Library of Congress
"Early Cinema in the National Digital Library"
“Uncle Josh at the Moving Picture Show” (1902, Edison Company) http://memory.loc.gov
Figure 1. “Uncle Josh at the Moving Picture Show” (1902, Edison Company)
The Library of Congress, through its National Digital Library program, has been digitizing many types of items from its historical collections and has made them available on its American Memory worldwide web site. These items include documents, photographs, manuscripts, and motion pictures. There are culled from various custodial divisions of the Library as a whole. The online material together comprises a digital archive referred to as American Memory. It attempts to describe and document the American experience, both our history and our culture. There are currently over 1 million items online and there is free access to these items. Anyone with a computer hooked up to the world wide web can access these items at no charge.
What I would like to concentrate on today since this is a symposium on orphan films are the motion pictures that the Library has online. I will give you some background on the collections, how they are chosen, the methods we have chosen for presentation and for access, and then we'll navigate a little bit through the site.
Lastly, I will focus briefly on the Library of Congress' future plans in terms of digitizing motion pictures, which include establishing a digital audio visual repository. The Motion Picture, Broadcasting & Recorded Sound division of the Library, in conjunction with the National Digital Library, has put nine film presentations online, for a total of almost 600 films. Some examples of the presentations include the films of the Edison Company, films of the Spanish and American war, and our last presentation, films of Theodore Roosevelt. All the films the Library has put online so far qualify as orphan films in that they are silent and that they are out of copyright protection. The process of digitizing these films and making them available online is a way of providing greater access to these orphan films and a way of informing the public in their own history.
Access, and not preservation, has been the chief goal of this digital project. The National Digital Library program desires to provide increased access to the Library's collections as a way of reaching congressional constituents around the country. We are, after all, the library to Congress and our funding comes from Congress. Another important aim is to provide educational materials. Especially for students from six to eighteen years old, an audience heretofore not served directly by the Library. Pilot tests of the American Memory Project reveal that teachers are finding these materials useful for school lessons due to the increased demand.
Another factor of consideration was to select items that were in the public domain and subject to copyright restrictions since the means of distribution was going to be over the worldwide web. The result was that the majority of films we have put online have come from the Library's paper print collection. The films in the paper print collection are mostly 100 or so years old and were deposited for copyright at the Library of Congress as paper positive, much as a photograph would be instead of on celluloid or nitrate. Many of these films are even deposited on rolls. So you have these long rolls of photographs, just like in film, but they are all positive prints. The fact that they were deposited in this manner accounts for many of these films existing at all today, since the original nitrate films would undoubtedly have been destroyed long ago.
Starting in the 1950s, the Library began to re-photograph these paper prints to convert them back into films again. Initially onto 16 mm film and then onto 35mm film. The digital project has been impetus to make more 35mm prints of these films in order to have a superior digital image to work from, and many who have seen them can vouch for the improvement of the 35 mm prints over the 16mm. The 35mm prints are made by the Library's Conservation Laboratory in Dayton, Ohio. The majority of films that we have used from this collection have been films made by the Edison Company and American Mutoscope and Biograph, although other company's films also exist in the collection. Most of the films we have used have been actuality films, although we have also put some fiction titles online. Other collections we have mined for the web presentations include the George Kleine collection and the Theodore Roosevelt Association collection, both of which contain 16mm and 35mm prints as the original source material. These films are also in the public domain. Kleine was a film producer and distributor who donated his collection to the Library, and the Theodore Roosevelt Association collected films pertaining to Roosevelt. They also donated the films to the Library.
It was clear looking through these collections, all three of which have published catalogues for them, that there are certain topics on which we had many films. The initial three presentations made during the pilot project in the early 1990s were “The Last Days of a President, Films of McKinley and the Pan-American Exposition,” “The Life of a City: Early Films of New York 1898-1906,” and “Before and After the Great Earthquake and Fire: Early Films of San Francisco 1897 – 1916.” After the pilot project, the National Digital Library was started on this basis, and I and my other colleagues were hired to work on it. Since then we have put online “Inside an American Factory: Films of the Westinghouse Works 1904,” “American Variety Stage: Vaudeville and Popular Entertainment, 1870 – 1920,” “The Spanish American War in Motion Pictures,” “Inventing Entertainment: The Early Motion Pictures and Sound Recordings of the Edison Companies,” and “Theodore Roosevelt: His Life and Times on Film.”
Consider the home page for the American Memory Project. There is a collection finder that you can go into, and I am going to scroll down to the motion pictures. You see we have the list of all the motion picture collections that are available online. In these pages we have put online since the pilot project, we have tried to add as much additional material as we can to give context to the films, especially if it is primary material such as journal articles of the time or photographs. For the “Westinghouse Works” presentation, for example, we are concerned that just throwing films online with no accompanying information would be doing the films an injustice since most modern viewers would not understand what the workers were doing in the films or indeed why the films were taken. We are fortunate in the Library to have…You can see we have a special presentation there which you go into. We have some promotional material that the Westinghouse Company had put out during this time period, which we use to write background essays on the various companies, and these photographs were from the promotional materials. You can click on them to get a larger image. The promotional material not only gave background information on the projects the company was working on at the time, which include the New York subway and the Niagara Falls generator, but also on some of the working conditions. We also did search into a local newspaper, The Wilberton News, and got an article on life in Wilberton at the time, which added more to the presentation. You can see what kind of lifestyle the people in these films were actually living at the time.
In “American Variety Stage” (a multi-divisional project) there are manuscript collections, photographs, sound recordings, and motion pictures. For the motion picture part of it, we were able to find photographs and advertisements for many of the performers that appear in these films, which gave added emphasis to the presentation. As you can see, the performers in these films are bona fide vaudeville acts, not just a filmmaker's girlfriend.
For the Spanish American War we decided to take the viewer through the war logically to show how the films were linked to how the war progressed. So in here we used stills from the films themselves and explained about the filmmakers going down to Cuba, what they were taking pictures of, and letting people link onto the films themselves.
For the Edison presentation, which is our largest one, we have both motion pictures and sound recordings and a biography page that you can go to. For the motion pictures which there are over 300 on this site, you could go to lists of the motion pictures in alphabetical order, chronological order or by genre, or you could go to an essay on the history of motion pictures which owes a lot to Charlie Musser's fine research on Edison Films. We also used photographs from books of the collections, photographs from journals to highlight the text information.
For each collection each motion picture has a full catalogue record giving information on the film. This is also where you would choose to play the film. We have three formats: RealMedia, which is a streaming format; MPEG which is a higher quality format; and QuickTime, which is a little faster, but not as high quality. A lot of people ask why we chose these three options. We wanted the web site to be accessible to the widest possible audience. Therefore, we tried to choose formats that seemed to have the widest popularity and those that might be especially available in schools. It was pointed out that Macintosh-compatible QuickTime is usually the format of choice in schools.
How do we digitize these films? When the film titles are selected, we go to an outside production house to copy the data to Beta SP tape. We then make an edited tape from the original video in which we add title slates and segment films longer than four minutes. The reason for this is that the download period for films, as many of you may have already experienced, can be tremendous. A four-minute film, depending on one's modem speed and transfer lines, may take several hours for the home user to download. We try to leave a film as complete as it is with any perforations, or original tagging or marks that may have been on it. I have to admit though that we do take advantage of the production house's capabilities to enhance contrast, to lighten prints that are a little dark, to make the image more visible on the computer screen, but this is in no way restoration work. After we have the finished, edited tapes, we send them to another outside vendor, a digital company, to have them make digital files of the film which they return to us on CD-ROM's. This is for the MPEG and QuickTime files. Some of the requirements in the statement of work provided to the company include the capability to convert video to RealTime, MPEG, and QuickTime files; to provide cross platform files, which means they have to be compatible with both Macs and PCs; and there be no stretching or distortion of the images.
Technical specifications: for MPEG files we require 320 x 240 resolution, and for Quick Time the resolution is 160 X 120. The RealPlayer files are made in-house at the Library from the QuickTime files by a colleague of mine using the software Media Pallet. We conduct a quality review of the files after we receive them from the contractor and if they pass muster, they are loaded onto the server. Files have to be checked carefully, because in every batch there are computer glitches and errors.
Since the symposium is about the importance of orphan films and how we can use them, I would like to just ask who uses our website and how have they used it? The most obvious users have been schools, especially secondary and high schools. A learning page for educators is supplied on the website which gives suggestions on how teachers can incorporate these films into their curriculum. That is written by another part of the staff. We also realize from feedback that we are reaching what we call a general audience comprised of different ages and backgrounds from around the world who are accessing the website because of their own interest and not because of a particular class or curriculum. It is difficult and challenging to try to write for a general audience, but you have to keep things as understandable as possible because you are not only reaching an American audience, we are reaching a foreign one also. Film and television producers seem to be using the site as an online resource. They view the films online and become interested in obtaining higher quality film or tape versions for use in the documentaries they are presenting, since the digital files are normally not a sufficient quality for most television productions.
To summarize: what we have is an access format in which we are rescuing orphan films from obscurity and making free access to them over the worldwide web. The limitations to this project are that we are currently only using what we already have in the collections. We are only using materials that are in the public domain and we must either use films shorter than four minutes or segment longer films which makes it difficult for us to use feature films. The digital image, while fairly good resolution in MPEG, is still not even close to being a preservation standard for films to use.
What does the Library have in store for the future in terms of digitizing its films? It is likely that we are going to continue to put online thematic presentations such as the ones I have shown you. Several feature presentations are likely to be multi-divisional, like the “American Variety Stage” one I showed you. This means there will be a mixture of items from various divisions such as films, photographs and manuscripts supporting a certain theme. The Motion Picture, Broadcasting & Recorded Sound Division (MBRS) has other digital plans on its horizon. These include a digital audio- visual repository. This initiative is being organized by MBRS technical expert Carl Fleischhauer, whose ideas I convey to you here.
MBRS is moving its collections from Washington, D.C., to Culpepper, Virginia. We have already moved some of our collections there, but they are intending on moving a good majority of the motion picture processing unit there also. The place is intended to be an audio visual conservation facility where all preservation, processing and cataloguing work will be done. A reference center will remain in the D.C. location where researchers can come and view the films.
What they are planning is a digital audio visual repository where the films can be digitized and access provided to them at the D.C. location of the Library over fiber optic networks. They are currently setting up a prototype project. The initial stage will focus on the sound recording portions since that has been deemed easier to do, and more practical. In the repository digital files can be assembled to create a cohesive digital object. For example, for recorded sound albums, there would be digital files for each sound recording selection plus additional files for the album cover images, for any images in an accompanying booklet. Use of the proper metadata about each file could enable them, in a repository, to be separate items yet also unify them into one digital object for the user. The current thinking for the project is that there would be several manifestations of the same content unit, master and derivative versions. In other words, for a film there would be a master version and then several derivative versions. Since digital images are still not considered the preservation equivalent of film, the master for film would probably be another film copy or there could possibly be a master digital Betacam tape from which to create the digital files from. These masters would be the source of lower resolution derivative manifestations of the film for convenient presentation in the Internet environment. Like our current online presentations, several different versions can be made of a film. For example, an MPEG to high quality version for access and a lower quality with smaller file size RealPlayer or other streaming format version. Theoretically, the researcher would have a choice of accessing higher quality or lower quality versions depending on their needs. Those needing higher quality copies of films, like film producers, would have to have a copy made from the master.
Digital objects would be managed and stored by a repository. A repository should support the storage of digital objects so they can be managed, retrieved, and tracked either as components, i.e. the separate digital files, or as a complete object. Digital objects and their components must be controlled in the repository using persistent, unique global identifiers. It is particularly important to construct a repository where it is possible to migrate media from one media to another. As the old forms become obsolete in this age of ever changing technology. There is also the question of whether we should have an uncompressed master file for video. The sizes of the resulting files are in and of themselves a barrier to computer storage, at least in the current environment. A 30-minute program in uncompressed digital component video have a theoretical extent of about 60 gigabytes.
The question of which format to select for preservation reformatting is also challenging. Recorders for uncompressed signals are very expensive and budgetary concerns may argue for minimally compressed recordings instead of uncompressed. You might wonder why this is an issue. If it is uncompressed, you loose more information. If it’s compressed there is the possibility of losing information, but again if a file size is too large to store, it is a problem.
The challenges to the effective delivery of video, even compressed video, over computer networks are well known. At the same time, the computer and online industry is a very high interest of overcoming these challenges and many initiatives are under way to find better ways to compress and stream video at a variety of data rates. The Library plans to take advantage of these initiatives and wishes to use the digital audio and video repository prototyping project to continue its investigation of this matter understanding that new approaches and options will become available every year. The Library foresees the cost factor as a lead to the offering of varying levels of quality to researchers at different workstations. The constraints will probably be felt most sharply in the case of video materials because of their high data rates. In the future the Library foresees offering researchers a small number of high performance work stations and a larger number of moderate performance work stations. At this time the Library of Congress is not seeking digital solutions for all of its preservation work and is mostly concentrating on access solutions. Interest is high however for a magnetically recorded audio and video. Many tape media and formats traditionally used for audio and video reformatting activities are no longer being manufactured. The condition of tape recordings from the 1960s and 1970s has reached a near crisis state. At the same time, the development of digital technology suggests that the current environment is conducive to the adoption of computer based digital reformatting.
The Library anticipates carrying out a three pronged approach to audio and video preservation, at least until there is a higher level of confidence in computer style preservation.
1) The original items will be retained, properly housed and stored in a suitable environment.
2) Conventional tape-to-tape copies will continue to be produced, often in analog form.
3) New computer digital copies will be made, mostly for access.
A prototype of the moving image and audio system should begin by 2001. The prototype will embrace 500 audio examples and 100 or so moving image samples. Initially, the prototype will be set up in the D.C. location of the Library for a few years. If it proves viable, then a bigger version will be built for the Culpepper facility in 2003 or 2004. So, we are at the very beginning of what will be an exciting digital experiment, through which we hope to offer increased access to the collections and safeguard the original film prints from repeated use. For the time being, the Library of Congress will also continue to retain and preserve the original items and make conventional copies of them when needed, since the digital formats are still not considered to be of preservation quality for us. In the future we can hope to use digital media even more to use and safeguard the films in the Library's collections. •
n Karen Lund
n Sept. 1999