David Bordwell's website on cinema   click for CV




Perplexing Plots: Popular Storytelling and the Poetics of Murder

On the History of Film Style pdf online

Reinventing Hollywood: How 1940s Filmmakers Changed Movie Storytelling

Film Art: An Introduction

Christopher Nolan: A Labyrinth of Linkages pdf online

Pandora’s Digital Box: Films, Files, and the Future of Movies pdf online

Planet Hong Kong, second edition pdf online

The Way Hollywood Tells It pdf online

Poetics of Cinema pdf online

Figures Traced In Light

Ozu and the Poetics of Cinema pdf online

Exporting Entertainment: America in the World Film Market 1907–1934 pdf online


Hou Hsiao-hsien: A new video lecture!

CinemaScope: The Modern Miracle You See Without Glasses

How Motion Pictures Became the Movies

Constructive editing in Pickpocket: A video essay


Rex Stout: Logomachizing

Lessons with Bazin: Six Paths to a Poetics

A Celestial Cinémathèque? or, Film Archives and Me: A Semi-Personal History

Shklovsky and His “Monument to a Scientific Error”

Murder Culture: Adventures in 1940s Suspense

The Viewer’s Share: Models of Mind in Explaining Film

Common Sense + Film Theory = Common-Sense Film Theory?

Mad Detective: Doubling Down

The Classical Hollywood Cinema Twenty-Five Years Along

Nordisk and the Tableau Aesthetic

William Cameron Menzies: One Forceful, Impressive Idea

Another Shaw Production: Anamorphic Adventures in Hong Kong

Paolo Gioli’s Vertical Cinema

(Re)Discovering Charles Dekeukeleire

Doing Film History

The Hook: Scene Transitions in Classical Cinema

Anatomy of the Action Picture

Hearing Voices

Preface, Croatian edition, On the History of Film Style

Slavoj Žižek: Say Anything

Film and the Historical Return

Studying Cinema


Book Reports

Observations on film art

The Celestial Multiplex

Tuesday | March 27, 2007   open printable version open printable version


Kristin here—

The Internet is mind-bogglingly huge, and a lot of people seem to think that most of the texts and images and sound-recordings ever created are now available on it—or will be soon. In relation to music downloading, the idea got termed “The Celestial Jukebox,” and a lot of people believe in it. University libraries are noticeably emptier than they were in my graduate-school days, since students assume they can find all the research materials they need by Googling comfortably in their own rooms.

A lot depends what you’re working on. For The Frodo Franchise, studying the ongoing Lord of the Rings phenomenon would have been impossible without the Internet. A big portion of its endnotes are citations to URLs. On the other hand, my previous book, Herr Lubitsch Goes to Hollywood, a monograph on the stylistic and technical aspects of Ernst Lubitsch’s silent features, has not a single Internet reference. Lubitsch can only be investigated in archives and libraries, where one finds the films and the old books and periodicals vital to such a project.

Vast though it is, the Internet is tiny in comparison with the real world. Only a minuscule fraction of all the books, paintings, music, photographs, etc. is online. Belief in a Celestial Jukebox usually works only because people tend to think about the types of texts and images and sounds that they know about and want access to. Yes, more is being put into digital form at a great rate, but more new stuff is being made and old stuff being discovered. There will never come a time when everything is available.

Even so, every now and then someone proclaims that in the not too distant future all the movies ever made will be downloadable for a small fee, a sort of Celestial Multiplex. A. O. Scott declared this in “The Shape of Cinema, Transformed at the Click of a Mouse” (New York Times, March 18): “It is now possible to imagine—to expect—that before too long the entire surviving history of movies will be open for browsing and sampling at the click of a mouse for a few PayPal dollars.”

Not only that, but Scott goes on, “This aspect of the online viewing experience is not, in itself, especially revolutionary.” He’s more interested in the idea that online distribution will allow filmmakers to sell their creations directly to viewers. That would be significant, no doubt, but as a film historian, I’m still gaping at that line about “the entire surviving history of movies.” Such availability would not only be “revolutionary,” it would be downright miraculous. It’s impossible. It just isn’t going to happen.

(The image above was generated to publicize the “Search inside the Music” program that is an important part of the Celestial Jukebox. Imagine movie screens with actors’ faces in those little boxes, and you’ve got the Celestial Multiplex.)

I will give Scott credit for specifying “surviving” films. Other pundits tend to say “all films,” ignoring the sad fact that great swathes of our cinematic heritage, especially in hot, humid climates like that of India, have deteriorated and are irretrievably lost.

Dave Kehr has already briefly pointed out some of the problems with Scott’s claims, mainly the overwhelming financial support that would be needed: “Tony Scott’s optimism struck me as, well, a little optimistic.” On the line about “the entire surviving history of movies,” Kehr suggests, “That’s reckoning without the cost of preparing a film for digital distribution — the same mistake made by the author of the recent vogue book ‘The Long Tail’ — which, depending on how much restoration is necessary, can run up to $50,000 a title. None of the studios is likely to pay that much money to put anything other than the most popular titles in their libraries on line.” As Kehr says, the numbers of films awaiting restoration and scanning isn’t in the hundreds, as Scott casually says. No, it’s in the tens of thousands even if we just count features. It’s more like hundreds of thousands or more likely millions if we count all the surviving shorts, instructional films, ads, porn, everything made in every country of the world. To see how elaborate the preservation of even one short medical teaching film can be, go here.

Putting aside the need for restoration, newer films present a daunting prospect. To help put the situation in perspective, let’s glance over the total number of feature films produced worldwide during some representative years from recent decades (culled from Screen Digest‘s “World Film Production/Distribution” reports, which it publishes each June): 1970, 3,512; 1980, 3710; 1990, 4,645; 2000, 3,782; and 2005, 4,603. For me the numbers conjure up the last shot of Raiders of the Lost Ark, only with just stack upon stack, row upon row of film cans.

Scott isn’t the first commentator to prophesy that all films will eventually be on the Internet. It’s an idea that crops up now and then, and it would be useful to look more closely at why it’s a wild exaggeration. It’s not just the money or the huge volume of film involved, though either of those factors would be prohibitive in itself. There are all sorts of other reasons why the advent of practical digital downloading of films will never come close to providing us with the entire history of cinema.

Coincidentally, two experts on this subject, Michael Pogorzelski, Director of the Film Archive of the Academy of Motion Picture Arts and Sciences, and Schawn Belston, Vice President, Film Preservation and Asset Management for Twentieth Century Fox, visited Madison this past week. Mike got his MA here in the Dept. of Communication Arts and now returns about once a year to show off the latest restored print that he has worked on. Schawn isn’t an alum, but he has also visited often enough that most students probably think he is. The two brought us the superb new print of Leave Her to Heaven that Fox and the Academy have recently collaborated on.

I figured it would be very enlightening to sit down with these two and talk about why the Internet is never going to allow us to watch just anything our hearts desire. They kindly agreed, and with my trusty recorder in tow we went for burgers—and fried cheese curds, a commodity not available in Los Angeles—at the Plaza Tavern. I’m grateful for the fascinating insights they provided into some of the less obvious obstacles to putting films on the Internet en masse.

No Coordinating Body

Before we launch in, though, one point needs to be made: there is no single leader or group or entity out there organizing some giant program to systematically put all surviving movies on the Internet. There isn’t a set of guidelines or principles. There’s no list of all surviving films. How would we even know if the goal of putting them all up had been achieved? When the last archivist to leave turned out the light, locked the door, and went looking for a new line of work?

Most of the physical prints that would be the basis for such transfers are sitting in the libraries of the studios that own them or in the collections of public and private archives (including many individuals). The studios would make films available online for profit. The archives might be non-profit organizations, but they still would need to fund their online projects in some fashion, either by government support, from private grants, or by charging a fee for downloads. Most archives are more concerned about getting the money to conserve or restore aging, unique prints than about making them widely available. Preservation is an urgent matter, and making the resulting copies universally available for public entertainment or education is decidedly a secondary consideration.

Mike works for a non-profit archive, Schawn for a studio, so together they provide a good overview of some key problems facing the creation of an ideal, comprehensive collection of movies for download.


People who claim that all surviving films could simply be put on the Internet don’t go into the technology and expenses of how that could be done.

Of course, Schawn says, studios want to “digitize the library.” That phrase is highly imprecise, however. He specifies, “For the purposes of this idea of media being online, available, downloadable, streamable, whatever, that’s something that we’re dealing with now using existing video masters. So there isn’t an extra cost to quote, unquote digitize. But there is a cost to make the compression master, what we’re calling at Fox a ‘mezzanine file,’ which is basically a 50 megabit file. That’s the highest-quality ‘low’ quality version of the content from which you can derive all of the different flavors of compression for the various websites that have downloadable media.

“There’s a huge problem with this, in that Amazon, iTunes, and Google all have a slightly different technical specification of how they need the files delivered to them. If you don’t have this kind of mezzanine file, you have to make a different compressed version for each one of these, which costs something, certainly. It’s not incredibly expensive, but it’s not free.

“So why? What’s the motivation to us to compress at Fox the entire library? I don’t know. Are we going to sell enough copies of Lucky Nick Cain [a 1951 George Raft film] compressed on iTunes to cover the costs of making the compression? I don’t think so.”

Compatibility and the Onrush of Technology

Schawn’s mention of the variety of files needed by the big download services raises the problem of compatibility. It’s not just a matter of supplying the files and then forgetting about the whole thing, assuming that the film is available to anybody forever. What about new standards and formats?

Shawn: “Just as with consumer video, the standard changes, so what used to be acceptable yesterday isn’t acceptable now in terms of technical quality.” As time passes, plug-ins make access faster and cheaper, and eventually the original files don’t look good enough. He points out that currently iTunes can’t download HD. If it becomes possible later, “If you want to get 24 in HD, what Fox will have to do is go and re-deliver all the files in HD.” And presumably re-deliver again when the next big format revolution occurs.

Mike explains further, “Using the mezzanine file, you would just have to continue to reformat it to whatever the players demand. The lowest of the low quality will keep going up as people have broadband and can handle larger chunks of data faster.”

What about the film you’ve already downloaded? You acquire films using plug-ins, which change. Think how often you’re told that an update is now available. If you go ahead and keep updating, the changes accumulate. Eventually you may not be able to play the download you paid for. Quicktime will have moved way beyond what the technology was when you made your purchase.

Schawn and Mike both point out that at this stage in the history of downloading, the level of quality is still pretty bad in comparison with prints of films in theaters or on DVDs. It would be nice to think that the virtual film archive could provide sounds and images worthy of the movies themselves, but that will take a long, long time–not the “before too long” that Scott envisions.


I raised another matter: “But what about copyright? Every time I hear something about restoration or bringing something out on DVD, it’s, ‘Well, there are rights problems.’ And some of those rights problems don’t get resolved. I assume that quite a few films that they blithely believe can be slapped up for downloading can’t be slapped up.”

Schawn: “Sure, and there are often not any kinds of provisions in the contract about Internet distribution—obviously! So you’re right, how do you deal with that?”

He pointed to Viva Zapata as a film that Fox has restored but can’t make available due to rights issues. The potential sales are not thought to warrant paying to resolve those issues. “I’m sure there are lots of titles in everybody’s libraries that you can’t just pop up on the Internet and start selling.”

The copyright barrier is worse for archives, which seldom own the exhibition or distribution rights to the films they protect. Usually—though not invariably–the studios do not object to archives owning and preserving prints. Making money through showing them or selling copies would be quite another matter.

Mike described the online presence of archival prints. “On a much smaller scale, this is being attempted in the archive world already, like on Rick Prelinger’s site [Prelinger Archives] or on the Library of Congress’s site, where they’ve put up dozens of moving-image files. True, it’s not independent cinema and it’s mostly commercials and the paper prints that have recently been restored. On Prelinger’s site it’s all the industrial films that he’s collected over the years. That has seen a good amount of traffic, but it hasn’t created new audiences for these films, I would argue. At least, not on a huge scale.”

Prelinger’s site contains only public-domain items, including the ever-popular Duck and Cover, allowing him to avoid the problem of copyright. Similarly, the Library of Congress gives access primarily to films in the pre-1915 era.

Suppose an archive and a film studio both have good-quality prints of a minor American film made in the 1940s. The archive does not have the legal right to put it online, and if the studio decides that it does not have the financial incentive to do so, that film will not be made available for downloading. A private collector might possibly create a file and make it available, but he or she would risk being threatened by the copyright-holder.


We briefly discussed the methods used to prevent pirated copies being made from downloaded films. Like DVDs, downloads can be pirated, with people sending copies to their friends or even offering downloads for a fee in competition with the copyright holder. Copy-protection codes might make it necessary for a purchaser to keep a downloaded copy only on a hard-drive without being able to burn it onto a DVD. Another type of code could erase the file once the film had been viewed once or twice. That’s not exactly conducive to the ideal archive of world cinema, where we would hope to be able to study a film in detail if we so choose.

One might think that piracy protection mainly applies to studios, with their need to make money. Mike points out, though, that the need for such protection “even applies to the archival model, too. For films that you mention that have gone out of copyright, there’s still the same costs associated with putting a silent film that no one owns–digitizing it, creating the compressed master that goes on the Web—and they aren’t the kinds of subjects that people are going to get rich on at all, so there isn’t a lot of piracy. You don’t hear many archivists complaining, ‘Hey, you took my 1911 Lubin film, damn you! You can’t put that on your Website. That belongs on the archive’s Website.’ But that scenario becomes more of a likelihood for more popular titles. The obscure 1911 Lubin film is on one extreme, but Birth of a Nation, a well-known silent film a lot more people would like to see, is on the other. Let’s say the highest-quality copy is on the Museum of Modern Art’s Website and can be easily lifted and posted on your own site. And even if you just say, ‘Oh, I’ll charge 99 cents or I’ll charge 50 cents to stream it,’ it’s still going to be someone taking over something that the archive put all of the high-end effort and money into doing. Frankly, I think unless it’s an archive with a national mandate and a little bit higher budget to digitize and to put the contents of their archives online, there’s not going to be any motivation to make that high-end investment up front.”

Thus an archive may simply not bother to put a film on the internet because it can’t guarantee recouping the costs that would be generated.

Language and Cultural Barriers

Scott’s notion of easy access to all of surviving world cinema implicitly depends on an idea that all these films are either English-language or already subtitled or dubbed for English-speaking users.

That’s not true for a start, so there would remain a great deal of work to translate films that have never been released in English-language markets. That’s another huge, expensive task.

Then there’s the opposite side of that coin. For truly complete access, everyone in the world, whatever language they speak, would be able to download and appreciate every film. Of course, there are billions of people without computers or Internet access, and it looks unlikely that being able to go online will become universal anytime soon. (For figures on numbers of people with internet access, check here; percentages can be calculated by clicking on each country and finding the total population. In Tajikistan, for example, .07% of the population was online in 2005. One has to assume that a lot of connections in some of those countries are dial-up, so downloading films would be virtually impossible. [June 21, 2010: In 2008, Tajikistan’s online population has grown to .08%.])

So let’s just say that for the foreseeable future downloadable films would “only” need to be subtitled in the languages of countries or regions where significant numbers of people subscribe to PayPal.

In our conversation, Schawn pointed out that digital compression files mean that there can be huge numbers of versions, with different soundtracks dubbed in or different subtitles added. Technically it’s possible to do all that translation. Still, “it’s very complicated. So for worldwide distribution of anything, like you’re talking about your silent film. If you’re in Pakistan, do you get the American version of the movie or do you get the version of the movie with the intertitles appropriate to wherever it is that you’re showing it? If so, that quickly compounds the amount of stuff that’s digitized.”

I responded, “Yeah, or a 1930s Japanese film put on the Internet for downloading, subtitled in every language where there are people that can pay for it. The more you think about it, the more absurd it becomes.”

Scott must be implying as well that there is some single “original” version of a film and that that version would be the one available in this ideal collection in cyberspace. Yet any archivist or film historian knows that multiple versions of a given film are typically made, depending partly on the censorship laws of the different countries where it is originally shown. In making downloadable files available, does a studio or archive use only the original version of a film made for its country of origin and thus risk having it include material offensive to viewers in some places where it might be downloaded? Or does a whole slew of different versions, one acceptable in, say, Iran, another inoffensive to the Danes, and still another compatible with Senegalese social mores, get put online? How could one even gather all such versions and digitize them? National film archives tend to have government mandates to concentrate primarily on preserving their own countries’ films. Not every version of every film gets saved.

The Bottom Line

For all the reasons noted here and others as well, film availability for download will follow pretty much the same economic principles that have governed film sales in other media. Mike’s opinion is, “Whether they’re from an archive or a studio, I think things’ll start going online in the same pace that they came onto DVD, in an eight to ten-year cycle. And there still will be large gaps.”

Schawn interjects, “Just as there are on DVD.”

Mike concludes, “There’s stuff that will never go online. Yeah, just as on DVD.”


A truly celestial screening: a restored 35mm print of Bertolucci’s 1900 projected free, under the stars in the Piazza Maggiore, Bologna, summer of 2006


Comments are closed.

David Bordwell
top of page

have comments about the state of this website? go here