Prepared remarks of Archivist of the United States David S. Ferriero, “Losing Our Memory,” as part of the Provost’s Lecture Series, Duke University
March 22, 2010
Delighted to be here and to participate in this year’s series, "The Future of the Past, the Future of the Present: the Historical Record in the Digital Age."
The provost, in his letter of invitation, puts forth an intriguing syllogism. He writes, "We live in the age of the archive. Our ability to capture text, video, audio, and electronic communications is unprecedented. This new power has also raised questions about security, privacy, relevance, access, selection, cost, and long-term preservation. In the digital environment, everything is saved yet little is preserved."
Now, during my academic career, I made it a point to never argue with the provost, but implicit in this description are several ideas about the changing nature of information, how it is distributed, and how it is persistent. And it is at this transformational point—the digital age as it is often called—that calls the questions: what can be done about losing our memory? And what is the right role for archives at the federal, or national, level as well as at institutions such as Duke University.
Let’s first look at the changing nature of information. Back in 1997, David Shenk published about information overload called Data Smog, which has entered the parlance as a short-hand description of the sheer volume of information each of us deals with every day. For example, Shenk wrote that in 1971 the average American was targeted by at least 560 daily advertising messages. Twenty years later, that number had risen six fold, to 3,000 messages per day. (For those who are counting, that’s one million per year.)
But this was written in 1991. Before the rise of email, twitter, IMs, Facebook, YouTube and so on. Who knows how many messages we are exposed to each day?
Contrast our experience with that of our parents and grandparents, who were limited by access to information—the morning and evening newspapers, the mail—and by the speed of communication.
While none of us, I would guess, want to go back a century and live without the extraordinary tools and communications delivery systems—off the grid as they say—we must acknowledge that the data smog requires some filtration. The best filter is going to be our own individual capacity to turn off and tune out occasionally, but we need as well as information filters such as search engines, libraries, museums, and archives.
Not only does information become more pervasive through multiple distribution systems, but—thanks to us—it becomes more persistent. That is to say, over the past century and a half, historians and archivists have done an excellent job in demonstrating the value of historical records of all sorts. We have created the age of the archives—the ways and means to save more stuff and for longer periods of time.
The age of the archive corresponds with the rise in America of "high" culture. In his seminal book Highbrow/Lowbrow, Lawrence Levine traces the emergences of cultural institutions—first in communities, and later on the national level—designed to collect and preserve art and artifacts. The great museums and libraries and historical societies—and yes, archives—developed as the country grew and matured. As Levine argues, one purpose of these institutions was to save knowledge and information for its own sake, its intrinsic value.
Government followed suit and acknowledged the need to capture and retain knowledge and information, and more recently corporations and business began to better understand, value, and exploit historical information. Perhaps no better example exists than that of the Disney Corporation, which so values its history that it was instrumental in getting Congress to extend copyright. Save Mickey Mouse. We might be able to sell him again.
As a society, we have become very adept at saving stuff. Virtually every business and corporation has some records management system, and archives are ubiquitous at all levels—from colleges and universities to governments large and small to NGOs of all stripes. Shown the value of something as simple as a baseball card or comic book, mothers no longer toss out the paper ephemera of childhood, but encase it in plastic and stash it the safety deposit box.
Information has become extremely persistent at all levels. Everything is saved yet little is preserved.
That presents these basic challenges for archives:
- How do we deal with the scale of number of records created?
- What is a record in this day and age, and how do we handle the kinds of records that we now create?
- And finally, is saving it all enough? Or what are the purpose(s) of an archives?
A century ago
These, of course, are not new questions. A century ago, the American Historical Association was advocating with the Congress and President Roosevelt—Teddy Roosevelt—for the creation of a National Archives and a National Historical Publications Commission. The AHA had been focused on federal government records since its founding in 1884, and for decades, the association and in particular J. Franklin Jameson, who at the time was with Department of History and Political Science at Johns Hopkins, badgered and cajoled and argued for some system and order to be created to preserve federal records and make them accessible to the citizenry.
Jameson wasn’t the first advocate, though perhaps, he was the most ardent. The history of the idea of a national archives is as old as the nation itself.
When the Constitutional Convention of 1787 had completed its work, delegate Rufus King of Massachusetts suggested that the Journals of the Convention be either destroyed or deposited in the custody of the President. He thought "if suffered to be made public, a bad use would be made of them by those who wish to prevent the adoption of the Constitution." The Convention voted to entrust the records to George Washington with instructions that he retain them "subject to the order of Congress, if that body was ever formed."
In 1796, Washington turned them over to the new Department of State. The exceptions, of course, were the Constitution and the Declaration of Independence, which were sent to the Congress in session in New York. It was not until 1800, when the seat of government moved to Washington, DC, that those charters joined the other records at the Department of State.
In 1810, Josiah Quincy of Massachusetts, recognizing the risk associated with the growing volume of documents of the government scattered around the city, told the House of Representatives that "the public records of the country were in such a situation as was disgraceful to the House and to the Nation."
Not only were they in disorder, and in a state of decay, but all the records of the Revolutionary War lay under the eaves of the building in a condition extremely unsafe, and daily exposed to destruction by fire." In fact during the fire of 1801, files of the War and Treasury Departments were destroyed and President John Adams himself joined the bucket brigade to quench the flames.
In true government tradition, a committee was formed and a study conducted and a bill with a $20,000 appropriation was signed by President [James] Madison to construct "as many fireproof rooms" as were necessary to house the archives of the Government. Before the law could be enacted, the British army invaded with instructions to burn the city. Clerks in the State Department working through the night packed the records of the government into linen sacks and loaded them onto creaking carts to be carried off into the hills of Virginia. The next morning, August 25, 1814, the Capital and other public buildings were in ashes. When the British left Washington, the records were returned to their former home and languished for another century.
From that point till just after the Civil War, federal records were still relatively under control. Some 108,000 cubic feet were in storage within various federal agencies in 1861, but that number grew to 1 million cubic feet by 1916. Just to give you a rough idea, there may be between 2,000 and 2,500 pages of material in a cubic foot. So as early as World War I, the federal government had some 2 billion pages under the control of several agencies. The call for centralization through a National Archives persisted.
In 1926, Congress finally acted and authorized the construction of the building, which began in September of 1931. President Herbert Hoover laid the cornerstone in February of 1933, though there was not yet an actual agency to be put in the building. It was not till the following June that Congress passed and President Franklin Roosevelt signed the "Act to Establish a National Archives of the United States Government."
The Act generally gave the National Archives the scope and authority to preserve the records of the federal government, to "make regulations for arrangement, custody, use, and withdrawal," to acquire and preserve motion pictures and sound recordings, and to publish and make available historical works and collections of sources.
At the same time, however, nothing of this sort had ever been done in the United States, leaving the new staff at the archives unfettered in devising systems and means to bring the information and material under control. These must have been high times at the agency, times of tremendous freedom and opportunity.
The head of the enterprise was Robert Digges Wimberly Connor, the first Archivist of the United States. Dr. Connor was a native of Wilson, North Carolina, a graduate of, and history faculty member at, the University of North Carolina. And I'd like to think that if Duke had been founded when Mr. Connor was going to school, he would have opted for Duke.
Connor had commissioned a study of the state of the government’s records during the two years before construction was complete. Describing the situation from the perspective of the researcher he writes:
...conditions make it impossible for officials to find adequate room for both their files and their staffs. Few facilities can be furnished the student and his presence is tolerated rather than encouraged by staffs already sufficiently burdened with the routine duties of the day. He finds the records he desires to use scattered throughout the country, stored wherever space can be found for them, in cellars and sub-cellars, under terraces and over boiler-rooms, in attics and corridors, piled in dumps on floors and packed into alcoves, abandoned carbarns, storage warehouses, deserted theaters, or ancient but more humble edifices that should long ago have served their last useful purpose. Typical is the case of valuable records relating to Indian Affairs which were found in a depository in Washington piled on dust-covered shelves mingled higgledy-piggledy with empty whisky bottles and with rags and other highly inflammable trash. In another Washington depository packed with documents the most prominent object which meets the eye as one enters the room is the skull of a cat protruding from under a pile of valuable records. I think it is a fair question that if a cat with nine lives to risk in the cause of history could not survive the conditions of research in the depositories of government records, can we justly blame the poor scholar who has only one life to give for his country if he refuses to take the risk?
Connor and his colleagues created systems for records management, accessioning and processing, preserving, and encouraging use. They cleaned out the whiskey bottles and put out the cat. They brought order to the chaos, just as the storm of records grew into a tempest.
As early as December 1935, records began to arrive at the Archives building, and by June 1941, over 300,000 cubic feet were on file. Eight years later, the space in the building was exhausted, and by the mid 1940s, one million cubic feet of Federal records were being created each year. "It is almost inconceivable," U.S. Archivist Wayne Grover observed in 1953, "that the federal government, in the 22 years from 1930 to 1952, should have created more than seven times as many records as it did during its previous 155 years of history."
You ain’t seen nothing yet, baby. The numbers are even more daunting today. As you know, the National Archives keeps only those Federal records that are judged to have continuing value—about 2 to 5 percent of those generated in any given year. By now, they add up to a formidable number, diverse in form as well as in content. There are approximately 9 billion pages of textual records; 7.2 million maps, charts, and architectural drawings; more than 20 million still photographs; billions of machine-readable data sets; and more than 365,000 reels of film and 110,000 videotapes.
We outgrew the building in Washington, DC, long ago and now have a second Archives that’s also filling up, records centers across the country, 13 Presidential Libraries, and underground centers—all filled with historical records that have been saved. Tangible stuff. Before I get to the challenges of electronic records, I’d like to say a few words about the lessons we can learn from dealing with this sea of paper.
Space: the first frontier
Once upon a time, in the antediluvian days before the Internet (imagine such a time, boys and girls), historical records primarily were written on paper. Anyone with a filing cabinet or desk drawer knows just how quickly paper takes up space. For the National Archives, space—or the lack of it—was the first frontier.
The first approach to the lack of space was to create more of it. Necessity became the mother of retrofitting existing space within the National Archives building. A kind of warren exists in the stacks, half-floors and secret passageways that may well contain a secret cat or two. [Reminiscent of 28-48 Perkins.] More buildings were built.
At the same time, engineers realized that another solution would be to make the paper itself smaller. Microfilm. Archives and libraries began using microfilm in the mid-20th century as a preservation strategy and as a space-saving measure. In his 1945 book, The Scholar and the Future of the Research Library, Fremont Rider, the director of the Olin Library at Wesleyan University, calculated that research libraries were doubling in space every sixteen years. His suggested solution was microfilming.
Compression is obviously a great space-saving idea. Microfilm also allowed for access to fragile materials and is a stable archival form. Preservation-standard film has an estimated life of about 500 years under appropriate storage conditions. It is an analog version of an analog original, requiring no software to decode the data. One only needs magnification.
Problem solved. Raise your hands if you prefer sitting at a microfilm reader over sitting in front of a computer screen. The disadvantages of microfilm are also fairly obvious. One needs magnification, usually on a difficult-to-use machine, requiring careful winding and rewinding. And often the quality is poor, especially in terms of photographs and color. Nicholson Baker raised that point in his book Double Fold, which quite rightly took on the shame of microfilming newspapers while discarding the originals, and the resultant loss of information. Furthermore, like all analog media formats, microfilm is lacking in digital’s ability to be indexed and searched easily.
The Museum of Obsolete Technology
If the future of microfilm is in some dispute, then what can be said about some truly obsolete technologies. How many archives have their own hardware morgues or Computer Museums, where machines are kept on hand able to read floppies in 8-inch, 51/4 -inch or 31/2 -inch formats?
At Archives II in College Park, Maryland, we have our own museum in the Department of Special Media Preservation. Here you’ll find a recording device that uses coils of thin steel wire instead of tape. There are 70,000 18-inch glass discs —each with two hours of enemy radio broadcasts from World War II. They play on a Memovox. There are push-pull movie sound-tracks —1800 reels. There are a quarter million optical discs—the cutting edge technology of the 1980s—that depend on software and hardware no longer on the market. All of these technologies are less than a century old, and yet, the materials may be gone.
While the quantity of information saved has increased dramatically over the centuries, the durability of media has decreased exponentially. That is to say, the clay tablets from ancient Sumeria can still be seen today, medieval manuscripts on animal parchment are perfectly readable, and paper correspondence from the Renaissance is still in good condition. But the floppies of our recent past are all but useless. What is the shelf-life of an 8-track tape? Can anybody here remember WordPerfect? FoxPro? Netscape Navigator? Where have you gone, MS-DOS?
But, as Alexander Stille points out in his book The Future of the Past, "books printed on modern acidic paper are turning to dust. Black-and-white photographs may last a couple of centuries, while most color photographs become unstable within thirty or forty years. Videotapes deteriorate much more quickly than does traditional movie film. And the latest generation of digital storage tape is considered to be safe for about ten years, after which it should be copied to avoid loss of data."
The lessons for the digital age
The challenges of paper records and those on obsolete media pale in comparison with the whole question of electronic records. They answer the question of "space" by compressing data into ones and zeros, but at the same time—and perhaps as a consequence—electronic records are even more likely to be saved than paper records. They answer the question of technological obsolescence by being hardware independent, but at the same time, electronic records are just as likely to be lost if maintained solely in their proprietary formats.
The National Archives long ago recognized the need to develop a means of dealing with the data smog. Perhaps the largest undertaking in the history of the National Archives is the creation of the Electronic Records Archives—or ERA. Established in partnership with the private sector and developed using the best available research from around the world, ERA is intended to preserve and provide access to any type of electronic record created by a federal agency. It is intended to handle any changes in software and remain a collaborative laboratory with the public to provide access to electronic records.
Progress has been slow, but steady. Records are being ingested into the system. Just this past fall, 77 terabytes of presidential records from the George W. Bush administration were entered into the system. And e-mail from the Bush White House was converted from Microsoft's Exchange server format into an "open source" format, in order to make all the e-mail messages of administration staff searchable. We look forward to a day when data miners and historians can have at this trove of information.
We must be mindful of the special needs of electronic records, just as for paper and special media, which require both access and protection, the preservation and conservation routines which care for the physical object to extend its life, usually through providing an appropriate storage environment. And we must be on guard against unintended consequences brought about by conversion from one medium to another.
And we must be prepared for the new ways in which government and citizens interact and the new kinds of records created through social media. Two years ago, President Obama and his campaign staff began the first truly Internet campaign, and now the Administration is using social media and new communications techniques to reach out to citizens in ways once inconceivable, especially from the federal government.
We are following suit. The National Archives first tweeted in the spring of 2009. It has web pages for most of the Presidential Libraries, several regional centers, the Federal Register, the National Historical Publications and Records Commission, and the Foundation for the National Archives. We are up on Flickr—with some 360,000 visitors, and we are on YouTube. One day soon, I predict, we will have the first Social Media Archivist, whose job it is to figure out how to tame the dragons we have set free.
Web 2.0 is a wild and crazy place. Let’s just look at just one part of the new social media dynamic: Facebook. Some 300,000 businesses in the U.S. are on Facebook, including the Federal Government. The White House has nearly 500,000 fans, the National Guard nearly 400,000, the U.S. Army 200,000 and the Centers for Disease Control over 54,000. Facebook is spreading quickly. It took radio 38 years to reach an audience of 50 million; television reached the saturation point in 13 years; the Internet in 4 years; and the iPod in 3 years. In less than 9 months, Facebook added 100 million users. QZone—the equivalent of Facebook in China—has 300 million users.
More than 1.5 million pieces of information are shared on Facebook each day. Generation Y—who just this year blew past the Baby Boomers in terms of overall population share—uses this stuff the way our generation relied on the daily mail and the newspaper. The whole commerce of information has changed permanently. The transactions between organizations—including the federal government—and users will continue to spread and grow. We are struggling to come up with the right framework for preserving the most basic records for the short term, and we need to be good stewards of this new dynamic.
We have a lot of federal records at the National Archives, and I suspect that despite our best efforts the number will continue to increase exponentially. Do you know the sign on the McDonald’s restaurants? Eighteen years after the first one opened, they put up a sign that said over 100 million hamburgers served. Fifteen years later, they were up to a billion, and now they’ve simply stopped counting and just say billions and billions served. I suspect we will stop counting someday as to the numbers of records saved. Certainly if you look at the national network of archives—the records at the federal, state, local and academic levels—you would count well into the billions. Electronic records are only going to blow the top off.
Making it easy for researchers, students, and the general public to learn about and make use of the billions of items in our collection is clearly a challenge. Free and open access to the records of government will always be the work of this agency. Exhibits, classes, lectures, and digitization activities all contribute to our mission of encouraging the use of the records of government and exciting the next generation of historians to become a Michael Beschloss, or an Anne Firor Scott, or a T.J. Stiles or a Drew Faust, to choose a life of scholarship.
And to help everyday Americans better connect with their government. Let me share some stories with you about ordinary extraordinary moments in the Archives:
A man walked into a regional center with a letter dated May 1946, recommending his dad for the Bronze Star. The medal had never been awarded, and the son wondered this was an oversight or had the recommendation not been approved? Staff at the St. Louis Military Personnel Records Center made this case a priority and found additional documentation. Through our efforts, it was determined that he was entitled to the Bronze Star. Just two days after his 100th birthday, and 63 years after the recommendation was written, in a ceremony arranged by the National Archives regional office, local Army officials presented Walter Pierce with the Bronze Star.
Last February in Alaska, the granddaughter of an 88-year-old Anchorage resident visited to obtain a certified copy of his 1958 divorce decree. Prior to Alaska statehood in 1959, the Federal U.S. District Court handled divorce cases, so the file was in federal hands. We quickly found the final decree, and "made her day." Her grandfather in the nursing home was waiting for proof of his divorce . . . so he could remarry.
In New York, a patron contacted us looking for a relative who she and her entire family had believed perished in the Holocaust but apparently survived, although his wife and baby did not. Our records confirmed that he came to the U.S., remarried and named his new baby girl after the little girl who had died.
All over the country, archives are reuniting citizens with their rights and helping historians research and tell the American story. Through the work of the National Historical Publications and Records Commission, we’ve helped fund documentary histories of Americans from Benjamin Franklin to Martin Luther King, from the records of the Florida everglades to the civic history of Los Angeles.
Archives play such a vital role in our lives. President Franklin Roosevelt, in dedicating the FDR Library in Hyde Park, said it best:
To bring together the records of the past and to house them in buildings where they will be preserved for the use of men and women in the future, a Nation must believe in three things. It must believe in the past. It must believe in the future. It must, above all, believe in the capacity of its own people so to learn from the past that they can gain in judgment in creating their own future.
Among democracies, I think through all the recorded history of the world, the building of permanent institutions like libraries and museums for the use of all the people flourishes. And that is especially true in our own land, because we believe that people ought to work out for themselves, and through their own study, the determination of their best interest rather than accept such so-called information as may be handed out to them by certain types of self-constituted leaders who decide what is best for them.
The Archives needs to continue to be above the fray, interested more in history, democracy, and the protection of both documents and individual rights. To that end, we are involved in several efforts to increase public access.
The Office of Government Information Services, which opened its doors just before I arrived, was established to act as the government’s Freedom of Information Act ombudsman by reviewing Freedom of Information Act activities government-wide and by helping to resolve disputes between requesters and agencies.
And, most recently, on December 29, President Obama established the National Declassification Center within the Archives. This is an inter-agency effort, led by the Archives, to streamline declassification processes, facilitate quality assurance, and provide training for declassification reviewers.
Ultimately, the Center will usher in a new day in the world of access, allowing us to make more records available for public scrutiny much more quickly. To give you a sense of the task ahead, there are now some 2,000 different security classification guides at work in the government and more than 400 million pages of records awaiting declassification and public access by December 31, 2013.
A couple of recent "memory losses" to illustrate the challenges associated with one form of Electronically Stored Information (ESI)—the new legal term of art under the Federal Rules of Civil Procedure, both relating to email.
The case against the Bush 43 White House was recently settled. CREW (Citizens for Responsibility and Ethics in Washington) and the National Security Archive sued over 22 million missing e-mails representing 94 days of business. The suit and the work to recover the missing e-mails uncovered serious flaws in the e-mail archiving system in the Bush 43 first term. In fact, Microsoft was called in to help find electronic messages as early as October 2003, more than two years before the problem surfaced publicly.
More recently, a footnote in a Justice Department report released last month on the crafting of the torture memos disclosed that many of John Yoo’s e-mails has been destroyed. Those missing e-mails are Justice Department "records" and highlight the challenge of the National Archives in ensuring that each agency is creating electronic records capabilities which don’t allow this to happen.
Just last week the National Security Archive awarded its Rosemary Award (named after Rosemary Woods, President Nixon’s secretary) to the Federal Chief Information Officers Council, the senior federal officials responsible for $71 billion a year of IT purchases who have never addressed the failure of government to save its e-mail electronically. A 2008 survey by CREW, in fact, could not find a single federal agency policy that mandates an electronic record keeping system agency-wide, rather a "print and file" system is pervasive. A system that is highly selective and serves as a serious memory loss vehicle.
A bill passed in the House just last Wednesday, H.R. 1387, is entitled the Electronic Message Preservation Act, which gives the Archivist the authority to "promulgate regulations governing agency preservation of electronic messages that are records, regulations which will require the electronic capture, management, and preservation of electronic records."
Tomorrow I will be testifying before the Senate Subcommittee on Federal Financial Management, Government Information, Federal Services, and International Security in a hearing entitled "Removing the Shroud of Secrecy: Making Government More Transparent and Accountable." Testifying with me will be the White House Chief Information Officer, Vivek Kundra, and Chief Technology Officer, Aneesh Chopra. While the theme is Open Government, I will be making several points:
- The backbone of the Administration’s Open Government initiative is effective records management.
- There is a disconnect between those responsible for agency records and the information technology folks in those same agencies.
- The CIO Council, composed of the CIOs of all the agencies, need to work together to develop the IT tools necessary to manage electronic records in a cost effective way.
In addition, I will be sharing the preliminary results of a government-wide self assessment which my agency recently administered that illustrates how far short Federal agencies are in meeting their statutory responsibilities under the Federal Records Act and the E-Government Act of 2002, with particular attention paid to electronic records, especially email and web records. Almost 80 percent of the respondents (90 percent response rate to a mandatory assessment) self identified to being at a moderate to high level of risk.
All of this is tied into a single effort: the Open Government Initiative. In December 2009, the Obama Administration issued guidance to promote new lines of communication and cooperation between the Federal government and the American people through the principles of transparency, participation, and collaboration. Four basic goals are part of the plan:
- Publish Government Information Online — particularly high-value data and through an open gateway for agency activities.
- Improve the Quality of Government Information
- Create and Institutionalize a Culture of Open Government
- Create an Enabling Policy Framework for Open Government
We are working through the mechanisms at the National Archives to embrace these goals and principles, and we will release a first draft of our plan on April 7. At the heart of the plan is a flagship initiative to develop web and data services that are worthy of the American people. This includes:
- a redesigned archives.gov website that is simplified and inclusive of user communities
- an easy, streamlined online public access experience that unlocks electronic and digitized records from sources across the agency
- a strategic approach to digitization at NARA
- social media tools leveraged to accomplish our goals within all sectors of the agency
- an inventory of data sets
- an updated FOIA Electronic Reading Room
- an updated records management website that will be an example for records management throughout the federal government
I am excited about this, for I see openness as one of the more important ways the federal government can ensure the basic rights of citizens in a democracy. You can find out more about it on our website at www.archives.gov/open.
I am reminded of those rights every day. The Charters of Freedom—just around the corner from my office in the National Archives—are a bold reminder of the ideals of the nation, and the ideals of a national archives.
And I am reminded daily of what is at risk.
Are we losing our memory?
According to the Report of the Committee on the Records of Government, a bipartisan federal committee working with the scholars, "The United States is in danger of losing its memory." The federal government and state and local governments have huge quantities of paper records at risk. Historically valuable electronic records compound the problem. So says the Committee Report issued in 1985. Twenty-five years ago. Before the Web really took off. Before the rise of social media.
If the past teaches us anything, it is that challenges persist. A century ago, J. Franklin Jameson and the American Historical Association were crying out for the establishment of a National Archives, for we were losing our national memory. Fifty years ago, the Archivist of the United States was shaking his head in disbelief over the rampant expansion of federal record-making, putting our national memory at risk. Twenty-five years ago, we were in danger of losing it again. Now, I am here to say that the problem has finally gotten serious.
There are huge risks and challenges associated with what archives are doing. Particularly in the digital environment, where everything is saved yet little is preserved. Clearly we need to save better and preserve more (and not the other way around). The National Archives can play an important role in the evolution of the digital archives environment, with ERA as a cog around which much additional research and experimentation can grow, including R&D from the academic community. But we need the best minds at colleges and universities and the private sector working in concert to develop new ways of not simply saving the records, but preserving them and figuring out ways to make sure they are accessible. Without such judgment and collaboration, we are in danger of losing our memory, again.