What You Need to Know about the New Discard Policy
GPO’s new Regional Discard Policy and GPO’s recent presentations about it are full of hopeful words and good intentions. We applaud GPO for having good intentions and high hopes, but we question if the Policy can meet those expectations.
- Introduction of the Policy and its Implementation
- One definite Goal. Some questionable objectives
- Preservation of and Access to Paper Copies
- Next steps
Here is what you need to know about the Discard Policy. GPO’s caveats and assurances about the new policy aside, there will no longer be any Regional Depositories for documents more than seven years old. It removes the requirement that there be access paper copies of all documents in the FDLP. It weakens the FDL Program by continuing the shift of responsibility away from FDLP members and toward GPO. It does not increase flexibility (as advocates of the policy claim), it shifts flexibility away from Selectives and gives it to Regionals. It puts new burdens on Selective Depositories. It establishes a new model for the preservation of paper copies of documents that is undocumented, unproven, and risky. It ignores long-term implications in favor of short-term benefits to a few large libraries. It makes GPO’s “guarantee” of long-term, free access to government information nothing more than a hollow promise.
We believe that the Policy actually weakens the FDLP and damages both access and preservation. We believe that the Policy provides no guarantee of meeting those expectations, and will make it more difficult to do so. Below, we explain why.
At the Fall Depository Library Council (DLC) meeting, GPO gave a general outline of how it will proceed to allow Regional Depository Libraries to start discarding paper copy documents.1 GPO has, so far, provided the following information about the Policy itself and how GPO intends to implement the policy:
- Government Publications Authorized for Discard by Regional Depository Libraries [draft policy, 07/09/2014]
- Vance-Cooks, Davita. [letter (July 10, 2015) from GPO to Gregg Harper, Chairman Joint Committee on Printing (JCP) requesting approval of policy to give regional Federal depository libraries the option to withdraw tangible depository materials]. and Harper, Gregg. [letter (August 5, 2015) to GPO] Both documents in one PDF file here.
- Council Session on Discards audio recording and presentation slides (10/20/2015)
For additional background, links, and commentary, see: Information sharing and the National Plan by Shari Laster.
The new policy has only one stated goal: To allow regional depository libraries the option to discard paper copies of government documents.2. To be clear, this is not a substitution of one format for another, like microfiche for paper. Regionals will not be required to uphold their Title 44 obligations to “retain at least one copy of all Government publications either in printed or microfacsimile form.” (44 U.S.C. §1912).
In addition to this specific goal, GPO has expressed a variety of objectives, which it apparently hopes the new policy will help accomplish. But GPO has been both inconsistent and vague in its expression of these objectives and how it will actually implement the policy.3 Six Regional Depositories will participate in a test of the policy in early 2016; presumably, this will produce more implementation details.
Some of GPO’s objectives (such as giving Regionals “the ability to expand their capability to serve the increasing number of remote users” [Vance-Cooks]) can be accomplished without the new policy.
Most of the objectives relate to giving Regionals the “flexibility” to discard paper copies of documents. GPO does not claim that this will have any positive effect for users. On the contrary, GPO acknowledges that regionals that are already relocating tangible collections to offsite storage are impairing the goals of the FDLP.4 GPO implies that Regionals will use resources that will be freed by discarding documents “to focus on the needs” of users of government information [Vance-Cooks]. But GPO does not specify what the resources are, or explain how it expects freed space to be reallocated to services or collections for users of government information, or require any such reallocation. Furthermore, some Regionals have admitted that any savings brought on by this policy will not go toward public service of government information, but will go toward their library’s central operating budget. Since the Policy does nothing to further such objectives, we should not read them as objectives of the Policy but as wishes of GPO.
The Policy does have one objective that has a direct relationship to its Goal. While the Goal allows discarding, this objective will retain paper copies in FDLP. Since the policy will only accomplish its Goal if documents are discarded, it is essential to know how many copies will be retained and how they will be used. This will illuminate how having fewer paper documents in FDLP will affect access and preservation.
Unfortunately, GPO has been inconsistent in both its intent and its support for this objective and, at every juncture, it has acted to weaken requirements for both access and preservation. Here is a timeline that demonstrates this:
- GPO requires retaining some copies. In its draft Policy of July, 2014, GPO recommended retaining a “requisite number” of copies without explaining how it would determine what that number was.
- GPO includes paper preservation and access. In the 2014 draft policy, GPO also expressed its reasons for retaining some copies in FDLP as “for access and preservation.”
- GPO rejects research. In August 2014, GODORT made several recommendations in response to GPO’s draft Policy, including that discarding should not be done without research that determines the “appropriate number” of paper copies of a publication necessary to ensure access, preservation, and re-digitization.5 GPO never responded to this recommendation or explained why it rejected it.
- GPO drops paper access and preservation requirement. The policy that GPO sent to JCP in 2015 when it asked for approval, did not include a requirement to retain any paper copies and did not include any mention of either access or preservation of paper copies. This new version of the policy only had two requirements: that the document must have been retained for seven years and that there was an authentic signed digital copy in FDsys. GPO asked JCP to approve a policy in which no paper copies would be guaranteed at all and the digital FDsys copy would be enough. [Vance-Cooks]
- JCP imposes the requirement. JCP approved GPO’s policy, but added a new requirement that GPO had not requested. JCP said that “A minimum of four tangible copies of the publication [must] exist in the FDLP distributed geographically.” JCP also specified why it wanted this requirement: it would “enable the Program to accomplish its goals while also providing the necessary access to the materials” [emphasis added].
- GPO ignores JCP. Although GPO accepted JCP’s requirement that it would retain a minimum of four paper copies in the FDLP, it ignored the explicit rationale of access that JCP gave for it. During the question and answer period of the Fall DLC meeting, GPO said that the four copies will be for “preservation” only and they will be stored in “dark archives.” It is not clear how GPO can ignore this aspect of JCP’s response.
The JCP-approved policy sets the “requisite number of copies” that FDLP will be required to retain and GPO has defined the purpose of those copies.
- Four Preservation Copies: The new policy requires that GPO will ensure that “A minimum of four tangible copies of the publication exist in the FDLP distributed geographically.” The four copies will be for “preservation” only and will be retained in a “dark archive.”
- Zero Access Copies: The policy does not require any paper access copies.
Implementation of the Policy will begin with a testing phase in January 2016. Regionals will have to get approval from the Superintendent of Documents to discard documents and GPO will set its own criteria for accepting or rejecting discard requests. What we know at this point about the criteria and procedures GPO will use to accept or reject requests comes from GPO’s DLC presentation.
Although the presentation slides list a variety of items as the “Basis for GPO Discard Decisions,” neither preservation nor access are among those criteria. Instead, GPO states that the Policy only requires GPO to “To ensure a minimum of 4 geographically distributed tangible copies of publications exist in the FDLP.” As noted above, GPO says these will be for preservation, not access.
Significantly, GPO also commits to maintaining an inventory of regional depository holdings, but not an inventory of all FDLP holdings.
The other details revealed by GPO in its DLC presentation and during the question and answer session following the presentation raise more questions than they answer.
- Ten regions, not four. GPO says it will use its own ten “Regional Printing & Procurement Office” regions to help ensure geographic distribution of paper copies retained within the FDLP. It is not clear if GPO intends to store preservation copies in its own regional offices. It is not clear if GPO intends to have one paper copy in each of those ten regions. It is not clear how the ten GPO regions and the ten GPO regional offices will be involved in implementation of the policy. GPO did make it quite clear that it will enforce a “minimum of four” paper copies, not a minimum of ten. In other words, although it will use ten regions, it only guarantees that there will be at least four paper copies of any discarded document held within the FDLP.
- What will trigger preservation storage? GPO did not describe any procedure for identifying titles to be put in to “dark archive” storage. Will GPO put the first four copies of any withdrawn title into dark storage? Or, will it wait until there are only four copies in all Regionals before requiring those copies be moved to dark storage? Or, will it wait until there are only four copies in all of FDLP? Will GPO use some other, as yet unspecified, number of copies in Regionals, or Selectives, or both to trigger dark storage?
- Dark Archive — or not. GPO used the term “dark archive” on one slide that described “preservation copies.” That slide simply stated: “Presume dark archive.” Does this mean that storage in a dark archive is still being discussed? During Q&A, GPO was clear that the four copies will be preservation copies — not for access or use. Research shows that storing paper copies in dark archives increases the likelihood of successful long-term preservation.
- Who will preserve the four preservation copies? Apparently, this is an open question. The only unambiguous answer to this question during the DLC presentation, was that GPO was not requiring (or, apparently, anticipating) that there will be four, discrete, complete collections. GPO spoke of its own regional offices as being more involved with this FDLP policy. Does that mean that GPO is considering storing preservation copies outside of FDLP libraries in its own regional offices? Is GPO considering special non-FDLP “partnerships” for storage? Would such arrangements meet JCP’s requirement that the copies “exist in the FDLP”? Will GPO require and enforce a strict definition of the conditions of a “dark archive”? If so, will that limit the facilities that are eligible for storing preservation copies? Will Regionals that are anxious to discard their paper copies be willing to sign up for new storage requirements? Will Selectives?
- The Inventory. GODORT recommended that GPO and the FDLP community should collaborate on the development of a national inventory of historical federal publications held in depository libraries before discarding documents from that collection. Evidently, GPO plans instead to consider only those copies in its inventory of Regional Depositories. GPO’s own policy implementation criteria specify that the preserved copies should be the “Best copy available.” To determine that, there would have to be some indication of the condition of all existing copies in order to determine which is the “best copy.” GPO did say that it is “working on” an inventory of the National Collection, and would prefer to have a union catalog of everything, but implicitly rejected GODORT’s recommendation to require a national inventory. Public Printer Vance-Cooks said in the letter to JCP that there are “more than one million titles” currently eligible for discard based on the contents of FDsys. Will GPO start with an inventory of FDsys titles and try to match those to discard requests? If so, will libraries discard based only on what is eligible for discard rather than on an evaluation of an inventory of their complete holdings? If so, will that lead to uninformed discard decisions? It is difficult to imagine how the Policy can be implemented effectively without a National Inventory.
- What constitutes a “title” eligible for discard? During the test period, the holdings to be tested will be limited to four titles in six Regional Depositories. Two of the four “titles” being tested are actually series: “GAO Reports and Comptroller General Decisions” and “Hearings from committees to be named.” Will GPO treat these as series to be discarded en masse, or will each title in each series be evaluated for discard individually? A third test title is described by GPO as “Title(s) determined by test libraries.”
- No requirement for access paper copies at all. GPO will no longer require FDLP libraries to contain any access paper copies of documents that are eligible for discard. By policy and procedure, it will, of course, be possible for a Regional Depository (or a Selective) to retain a paper copy of a document for local access, but, as far as paper copies of documents in the FDLP system, GPO’s only requirement under the new policy is that there will be at least four paper preservation copies nationwide of those documents that are eligible for discard. By policy, there will no longer be any Regional Depositories for documents more than seven years old with an FDsys digital surrogate.
The new policy weakens the strengths of the FDLP and fails to strengthen FDLP’s weaknesses.
One of the basic strengths of the FDLP is that it instantiates in rules under an existing legal mandate a simple formula for access and preservation of documents. These rules result in a simple formula of lots-of-copies distributed geographically, facilitating both access AND preservation. Regardless of the state of federal budgets or political decisions and regardless of choices or errors by individual libraries, these rules mean that it is very difficult to lose all copies of any title and it is relatively easy for readers to get a copy of any title, regardless of where they live. (Lots-of-copies also helps ensure interlibrary borrowing.) The policy spreads both the responsibility and the authority for preservation and access over many institutions. No single institution can cause loss of the information or access to it. The system is far from perfect (e.g., inadequate inventories and bibliographic records, cumbersome withdrawal and exchange procedures, inadequate staffing), but one of its strengths is that it does not have to be perfect to be effective.
The new rule will destroy this strength. It puts the burden of paper preservation on as few as 4 or 5 institutions for any given title. It removes any requirement for providing access copies anywhere within the FDLP. Any FDLP library that wants to continue to provide access will do so on its own and without the assurances of 47 Regional library collections.
The FDLP also has weaknesses. One digital-era weakness is self-imposed. GPO has chosen to set itself up as the sole FDLP institution responsible for preservation and access of born-digital documents. Since almost all federal government publications are now born digital, this puts onto GPO — and GPO alone — the entire burden of preservation and access of those documents. This model has to be perfect to be effective. If GPO does not exercise its power perfectly, the result would be a catastrophic loss of access or preservation or both. This system is very fragile. Any of a number of things could cause such a catastrophe: A cut in GPO’s budget; a change in policies by a different GPO administration; a technological failure or mistake; an administrative failure or mistake, and so forth. GPO could have used the new policy to fix this weakness by requiring Regional Depositories to store a digital copy of paper copies that they discarded, but it did not do so. When this idea was raised at the DLC question and answer period GPO was mute.
With these strengths and weaknesses in mind, here are our conclusions with regards to the new discards policy.
- No Paper Copies for Access. Perhaps the biggest change the discard policy makes is also a subtle one. No one at GPO said this out loud, but the policy is clear. For those documents that are eligible for discard, the policy removes the requirement that there be any paper copies for access and use within the FDLP. Although Regionals and Selectives are certainly allowed to keep paper copies for access and are not required to discard any, the new policy nevertheless flips the old policy on its head. Instead of guaranteeing paper copy access through the Regionals, now it removes any guarantee that any library will retain any paper copy for access. If, in the future, there are access paper copies, it will be in spite of this policy, not because of it.
- Continues the shift away from FDLP Libraries. For the last 20 years, GPO has taken over the responsibility and power to control both preservation and access of digital government information. FDLP libraries have largely accepted this situation. This has left individual FDLP libraries without the ability to select, acquire, organize or deliver collections to their communities. Since FDLP libraries have no digital documents collections, they have no way of curating collections for their communities and no control over discovery or access or utility of digital documents. They cannot guaranantee that documents their communities require will be preserved. They cannot guarantee that documents will not be moved or altered or deleted. This makes every individual FDLP library less able to meet the needs of its communities and that, in turn, makes each FDLP library less useful and less important to its communities. By creating a policy that shifts both collections and access away from FDLP libraries, GPO is continuing that shift of power to itself. It is even possible that GPO will shift preservation paper copies out of FDLP into its own regional offices or into the hands of “partners” outside of FDLP completely. Each FDLP library should consider whether this shift of power serves the needs of its communities adequately and whether it strengthens or weakens its own ability to meet those needs.
- Puts new burdens on Selective Depositories. The whole concept of Selectives that are “selective” and Regionals that are “complete” is now gone. If a Selective needs a paper access copy and cannot convince its Regional to retain it, its only option is to assume that responsibility itself.6 This puts Selective Depositories in a new and less clear position. Since some regionals have adamantly and relentlessly advocated for the “flexibility” to discard, will Selectives have to have complex agreements among themselves to guarantee access to paper copies if their Regional refuses to retain a document? The policy does not answer this question; it creates this new problem. The policy could result in Selective Depositories taking on the sole responsibility of providing access paper copies of discarded FDLP materials. The policy weakens the FDLP by having no requirement or procedure on which libraries can rely for long term access to paper copies of documents. This has one very important implication for Selectives that has not, to our knowledge, been discussed: Selectives have already withdrawn documents with the understanding that their Regional has copies. Selectives that need paper copies of those already-withdrawn documents will be facing a dilemma under the new policy: either lose access to those paper copy documents or re-acquire them when the Regional discards them.
- Risky Preservation of Paper Copies. While the policy allows more than four paper copies to be retained somewhere within FDLP, it does not require it. Are four copies enough to ensure the survival of at least one copy for the long-term? We do not know because GPO is implementing this policy without either the research that would answer that question or procedures in place to implement policies that such research might suggest. GPO says that it will use the preservation copies for digitizing and re-digitizing.7 Will those uses endanger the preservation of the paper copies? We do know that, although access copies are more likely to be lost or damaged over the long term, the policy of one copy in each Regional at least met the common sense criteria of lots of copies being better for preservation than fewer copies. If, for the long-term, more than four copies are needed they will be provided, not by FDLP policy, but by the good graces of FDLP libraries. If it is discovered in 10 or 20 or 100 years that we should have preserved more copies, it will be too late to fix the problem. (There is little or no duplication of titles in the FDLP Historical Collections anywhere outside of FDLP Libraries.) Replacing a policy that is successfully preserving documents with one that increases risk of loss and provides no evidence that it can work is irresponsible.
- Long-term implications. The policy places all its hope for success of preservation and access on the ability of digital documents to replace and substitute for paper documents. It is probably true (though GPO has not provided any data to demonstrate this) that most if not all of the more than one million documents that are currently eligible for discard are born-digital. To the extent that this is true, this should alleviate some of the above concerns because a lot of born-digital documents are designed to be read online rather than on paper. Loss of access to paper copies of such documents should not be a grave concern. But there are several caveats to this alleviation of concern. Not all born digital documents are created equal.
- Some documents, though created digitally, are designed for print distribution. For these documents, the print may be the better and preferred format for use. Allowing them to be discarded will not serve library users. Even GPO’s own policy document SOD 301 (“Dissemination/Distribution Policy for the Federal Depository Library Program”) requires document distribution to consider the characteristic differences between paper and digital including functionality, utility, reference value, requirements of special needs populations, and what format is commonly accepted by a user community. The Discard Policy has no such considerations.
- As far as we know GPO has not provided the FDLP community with examples of “best case” and “worst case” authentic signed documents. It is reasonable to assume that digital documents created 21 years ago are not as good as documents created today. If GPO accedes to pressure from libraries wishing to discard their paper copies and brands inferior digital copies as “authentic,” information will be lost.
- The digital and print copies of titles may not always be identical. It will be necessary to distinguish such cases to avoid loss of information.
- Given the problems of getting born-digital documents from agencies, GPO may accept whatever the agency provides as “authentic” even if it does not meet standards that GPO would require for documents it creates itself.
- Digitizations of paper documents may be accepted by GPO but not acceptable for use. We know that some of the same libraries that have relentlessly pushed GPO to adopt the discard policy have also pushed to allow digitizations of paper documents to be accepted as “surrogates” for the original paper copies. Documents already digitized and available (with restricted access to non-members) through the HathiTrust have been mentioned as possible candidates. We know, however, that one study of digitized books in the HathiTrust found that about 36% of the volumes studied were found to be so inaccurate and incomplete that they could not be considered “reliably intelligible surrogates” for the original paper copy.8 FDsys already includes digitizations (presumably making them eligible for discard) that have no digital text (OCR).9 GODORT recommended a rigid quality assurance program to ensure that digitizations are accurate and complete. GPO has not addressed this issue yet, but if GPO concedes to the demands of libraries that want to discard more volumes and allows such digitizations to qualify for the discard policy without adopting the GODORT recommendation, the result will be a recipe for discarding information without preserving it.
- Expanding the eligible universe. What is at risk? This is not about a limited number of documents. The scope of the policy covers the entire body of the FDLP Historical Collections. In the letter to JCP, GPO estimated that more than one million titles are currently eligible for discard. That is from one-third to one-half of the total number of titles estimated to be in the entire Historical FDLP Collection.10 This covers documents as old as 1994 (when GPO began online dissemination) but already goes back further than that and will go further still — by including any digitized Government publication made available on FDsys. GPO wants to develop a process for ingesting digitized copies into FDsys [Vance-Cooks]. In evaluating the policy, we must assume that documents earlier than 1994 will be added to FDsys and be qualified as documents eligible for discard. One reason for this is practical: We all want more digital documents! The other reason is political: The libraries that have pushed hard for the discard policy are libraries that insist they need to discard their historic paper collections. The great bulk of those collections were deposited before the digital era — before 1994. Thus the discard policy will not accomplish the one thing it was designed to accomplish (allow libraries to discard their historic collections) unless those pre-1994 volumes are made eligible. The main way that such documents can be made eligible is by digitizing the original paper copies. As noted above, without careful control to ensure that digitizations are both accurate and complete, discarding digitized paper copies will result in loss of information.
- A Fragile Guarantee of Preservation, no Guarantee of Access. If the discard policy is to guarantee no loss of information it must meet a higher standard than the Policy provides. Much of the power or weakness of the discard policy will be embedded in its implementation procedures. There are three reasons to worry about this.
- First, as noted above, GPO has already placed discarding paper documents as a higher priority than preservation of (or access to) those documents by ignoring and eliminating and weakening procedures that could have ensured their preservation. The result is an approach to preservation that is undocumented, unproven, and risky.
- Second, since we have only vague descriptions of GPO’s intentions, we have no way of knowing that the policy will be implemented in a way that will preserve the paper copies. By refusing to require retention of paper copies, GPO is removing the guarantee of paper copy access. GPO has given itself its own “flexibility” to implement the policy and has imposed no constraints or measurable goals for preservation or access on itself.
- Third, because the new policy throws out the old rules that have guaranteed long-term access, we are left with an FDLP that is subject to the whims of each new GPO administration and each new Congressional budget. This is a fragile policy. It provides no way of guaranteeing over the long-term what the current GPO administration says it intends to provide. (GPO’s legislative mandate is limited to providing “an electronic storage facility” and “a system of online access” without any explicit mention of historical documents or long-term preservation [44 USC §4101]; Title 44 also specifically authorizes GPO to charge fees for access [44 USC §4102], which it has done before.) The policy replaces a long-term, rule-based, many-institution guarantee, with a short term, few-institution implementation — without enforceable requirements — that can be changed without notice.
Some, but by no means all, of the concerns above can be addressed with rigorous procedures and tight enforcement of the weak rules of the discard policy. The test period will give us all, not just GPO and the six participating libraries, an opportunity to judge how this policy will be implemented. GPO has given itself the power to refuse discard requests, so it is possible for GPO to make the policy much more stringent in implementation than it is in wording. Unfortunately, the pressure to create the policy will not be alleviated unless GPO allows lots of “flexibility” — which translates into lots of discards and weak enforcement of preservation and access.
There is still, perhaps, time to convince GPO to create stable, consistent internal “SOD” policies that will be harder to break or discard in the future — policies that will do a better job than the bare requirements of the discard policy itself. During the test period in January 2016, all government information professionals should watch and evaluate not just the short term success of the Policy in enabling libraries to discard documents, but the long-term policies that will help prevent loss of information, or loss of free access to that information. Here are some suggestions that might help. We invite you to suggest your own.
- Require OAIS standards for paper preservation copies. The Open Archival Information System (OAIS) is the preservation standard often associated with and used for the design and evaluation of digital repositories. But the standard is explicitly designed to apply to all formats in all long-term preservation archives.11 Therefore, as GPO determines how paper preservation copies will be stored and preserved, urge GPO to adopt the OAIS standards for paper preservation.
- Slow down discard while determining preservation and access needs. As unlikely as it may seem that asking GPO yet again to consider using a research-based approach to address the question of how many copies are needed to ensure preservation and access will produce a positive response, this is the time to do so. Discard decisions are irreversible.
- Apply SOD 301 Standards to discard decisions. As noted above, GPO’s SOD 301 policy requires GPO to consider the functionality and utility of paper copy documents in comparison to their digital counterparts. Urge GPO to include similar criteria when it considers any discard requests.
- The Inventory. During the test phase, GPO is likely to discover two things. First, that it will be difficult — and, possibly, expensive — to match holdings records at the test Regionals to FDsys records. Second, it will be impossible to know how many copies actually exist in FDLP without an accurate, complete inventory. This is the time to make the case again for the much needed national inventory of the FDLP Historic Collections. It will actually be easier and cheaper to do (safe) discarding of paper documents with this inventory and harder and riskier to do it without.
- Quality Assurance of Digitizations. This is the time to investigate just what your communities are getting when you deliver digitized documents to them instead of paper. Urge GPO to establish a QA process and apply it to digitized documents before authorizing discard. At minimum, we recommend using the The Digital-Surrogate Seal of Approval for page-images complemented with complete and accurate OCR for digital text.
- FDLP Libraries First. Urge GPO to rely on FDLP libraries for preservation and access before it reaches out to new “partners” with secret contracts and unknown agendas. If GPO wishes to rely on private-sector partners, insist that GPO stand by its claim that all “partners” will be obliged (by publicly viewable contract, we hope) to abide by the long-term, free, public access standard that the current GPO administration favors.
- Participate. Be active in this process. Keep your communities as well as your library administrators informed about what is happening and what the implications may be for them (e.g., increased use of ILL? increasing number of ILL requests of your library?). Do your own evaluation and draw your own conclusions; don’t just rely on those who have a vested interest in discarding to tell you how great everything is going.
- Monitor The Plan. Some of the success or failure of the discard Policy will relate to the success or failure of GPO’s National Plan for Access to U.S. Government Information. The strengths and weaknesses of The Plan, along with GPO’s Federal Information Preservation Network (FIPNet) will affect the discard Policy. We’ll have more to say about both here soon, too.
- The Policy actually allows the discard of any “tangible” depository materials (print, CD-ROM, microfiche). GPO has not provided any data about how much if any data on CD-ROMs or on microfiche is, or will be, in FDsys. We will simply refer to “paper” instead of the awkward, somewhat misleading term “tangible.”
- The policy states that its “purpose” is “To allow regional depository libraries the option to discard certain tangible materials and provide permanent public access to the digital version available on GPO’s Federal Digital System that meets the standards of the Superintendent of Documents as authentic.” Before the Policy was approved, Regionals already had the ability to “provide access” to digital versions of documents that are in FDsys. So, the only change that the policy makes is allowing Regionals to discard documents that are already in FDsys. The policy will not add documents to FDsys or enhance or augment the ability of Regionals to “provide access” to documents through FDsys. It has been GPO’s continuing intent to “provide permanent public access” to digital documents since at least 1996. (See for example: the 1996 Study To Identify Measures Necessary for a Successful Transition to a More Electronic Federal Depository Library Program, the year 2000 GPO Annual Report, and the 2001 GPO Access Biennial Report To Congress On The Status Of GPO Access.)
- The objectives have appeared in the draft proposal, the letter to JCP requesting approval of the policy, and its presentation slides describing implementation of the policy.
- “These goals are impaired when space limitations result in the relocation of tangible collections to offsite storage…” (Vance-Cooks, letter to JCP. July 10, 2015).
- This is a reasonable criterion. Such research has been done to rationalize and inform the mass-discarding of academic journals. Those who have conducted and relied on such research have noted that existing research does not apply to more complex kinds of collections such as government documents. GODORT’s recommendation was a prudent step of due diligence to ensure that the policy will not cause the loss of documents in FDLP care. Since the discard policy will enable libraries to make non-reversible decisions it is reasonable to base that policy on evidence that indicates discarding will not result in loss of information. See, for example, Yano: Optimizing the Number of Copies for Print Preservation of Research Journals and Schonfeld: What to Withdraw: Print Collections Management in the Wake of Digitization and Nadal Scarce and Endangered Works.
- In answer to a question about access paper copies, Cindy Etkin said that if the users of a Selective Depository need “tangible” copies, then that library should not discard its paper copies.
- One presentation slide says that “preservation copies” can be used as a “Source of tangible facsimiles” and as a “Source for new digital copies.”
- Conway, Paul. “Preserving Imperfection: Assessing the Incidence of Digital Imaging Error in HathiTrust,” 2013. See also: “An alarmingly casual indifference to accuracy and authenticity.” What we know about digital surrogates.
- For example, a digitized version of this 1986 Senate Hearing is in FDsys, and the PDF format is digitally signed by GPO. Presumably it is eligible for discard even though the PDF version has no digital text and the “text format” is only the title page: Sioux Nation Black Hills Act : hearing before the Select Committee on Indian Affairs, United States Senate, Ninety-ninth Congress, second session, on S. 1453, S. Hrg 99-844. July 16, 1986, Washington, DC.
- See: Burger, John. 2013. Federal Documents as Content for DPLA? Association of Southeastern Research Libraries (ASERL), DPLA Content and Scope Workstream; and Petersen, R. Eric, Jennifer E. Manning, and Christina M. Bailey. 2012. Federal Depository Library Program: Issues for Congress. R42457. CRS Report for Congress. Washington, D.C.: Congressional Research Service.
- The OAIS standard famously says that the “data” to be preserved in an OAIS repository can include “a sequence of bits, a table of numbers, the characters on a page, the recording of sounds made by a person speaking, or a moon rock specimen.” Consultative Committee for Space Data Systems. 2012. Reference Model for an Open Archival Information System (OAIS). Magenta Book, issue 2. Washington, D.C.: Consultative Committee for Space Data Systems.
James A. Jacobs (UCSD)
James R. Jacobs (Stanford)