5. Metadata – Discussion Questions
The Government Printing Office (GPO) document, Digital Distribution to Depository Libraries: Exploring the Issues (9/6/2006) asks these questions about metadata:
4. Metadata a. Should GPO distribute bibliographic and other metadata as well as the digital publication? b. What other metadata should be routinely distributed?
FGI Responses Q: a. Should GPO distribute bibliographic and other metadata as well as the digital publication? Yes. But the issue is more complex than this question implies. We need to be thinking of "metadata" in new ways. It might be best to think of this question as two separate issues: First, should bibliographic and other metadata records be freely available? Secondly, should they be distributed (as opposed to being freely available) to FDLP libraries? We would answer "yes" to both those questions. We believe that a wide variety of rich metadata should be made available to the public without charge. The specific implementation of such a policy would require study, but we can imagine one specific instance that serves as an example: GPO could expose OAI records for harvesting using OAI-PMH. This would allow libraries, archives, digital repositories, citizens groups, and even individuals, to create indexes, guides, portals, bibliographies and even collections of government information, freely and relatively easily. We also believe that there need to be mechanisms for digital "deposit" of metadata with FDLP libraries. Such mechanisms will need to be developed, but they should include the ability of FDLP libraries to select what they want and receive what they select easily. This would enable FDLP libraries to easily and even automatically build their own indexes, guides, portals, bibliographies, and collections through the use of their OPACS and other tools. It would also facilitate FDLP libraries in their management of their own collections -- including government and non-government information -- in a common environment and with minimal special procedures for government information. Q: b. What other metadata should be routinely distributed? There are many different types of metadata: preservation, administrative, descriptive, technical, use, and others. (See for example, Gilliland-Swetland, Anne J. “Setting the Stage: Defining Metadata†in Introduction to Metadata: Pathways to Digital Information , Murtha Baca, ed. Los Angeles: Getty Information Institute, 2000, second edition. 12 pp. -- Third edition forthcoming, Summer 2006.) In addition, metadata can include Semantic Web, Thesaurus, and RDF-like information. Metadata can be highly structured and controlled, or user-contributed. A rich depository system needs methods of exposing, distributing, and sharing such metadata easily. One of the most valuable pieces of this puzzle will be the high-value, authoritative, metadata constructed by and for GPO for use in describing and managing government information for the long-term as mandated by Title 44 and as administered through FDsys. There should be a very high priority for distributing such metadata to FDLP libraries. Continue reading
4. Regional Responsibilities – Discussion Questions
The Government Printing Office (GPO) document, Digital Distribution to Depository Libraries: Exploring the Issues (9/6/2006) asks these questions about Regional Responsibilities:
3. Regional responsibilities a. What are the responsibilities of regional depositories for the digital publications GPO distributes? i. Should they be required to accept and maintain all digital publications that GPO distributes? ii. Should regional depositories be allowed to accept digital content as an alternative to the print or microform version?
FGI Responses We're somewhat hampered here at FGI because none of us has worked at a regional library. All of us have a selective perspective. So while we have opinions on this section, we're really hoping a Regional reader or two will come forth with feedback on these questions! Q: a. What are the responsibilities of regional depositories for the digital publications GPO distributes? We at FGI believe that regional depositories have the same responsibilities for digital materials as they do for all other formats. Q: i. Should they be required to accept and maintain all digital publications that GPO distributes? They should either be required to accept all digital publications or have selective housing agreements in place with selective depositories in their system. We should note that when we say "digital publications", we mean document-like objects. No one here is suggesting that any depository should be expected to be a full mirror for dynamic-database driven content. For example, we want to see deposit of Census publications and special reports, but American Factfinder would not be mirrored by any depository. Hosting databases usually requires software in addition to data. Even when the data is public domain, the software usually is not. Q: ii. Should regional depositories be allowed to accept digital content as an alternative to the print or microform version? As long as Regionals accept local storage of digital materials and have some kind of backup and/or preservation plan in place, they should be allowed to accept digital content as an alternative to print or microform version. Ideally such selection would take state bandwidth capabilities into account and be done in consultation with the local selective depositories. Continue reading
3. Infrastructure – Discussion Questions
The Government Printing Office (GPO) document, Digital Distribution to Depository Libraries: Exploring the Issues (9/6/2006) asks these questions about infrastructure:
2. Infrastructure a. Who determines sufficiency for local infrastructure to provide timely and effective public access? i. What are the criteria? ii. Should GPO develop guidelines as it did for workstations?
FGI Responses Q: a. Who determines sufficiency for local infrastructure to provide timely and effective public access? The same authorities who determine sufficiency in other areas of depository management. Q: i. What are the criteria? We would like to suggest a two tier criteria. A library accepting deposit of electronic government publications should be able to either: A) Provide download times for government publications similar to that from GPO. OR B) Be able to provide documents by CD/DVD, or by in-library use only IF the library is located in a low bandwidth area. The Pew Internet & American Life Project found that 75% of rural residents lacked broadband access as recently as February 2006. People in such areas aren't going to download documents easily from any web site. Q: ii. Should GPO develop guidelines as it did for workstations? We believe that GPO should develop guidelines for serving local digital documents like they have done so for workstations. GPO should focus on functional requirements rather than dictate specific systems, unless that hardware is required to receive automated distribution of digital publications. Continue reading
2. Access – Discussion Questions
The Government Printing Office (GPO) document, Digital Distribution to Depository Libraries: Exploring the Issues (9/6/2006) asks these questions about Access:
1. Access a. Is the major goal of digital distribution to improve public access to FDLP publications, and if so, how is that goal facilitated by libraries providing local access to a digital copy? b. Should libraries receiving digital distribution be expected to offer no-fee, anonymous public access to local copies of FDLP digital publications, and to minimize any restrictions such as user registration, location, etc? c. Should depository libraries take active steps, such as including metadata in their catalog or developing appropriate web pages, to enable users to identify and link to FDLP digital publications in their collections?
FGI Responses Throughout our responses we will treat the terms "digital distribution" and "digital deposit" as interchangeable. Q: a. Is the major goal of digital distribution to improve public access to FDLP publications, and if so, how is that goal facilitated by libraries providing local access to a digital copy? Digital distribution probably will improve public access to otherwise available data, but that is not the major goal of digital distribution. In the view of FGI volunteers, as well as many digital deposit advocates, the major goals of digital distribution are:
- To help assure the preservation of electronic publications through many copies. One copy or a small number of copies is liable to destruction or alteration, whether through accident or malice. But if we start out with hundreds of initial copies, it is very likely that at least a few institutions will successfully preserve a given publication.
- To assure perpetual no-fee access to publications acquired through digital deposit. The Future Digital System FDSys is required to be policy neutral and it will also support the needs of the GPO sales program, therefore the infrastructure for charging for electronic government publications will exist. As budget woes grow across government, pressure will build for "reasonable prices" for government publications. With a single-system solution like FDSys, Congress could mandate a subscription plan that would automatically deny access to previously published items. With digital distribution, even if the government eventually charges for new publications, the older publications would remain fee-free at the nation's depositories.
- To force any removal of government publications into a public process. When the government holds all copies of a publication on its own servers or those of commercial contractors, items can be removed quietly outside of normal processes. When those same electronic publications are deposited with libraries, then any removal must follow a public process like the sensible one outlined in ID 72: Withdrawal of Federal Information Products from GPO’s Information Dissemination (ID) Programs. Digital distribution is simply the people's insurance that withdrawal of information won't take place under cover of night.
- To hinder official efforts to collect personally identifiable information on users of electronic government publications without a court order. When publications are dispersed across the country, no one query can pull together a list of everyone who has used a file. By contrast, information directly pulled from federal servers can be tabulated by IT staff without any legal process at all. History has shown that federal governments of both parties tend to be overzealous in their data gathering efforts and wind up spending time investigating folks like Lucille Ball, Albert Einstein, and Thomas Merton. We should put a few checks on that tendency.
- Protection against partial Internet breakage - People can access local digital collections built by deposit when there are Internet problems in their area. This week, a fiber optic cable north of Talkeetna Alaska was cut and all areas north, including Fairbanks, lost Internet access for servers south of the cut for two days. If a local digital collection existed at the depository library at the University of Alaska Fairbanks, people in Fairbanks could have accessed those publications online even though they couldn't get to GPO Access.
- Insurance against natural disasters/terror attacks - We are going to have more disasters and terror groups haven't retired. While a single mirror site for publications on GPO Access has been bandied about for five years or so, we still see messages from GPO that GPO Access will be down for maintenance for hours at a time. Every once in awhile days at a time. Digital distribution would instantly provide dozens or hundreds of mirrors for a given publication.
- Help with bandwidth issues - Many electronic government publications are many megabytes in size, making them difficult to download over the dialup connections used by as many as 30% of the American population. Libraries with local digital collections housed on library services could easily save these onto DVDs, flash drives or even MP3 players brought in by patrons seeking access to large government documents. Download speeds from LANs are much faster than the general Internet.
For more on the issue of access in general, please see our access issue page, and of course, Government Information in the Digital Age: The Once and Future Federal Depository Library Program. What do you think about the Access discussion questions? Let us know! And if you are going to Council, leave a comment or send us a note about what you heard! Continue reading
1. General Assumptions
The Government Printing Office (GPO) document, Digital Distribution to Depository Libraries: Exploring the Issues (9/6/2006) frames the discussion with these six basic "general assumptions":
- Consistent with other formats, GPO will distribute authentic digital publications in formats intended for public access.
- The characteristics of digital publications vary so significantly from tangible products that new, more flexible, guidelines for managing them in depository libraries will need to be developed.
- Preservation of the source files, called Archival Information Packages in FDsys, will be the responsibility of GPO and its preservation partners.
- For a majority of content, digital distribution represents another format choice that can be selected, in addition to print, microfiche, and tangible electronic products.
- Libraries receiving FDLP digital publications would be responsible for providing sufficient infrastructure, including bandwidth and storage, to provide timely and effective public access.
- Libraries would need to ensure that they are providing access to the same version(s) of a digital publication that is/are available from GPO.
We at Free Government Information feel that assumptions 1, 2, 4 and 5 are extremely reasonable assumptions and hope they will form part of the philosophical foundation for any mostly digital depository program. We must strongly object to assumption three, that the preservation of documents will be the responsibility of GPO and unspecified preservation partners. This represents an upending of over a century of successful distributed preservation efforts and launches GPO into entirely uncharted waters. Uncharted territory for a chronicly underfunded federal agency subject to constant fiscal pressure. While depository libraries are also financially challenged, there are more libraries than there are GPOs and so we continue to believe that lots of copies keep stuff safe. We hope that FDSys and government contracted partners will be part of a preservation solution, but relying on any one entity for preservation of fragile digital objects is suicidal for information. There is no polite way to put this. We do not feel as negative about assumption six, but feel the community requires more information before they agree to it. Certainly for the latest available publication libraries should be serving the same version as GPO. They should also be responsible for marking previous editions as superseded and pulling them off their servers if appropriate. But we feel strongly that superseded editions in the digital world deserve the same treatement as their printed counterparts. Many libraries will only keep the most current version, and others will keep older editions for their historical value unless the agency has formally withdrawn them from the Federal Depository Library Program. If a GPO of the future decides to keep "latest only" for budget reasons, depositories libraries should not be obligated to do the same. If this is made clear, than assumption six becomes reasonable as well. Continue reading
Latest Comments