


 Little ability to work with content or even metadata cross-publisher and cross-aggregator. Or crosslibrary.
Digital couch potatoes versus rip/mix/burn:
We invite our users to visit sites and watch content channels (TV); they want to sample, re-use and repackage as a personal library, a classroom presentation (the music mix)







Google, Yahoo, Ebay, and Amazon are your friends
Rich personalization and communal ownership
OCLC/Worldcat
Yahoo and OAI via Michigan
Crossref/DOI and Google
Amazon and Google together at last: A9.com



Users are simultaneously over-whelmed with the time it takes to find relevant information in a “data silo” landscape, and (outside the sciences) under-whelmed by the lack of good material in their particular discipline.
TIME: 39% of all respondents (60% faculty) report insufficient time as their major problem
[Dimensions and use]



Much more content, and much richer, domainsensitive, finding systems are vital, as is the ability to enrich, re-shape, re-package, annotate, and contextualize the data once one has found it.
Respondents frustrated with finding information, determining its credibility, and analyzing it [ Dimensions and use ]






DLF Strategic Goal
New level of interdependence and deep sharing of master files
Two-phase Finding System, initially OAI
Digital Object Sharing or richer library services and better scholarship
New infrastructure and data creation needs – what are the characteristics of sharable content?





The transformation from isolation to integration is our central challenge and opportunity– with some enormous payoffs when we get it right.
Innovative users need malleable content with which to innovate; need to learn to re-shape content in a mutable library.
It is not sufficient simply to offer the current fragmented set of websites defined by a publisher or aggregator. This frustrates use, wastes time, and hogties library services.
Our collective response to frenzy of change is a demonstrably powerful force – DLF groups working together and across industries have had great success meeting local needs and global ones.