Populism, Digitization and Plenty: An Online Film Archives at 15
Talk delivered at the Shaping Access! conference, Berlin, November 13, 2014.
Published on: Mar 4, 2016
Transcripts - Populism, Digitization and Plenty: An Online Film Archives at 15
Populism, Digitization and Plenty:
An Online Film Archives at 15
Prelinger Archives / Internet Archive
University of California, Santa Cruz
Berlin, November 2015
Thursday, November 13, 14 1
Thursday, November 13, 14 2
It was 1982, and we were in a period of media transition.
Film was giving ground to video, and the world was filled with unwanted film. Producers and
distributors were quietly going out of business.
I had been a documentary film researcher and realized how much had never been collected by
I began to collect ephemeral films -- advertising, educational, industrial and government films, and
home movies. Established archives were remarkably passive in this area; in fact, the US is such a
media-rich country that no one seemed to care if film was lost.
Thursday, November 13, 14 3
I organized an unofficial, private archives and funded it by charging for access to our collection as
stock footage. This was considerably easier to do in the US than it would have been in Europe,
because we have a rich public domain and no serious moral rights laws. Business was good, and
many producers were interested in using this material that depicted US life, culture, landscape and
industry in such a fascinating way.
Thursday, November 13, 14 4
In 1999 I moved from New York City to San Francisco, and soon came into contact with Brewster
Kahle, philanthropist and founder of Internet Archive. He challenged me to put my archives online
for free. As a New Yorker, I'd never heard of the open-source movement, and I thought information
wanted to be expensive. So I argued with him.
But it seemed unfair that so many historical resources could only be accessed by a few people. We
agreed to move ahead, though I was very conscious this was an experiment. As my income and
wellbeing was almost completely derived by selling access to our collection, I had my doubts. But as
a contrarian, I had to move ahead.
Thursday, November 13, 14 5
I proposed to Brewster that we begin with 1001 films. I won't describe my curatorial decisions now,
but I could later. Internet Archive subsidized film-to-videotape transfer, hosting and bandwidth.
This was my first lesson in one of the quieter truths behind many open-access projects: peel back
the layers, and quite often you will find some kind of subsidy. Here is the online archive at 3
months, with 750 films.
After 14 years, our collection has grown to over 6,400 items. More are coming. A study of download
figures for our collection alone, plus items mirrored elsewhere, seems to indicate that around 100
million films have been accessed.
In total, Internet Archive offers a total of 1.8 million video items plus clips from 639,000 TV news
broadcasts since 2009.
Thursday, November 13, 14 6
When I visit Europe, the first, second and third questions I get are always about copyright. So let me
try to anticipate these so we can move beyond this subject.
U.S. copyright law differs from European in a number of key areas. First, there are many more works
that fell into the public domain. I would estimate there are perhaps 400,000 to 500,000 motion
pictures that are out of copyright. Second, moral rights plays a much smaller role in American law,
so we are not restricted from presenting edited clips or reformatting films. And third, if one of our
contributors uploads copyrighted material in error, we have the DMCA.
Thursday, November 13, 14 7
Most of the films I collected are not currently in copyright, so I had few concerns about digitizing
them and putting them online.
We HAVE put some orphan works online. Orphan works, for those of you who don't know, are works
under copyright whose author is unknown or cannot be located to grant permission for reuse.
Thursday, November 13, 14 8
Home movies are a special issue -- the quintessential orphaned works.
Just about every home movie is an unpublished work and remains under copyright. Who owns rights
is often a mystery. I have permission to reuse some, but we have some 13,000 items, and I cannot
morally accept the idea of keeping an entire area of culture from public access. So except for some
sensitive material, I've started to put our home movie collections online.
Thursday, November 13, 14 9
What kind of access model did we adopt?
As everyone who shares resources is, we were immediately faced with having to come up with an
actionable definition of "openness."
This was my first experience with open digital content, and I have to confess I gravitated toward a
restrictive model. You can see it above.
This was inexcusable, but I was also trying to comply with our representation agreement with Getty
Images. In a few months, we dropped all of these restrictions.
Thursday, November 13, 14 10
And in late 2002, we adopted the CC license. This has dropped out of the official history (like David
Bollier's book), but we were the first significant cultural collection to adopt the CC license. CC urged
us to use a special CC Public Domain declaration that allowed for the fact that we did not own rights
to most of the material we licensed, but declared it to be in public domain.
This is what our statement of rights looks like today -- it couldn't be more permissive or
Two-tier licensing system
• Creative Commons licenses
• user does own research
• tapes (later files) (high-res)
• specific written agreements
in producer’s name
• user gets research services
• lots of human intervention
• value added
Thursday, November 13, 14 11
I think you know this model, which some people call "freemium."
Thursday, November 13, 14 12
How would this affect our existing stock footage business? We did not know, and it turned out to be
a very interesting question. We believe it significantly increased our revenue in a time when license
fees were decreasing and the market for historical material was challenged. On top of that, it helped
us maintain the Prelinger brand name within the Getty Images portfolio.
best estimate: 62% increase
Thursday, November 13, 14 13
Brewster and I built the first webpage linking to the digitized film files on Saturday, December 28,
2000. Here is my not-so-beautiful interface, which I generated completely in FileMaker. On Tuesday,
January 2, 2001, we looked at the directories and found a number of unfamiliar-looking objects. As
it turned out, we'd left the write permissions unlocked and some users had contributed several
episodes of SOUTH PARK. While we saw this as an altruistic act of sharing, we were also worried that
someone might try to close down our emergent service, so I deleted the files and changed the
perms. This file "donation" tells me that some of our more sophisticated users construed our
collection as a hospitable destination (in the way that South African archivist Verne Harris speaks of
Jacques Derrida's sense of hospitality in the archives) and as a node of participation in the YouTube
Thursday, November 13, 14 14
While this would not last, IA then had user logs.
-- government agencies
-- universities, especially dormitories
-- and a few users at home with broadband. At that time the statistics alleged that 19% of homes
had broadband, but I don't think that was really true.
mpeg-2 was 28MB/minute
The site was a great success. it was the first public collection at Internet Archive (before 9/11 TV
News archive and Wayback Machine).
Students, makers, fans, community producers, activists, artists, scholars, teachers, etc. all picked up
Thursday, November 13, 14 15
But for archivists, this was a new idea. I think to some it was anathema.
Archivists were suspicious about:
building a large collection for access only
digitizing without first preserving the film and presenting video in non-archival formats (too low-res)
building an archives site without metadata that conformed to standards
propagating material in what were then high-res formats -- giving too much away
We were therefore criticized for low-res as well as for high-res
Some were peeved that we didn't restrict to streaming but encouraged downloading
Some archivists were concerned with letting material go without restrictions -- that it meant a loss
of control. But if you can't release your holdings, then you've REALLY lost control.
In any case, we continued, and now it is 14 years later.
Thursday, November 13, 14 16
Some significant issues I wanted to mention.
Accidental artwork, Yorkshire Sculpture Park, 2007
Thursday, November 13, 14 17
How persistent is this collection? Will it last?
We are exposing and storing copies for access. These are video or scanned copies of film originals,
not original materials themselves.
However: just as the films back up the scans, the scans back up the films. it is possible we'll lose
I do not take persistence for granted, and to me it's axiomatic that we will digitize film more than
once in its lifetime (I'm already working on a second go-round for some film).
Thursday, November 13, 14 18
Metadata, as always, is a problem. I didn't wait for all the metadata to be perfect before putting
records up, and I also did not build a mechanism to maintain it. I hoped the community would step
in, and for the most part they have not. The annotations/commentary/reviews are of uneven quality.
There is some hate speech, and lots of inaccurate info regarding films. To be sure, there is great
information as well. Fan culture is wonderful, but it is not infallible. Who will take the time to add,
improve and maintain metadata? That's unclear.
It is time-consuming. Most of us cannot generate (or research) metadata that is up to scholarly (or
even fan) standards.
Thursday, November 13, 14 19
We have only a single portal. This has a bias towards fan appreciation. While thousands of
educators, researchers and scholars use the site, they are almost never present as annotators or
reviewers. We could use an education portal, and maybe a scholarly portal into the collection.
Thursday, November 13, 14 20
Physical preparation of material for digitizing has always been a challenge. For me, it's the #1
roadblock to sharing materials online. It is expensive and requires trained people. I tried to crack
this nut by convening a group of emerging and would-be archivists who volunteered in exchange
for training, and it sort of worked, but it would be really difficult to do it at scale.
Thursday, November 13, 14 21
Putting digital collections online for public access is inherently a retail business, and one needs to
be prepared to offer some kind of support. Almost fifteen years later, I still get emails asking me
how to download. There are still many people who (1) don't realize that downloading might meet
their needs better than playing through a browser or (2) don't know how to right-click or control-click.
Thursday, November 13, 14 22
How do we want to think about copyright as laws evolve?
ECL is a real problem. The regularization of orphan works concerns me as well. Melissa Terras is
doing a great series of blogposts about the new UK regime. Who will these collective licensing
organizations be, how much will they cost to administer, and who will get the money? And why
should individuals go through the same process that a major publisher or studio must go through?
Could we have built this site under ECL or a regularized orphan works regime?
And as cultural custodians, we're obliged to take the long view, which means we cannot postpone
thinking about issues that last longer than copyright and pose more profound questions. For
instance, how can we temper openness with respect? Questions of indigenous or community cultural
and intellectual property rights, and the moral rights of creators, all pose issues that go far beyond
the bounds of copyright. These issues will outlive shorter-term discussions and are part of shaping
the kind of world we hope to live in.
Measures of success
Thursday, November 13, 14 23
Was this project successful? Well, we don't have objective metrics or scientific criteria to judge. I've
proposed in the past that we borrow a page from the environmental sector and require impact
statements for digitization projects. We might ask: How might a project affect the condition,
preservation status and longevity of the original materials? What kind of impact would access to
these materials have on the cultural and social spheres? But we don't yet do impact statements.
Instead I need to talk about outcomes and apply flexible criteria; ultimately you will be the judges.
The accessible archives:
Treats access as a key part of its mission, not an afterthought
Reconfigures its workflows to expand access and use
Limits access to collections only as required by law, respect, custom
and unavoidable constraint
Makes materials available before they're requested
Measures value by consumptive use
Seeks out new users
Brings archives into the community and community into the archives
Sees archival activity as a civic function
Builds transactional spaces
Avoids being hobbled by the precautionary principle
Thursday, November 13, 14 24
I don't want to repeat the arguments for making archives more accessible; I would hope we're beyond needing to
be convinced. But here are some of what I think the attributes of an accessible archives might be. Our online
experience led us in these directions.
Thursday, November 13, 14 25
Specific to our project, I CAN say a few things.
We made original historical films available to a very broad public, who used them in traditional and
quite unorthodox ways. This had never before happened.
We caused some 100 million films to be downloaded or viewed. This is striking given the obscurity
and specialized character of many of the works.
We played a role in lowering use fees for public domain material. This might be a double-edged
Thursday, November 13, 14 26
We influenced countless scholars who otherwise might never have considered incorporating moving
images into their research. The films have propagated into hundreds of papers, dissertations, and
We also strongly influenced the development of a fan culture focused on industrial and educational
-- mention MST3K
We played a key role in helping move ordinary film, useful film, quotidian film from the cultural
periphery towards the center.
broadening documentary aesthetics
youth and newer Americans
centrality of archives
Thursday, November 13, 14 27
Less tangibly --
We increased the sense of entitlement with which people regard the Internet and moving image
archives generally. We replaced a model of scarcity with a model of plenty, as Lawrence Lessig used
to put it in the early 2000s. We gave people a sense of how a gift economy might work.
We helped to expand makers' sense of how archival materials could be used, beyond the
overdetermined and historicized ways they are used in documentaries
We collected a decently large chunk of U.S. cultural and social history and made it available for
commentary and remixing by younger people and newer Americans
Along with the other Internet Archive access projects, I think we helped push forward a new sense of
the centrality of archives in the world. To engage in archival activity is to intervene in history's flow. I
think many more people understand the importance of archival work.
We helped to move the figure of the archivist from the margins towards the cultural center. We
helped to make archivists and archival practice cool.
Thursday, November 13, 14 28
And we had a band named after us.
Thursday, November 13, 14 29
As to the future:
We felt a little lonely when this project started, and I have to say we still feel a little lonely. While
many cultural digitization projects that offer one or another flavor of open access now exist, most
moving image collections are not online. There is a kind of exceptionalism that seems to apply to
films. Or if they are online, they're not available for most kinds of use. Viewing-only access is NOT
open access. We need to be thinking of moving images and sound as we think of code -- as objects
that users are able to freely access, download, modify and manipulate. This is especially true when
we realize that many of the "audience" members for moving images in the future will be machines,
"The objective of archival policy in a
democratic country cannot be the mere
saving of paper; it must be nothing less than
the enriching of the complete historical
consciousness of the people as a whole..."
Robert C. Binkley, 1939
Thursday, November 13, 14 30
I realize that we won't be able to offer free access to all cultural resources. And yet, as we enumerate
the reasons for cultural enclosure and make our excuses as to why certain materials are not publicly
available, we have the responsibility to specify (as clearly as we can) why we can't release, how and
when they will be released and by whom. If we can't offer openness, we must at least plan for it.
Thursday, November 13, 14 31
Projects like ours have helped to mainstream archival activity. Because today the spread of digital
culture has rendered archival activity universal. To quote archivist/provocateur/poet Kenneth
Goldsmith: "The advent of digital culture has turned each one of us into an unwitting archivist....The
ways in which culture is distributed and archived has become profoundly more intriguing than the
cultural artifact itself. What we've experienced is an inversion of consumption..."
Thursday, November 13, 14 32
We may not survive projects such as YouTube. Now, I am a tremendous fan of YouTube. It is hugely
paradoxical. It is an amazing repository that offers a hit for almost every query, that permits anyone
to upload their work and let it intermingle with the work of their cultural heroes. It's become a
magnet for personal media, filling the persistent gap between the world of massive personal media
production and the archival ecosystem that hasn't collected it. No established archives will ever
serve as many videos to as many people. It appropriates the eyes and ears of the public by offering
easy access to subsets of cultural expression unavailable elsewhere without great effort. It enables
countless instances of media and multimedia authorship & remix that legacy archives never wanted
to enable and could never have enabled on their own. But we might describe its relationship with its
users as a noncommittal handshake. Archival persistence, digital longevity and resistance from
outside interference are traded in for the appearance of openness, an absence of latency, and an
omnivirous collecting policy curated, if curated at all, by machines that know not to expose certain
videos to Chinese IP addresses. Just as we are willingly exchanging the resiliency of venerable
copper-wired telephone networks for the stimulation of app-driven but unreliable wireless
smartphones, we have exchanged the traditional archives for the apparent archives, gaining an
appearance of completeness that is in fact full of gaps. And I doubt established institutions will ever
be able to catch up with them or with the other massive commercial collections that will follow.
Detroit, February 2011
Thursday, November 13, 14 33
I end with a thought about Europe and America. We do a few things differently.
Our project was not the work of a government agency; it was supported by Brewster Kahle, a
philanthropist who wanted to help bring moving images to the Web six years before the launch of
YouTube. While Americans aren't very good at doing the great, national-level digitization projects
you do in Europe with the help of cultural ministries, we are great at bottom-up, entrepreneurial
initiatives. You work out the issues with stakeholders and standards; we are ready-fire-aim. We are
more disposed to question copyright maximalism; in Europe it often seems to be beyond question. I
have no opinion as to which is better, but I strongly believe this is the time for each of us to learn
from one another.
Thursday, November 13, 14 34
This talk is online at
Thursday, November 13, 14 35