In a session on the Digital Public Library of America (DPLA), Robert Darnton, Director of the Harvard University Library, likened the concept to Thomas Jefferson’s observation that often the use of something does not diminish its value. For example, using one candle to light another spreads light and does not diminish the value of the first candle. This idea acquired a 21st century luster with the spread of the Internet. The use of information does not diminish its value. Public good benefits the entire citizenry and one citizen’s benefit does not diminish another’s; it is not a zero-sum game. However in considering these concepts, we must not lose sight that the acquisition of knowledge as a public good is not without cost. (Someone had to purchase Jefferson’s candle!)
Moving on to the present, Darnton noted that the DPLA is an opportunity to realize the enlightment and goals upon which our country was founded. Google tried to establish a major digital library and demonstrated that today’s technology could be used to create a new kind of library which, in principle, could contain all the books in existence. But Darnton observed that Google Book Search is an example of a good idea gone bad because of copyright problems and the alleged infringement of it by the Author’s Guild. Google did not pursue a legal case which (if they won) would have provided a significant public benefit, but instead they chose a commercial approach and negotiated a settlement with the Guild. The settlement was rejected by a Federal Court. So the time has come to create a digital library to make our cultural heritage available to the entire world.
In 2010, a steering committee was formed to provide guidance. Working groups were set up and produced a way to create a master plan, which was presented to the public last week.
Here are some of Darnton’s thoughts about features of the DPLA:
1. The DPLA will be a distributed system aggregating collections from research libraries and institutions. It will not be a single database but will consist primarily of books in the public domain from the Hathi Trust, the Internet Archive, and digitized collections made by large libraries independent of Google. Government sources are also rich sources; all 50 states have digitized their major newspapers and given them to the Library of Congress. These can be given to DPLA. Because of copyright law, most current literature will not be in the DPLA. DPLA’s mission should be defined to make its service distinct from public libraries. Darnton suggests that the DPLA exclude anything published in the last 5-10 years.
2. At its launch in April 2013, the DPLA will probably contain a basic stock and will grow as funding permits. It will be interoperable with major digital libraries in other countries (it has already made an agreement to cooperate with Europeana). The example of Europeana suggests the bare minimum funding needed to get the DPLA going. Brewster Kahle of the Internet Archive estimates a cost of 30 cents to digitize a page, or $300 million to digitize the contents of a large library, but others think the cost is nearer to $1/page. The DPLA will grow in accordance with its budget, which nobody knows yet. If a coalition contributed $100M/year, a great library could be created in a decade. DPLA will cooperate with Europeana, which estimates 5 Million Euros/year for its operating costs.
3. The DPLA must respect copyright. The first copyright law struck a balance between authors and publishers by providing limitations on the term of copyright. The current limit tips the balance toward private commercial interests. Every book published since 1923 is now covered by copyright, regardless of whether it has been renewed, and many owners are unknown which has led to the orphan works problem. The DPLA could try to reach an agreement between authors and publishers of books that have gone out of print.
4. The DPLA steering committee established a contest to develop a technical infrastructure. The technical subcommittee will develop a draft prototype to go into operation when the DPLA is launched.
5. A governments committee has only begun to study the administrative issues of the DPLA. The present interim leadership at Harvard will continue until the final DPLA comes into existence. It will serve a very broad and diverse community and is meant to serve the entire country, so it probably will not be at any elitist institution. Most people think it should not be part of the Federal Government to keep it free from political pressures.
Rachel Frick, Director of the Digital Library Federation summarized the operational plans of the DPLA for the next 18 months.
- Where possible, existing free or open source code will be used.
- The DPLA will be freely accessible for others to port or replicate.
- Metadata is the core of the discovery framework. It will aggregate existing library data and operate in a global data environment. All metadata will be freely available except where it would violate personal privacy.
- Content will incorporate all formats, not just books. It will begin with already digitized works in the public domain. It will grow with orphan works.
- Tools and services like APIs will provide enhanced uses of the content. The platform will be open to public innovation and enable the creation of new tools and services. It will provide APIs. With Europeana it will share an interoperable data model and source code.
- Community will be a participatory platform that supports users and developers who wish to reuse content and metadata.
- A discovery layer will provide access to secondary sources.
More information is available on the DLPA website.
The DPLA beta-sprint was an aggregate of metada from 1400+ collections in 44 states. Visit dpla.granger.illlinois.edu to see it. 60 organizations submitted letters of intent, and several were chosen to demonstrate their systems at the DPLA plenary conference.
Columnist, Information Today and Conference Circuit Blog Editor
My thanks to Carol Tenopir, University of Tennessee, for her contributions to this post.