|
|
 |
|
Anybody Can Be TV: How P2P Home Video will Challenge The Network News
Drazen Pantic
To be a real-time video journalist, all you need is a blog, a camcorder, and a laptop with WiFi.
 Recently U.S. Secretary of Defense Donald Rumsfeld attested to the revolutionary power of the wireless uploading of digital images to the Internet. Testifying in Congress about the sudden widespread appearance of photographs and video of the abuses at Abu Ghraib prison in Iraq, he did not address this subject as a technological optimist. Rumsfeld is the farthest thing from a dotcom stock analyst circa 1999, or a computer visionary. Rather, he stuck to the brutal reality, explaining that the combination of cheap digital cameras and the Internet had fundamentally changed the dynamics of news making during wartime.
Today, everyone has access to the latest high quality consumer electronic devices. Every cell phone has the ability to capture images, even movies. Once people begin to use these devices to record the significant events in their lives, there is no way to prevent them from slipping cameras into any location. When sensitive material is captured in digital form, it takes on a life of its own. Circulating across the Internet, it becomes a fact in itself. It is impossible for a military organization to control the flow of disbursed, distributed content production in a network environment.
The mainstream media, even if they prefer to ignore troubling facts, are forced to respond. The story of torture at Abu Ghraib prison had been available to journalists for more than three months before the first disturbing photographs surfaced. The media could have reported on it at any time. But with images, it was established as a fact that could not be avoided.
Rumsfeld and his colleagues better prepare themselves, because this is just a taste of the emerging media ecology that is now on the horizon.
First Sightings of The New
Back in 1999, during the dotcom bubble, many of us shared in the delirium of hope that the Information Technology sector could solve all our problems. There was a belief that the Internet would quickly surpass television and become the main source of information for most people.
Web content producers borrowed the television news channel metaphor to design what were called information portals, or aggregator sites. These were vast web sites meant to provide comprehensive knowledge about a particular sector, as well all news associated with it. Internet portals were structured to resemble news channels. They took the idea of staff correspondents, and extended it to include a kind of participatory journalism -- content provided for free by non-professionals that would be vetted in some way by paid staff.
The Internet's distributed nature, and the potential low cost of entry for media makers, struck us as extraordinarily promising. Here is an excerpt from an essay I wrote at the time, which reflected this belief that the Internet was on the verge of radically transforming our media:
Programming produced by any big transnational TV network (CNN, BBC, etc.) is, from the standpoint of an Internet user, similar to an AGGREGATOR SITE distributing video material. It may also function as a portal providing a variety of material of interest to the viewer. Similarities abound - sections of a transnational TV network correspond to parts of an aggregator site: a program schedule is analogous to a web site index, news programs function as general information about the portal's community, shows represent particular web pages or sections on the portal. Most importantly, both a TV network and a Web portal try to fulfill the basic media mission: to define its own reality and broadcast that reality to potential followers - TV viewers or Internet users.
That is what is similar, strikingly similar. What is dissimilar is the nature of the different media. Classical TV and radio are linear and give an observer just two choices: to participate passively in a broadcast as is, or to switch it off. The Internet is more flexible and offers more choices, at least in the basic premises of the media. It is also interactive, allowing the viewer to actively participate.
TV and radio networks are also much more expensive in terms of distribution and production, and by their nature as one-way media, they are closed systems. A single corporation can dictate production costs of worldwide video or radio coverage. This immediately implies that discourse, basic ideological standpoint, and focus of coverage are fixed and are at the discretion of the producer. Every transnational, national and local radio or TV station covers the drama in Kosovo. This certainly propels some more or less peaceful solution. On the other side, wars, genocide and turmoil in Africa (Rwanda for example) is almost not focused upon, allowing events to take their course far from the eye or interest of the public. The New York Times Africa correspondent covers six or seven countries with populations as big as Europe, and with half a dozen wars in progress. The obvious question is whether better coverage or persistent webcasts could stop or minimize human casualties there. Can a camcorder attached to a satellite phone indirectly save thousands of lives?[1]
The portal sites worked fairly well, but they never attracted a large enough base of participants to be financially successful. As it turned out, too few people are informed and skilled enough to regularly contribute relevant, interesting catchy news content. More importantly, though, at the time neither the news media nor the IT sector were truly ready to support participatory journalism.
The news sector was trying to preserve the monopolistic position it had then, and still has. It was deeply reluctant to admit the possibility that anyone could be a journalist. Consider, for example, these remarks by Leslie Gelb, a former editor of The New York Times and, until recently, President of the Council on Foreign Relations. At a Columbia University conference about the role of journalism in war, referring to the possibilities of participatory journalism, Gelb said, "Journalists are in the business of news, not truth."[2] Gelb put it bluntly: for professional journalists, truth is important, but the top priority is to produce material that sells. Sure, it has to be vetted, the truth must be taken into account, but primarily journalism has to result in a salable product. This kind of professionalism leaves little room for outsider, eyewitness accounts as part of a news reader's daily diet.
At the same time, the IT industry was not yet ready to provide the tools and technology necessary for widespread participatory journalism. Even if the equipment was capable of delivering it, the mindset of the industry resisted making it available. In early 1999, I collaborated on a project with IBM to build a laptop for remote news correspondents. It included video editing, network capabilities, and other appropriate technology. But we struggled with IBM's engineers over the inclusion of Firewire. They insisted that it was impossible to fit a Firewire card in a PCMCA slot. As a result, the prototype models we outlined could not do the efficient video transfer necessary to be truly useful to remote journalists, let alone war correspondents (the intended recipients of the laptops), and the project collapsed. Then, a few months later, Apple came out with a laptop that included Firewire. The lessons of this experience have been confirmed repeatedly for me and others who see the potential contribution that digital communications could make to civil society: the state of mind of the IT industry is often more limited than what the technology itself is capable of delivering.
How long will it be before our news reports come direct from local sources with their own video production facilities, in real time, over the Net?
|
Another reason for the meager success of participatory reporting was that broadband proliferation was extremely limited at the time. Even though there was much talk and high hopes about broadband, its penetration to households, even in the US, was poor. While quite a few people in both media and IT saw the opportunity for convergence between the two sectors, the basis was not yet there for an effective convergence. In this light, it is worth reconsidering the infamous disaster of the AOL and TimeWarner merger. As we now know, once their corporate marriage was complete, the new entity did not know what to do with itself. Nonetheless, this event pointed in the right direction. It was an important development, even if it came too soon, signaling that convergence between the media and the IT industry is the way to go.
We're Not Gonna Take It
Today, the biggest problem faced by both mainstream media and the IT sector is a lack of trust on the part of consumers. Consumers no longer accept the passive position that was common in the last century. They want to participate in media production. Understanding this shift has been the secret behind the huge success of Apple. Steve Jobs and his team realize that the company's future depends on supporting this transition of their customer base from passive audience to digital producers. They sell the notion that anybody can be a content maker, wrapped in attractive, translucent plastic. That notion is Apple's Ur-product.
The consumer no longer gives a blank check to either the broadcast media or the IT industry. We have all experienced the unreliable reporting in mainstream news, most recently during the run up to the invasion of Iraq, and the visceral disrespect of the monopolistic software industry, which charges exorbitant prices for mediocre proprietary products.
On the positive side, however, we have also witnessed two developments, especially in the years after September 11, 2001. We have encountered the vast amount of direct, unmediated information available over the Internet -- on web logs, personal web sites, targeted e-newsletters, independent news portals, and other sources. This information, as a whole, is as reliable as any news seen on the cable networks, while often providing an independent perspective missing from corporate media. At the same time, we have also seen how Free and Open Source Software (FOSS) is on the rise in the IT sector, with companies like IBM and Hewlett Packard providing secure, high quality computing for corporate clients. With both the media and IT sectors, we see the emergence of participatory methods that have led to viable, practical, widespread uses.
Meanwhile, many IT products have improved significantly in the last few years, with a wide range of high quality consumer electronic devices for content production on the market. These are low cost, solid tools for audio/video production, editing, and the packaging of media into distribution units, such as MPEG. MPEG is software that compresses video into a format suitable for distribution over broadband networks like those widely available in US households today. MPEG files not only allow people to exchange video material over cable and DSL networks, but the quality is good enough for professional broadcasting on traditional television. In fact, as a technical standard, MPEG is identical to what is used in cable distribution. While MPEG is a proprietary technology, the open source community provided an MPEG 4 compatible video codec (or encoder), called Xvid, which is not controlled by any private entity or corporation. Xvid video files can play on any Quicktime or Real Media compatible player. And there are open source versions of the players as well. The appearance of open source versions of this technology insures its availability to all who want to use it -- so that indeed anyone can be a television news producer.
Additionally, broadband has now reached many homes. More importantly, wireless became available on a consumer level, and free wireless networks began to emerge in many urban areas.
Together, these factors contribute to the emergence of an entirely new, distributed media environment that can no longer be controlled from the top down.
You Too Are A News Producer
Not long ago, the production of high quality video was expensive and tied to proprietary formats (or codeces). Companies like Avid or Pinnacle produced video editing tools only for high end professionals from the broadcast industry; their products were too expensive for non-professionals. The expertise required to use these tools also went beyond the ability of most people. Just learning the basics meant investing much time into acquiring arcane skills. This alienated people from direct participation in the video documentation of events. If you wanted to record and edit a movie, you needed professional connections.
But over the past four years, a number of factors have contributed to making video production capabilities available to millions of people. This shift began with the introduction of the software product Final Cut Pro, the first consumer grade video editing program that could produce broadcast quality video on a laptop. In parallel, digital video cameras came down in price. Once they reached $700 or so, there was a huge proliferation of DV cameras. Other contributing factors include: the proliferation of open source software; simple open standards for video production (like MPEG 2 and 4); increased processing power in personal computers and digital cameras; the availability of consumer products that make MPEG video (like Tivo); and the proliferation of broadband Internet access. Open source tools for all aspects of video production keep getting better, more sophisticated, and easier to use. Some open source video editing tools allow for real time rendering of video and video effects -- a process that was unimaginable only a few years ago, even at expensive, exclusive production houses.
These resources are now within reach of the average middle class household. We are now seeing a new level of mass participation in video making. These independent productions, done outside of any institutional framework, are already being distributed in the form of high quality video on demand over the Internet.
How long will it be before our news reports come direct from local sources with their own video production facilities, in real time, over the Net? Who needs a cable network's team of celebrity reporters, with their jingoistic coverage of "Operation Iraqi Freedom," when I have unfiltered access to images and testimony from the war zone?
It is important to note that this proliferation of low cost tools, and their increasing quality, was made possible by the Free and Open Source Software movement and the push for open standards. Commercial software companies would not have made the commitment to open standards had they not been confronted by open source alternatives that are cheaper and more flexible. The emphasis on open standards came from the open source community. Companies then saw the advantages of the open approach, and began to adopt it. For example, Apple's Quicktime, which is not open source, makes use of open standards to encode video and audio. Apple made this choice because open standards offer more flexibility, while they also help solve problems with the licensing of proprietary intellectual property, since open standards, by their very nature, are free.
At the same time, open standards are a better guarantee that content produced today will be available for playback in the future. Content that is enslaved in a proprietary format has less of a chance of surviving over time, since if the IP holder goes out of business, all support for their format is likely to disappear with them. One glaring example of this problem is offered by Real Media. Once anything is encoded into Real Media, it cannot be transferred back into uncompressed video, or into any other format, without a significant loss of quality. This creates potentially huge problems. For example, at the Tate Modern in London, over the last five years they have encoded many of their interviews with artists and cultural figures into Real Video. Assuming the possibility that Real Media is one day no longer in business, their proprietary standard may no longer be supported and the Tate's video library may no longer be playable. It will effectively vanish, even if the Tate saves the original files.
It is now possible to use a laptop-Internet system for on-the-fly transmissions from remote locations for distribution over a television network.
|
For makers of rich media content, open standards are essential. And open source hackers have done much to extend the availability of high quality open standards. For example, the most widely used standard for the formatting of audio on the Internet is MP3, which is not open. It is a propriety standard. But the open source community has developed variations of MP3 that are compatible and non-proprietary. One example is OGG Vorbis, which both open source and open standard. Apple has cleverly included the OGG Vorbis standard in the Quicktime player and its iTunes music device.
It is impressive how far Apple has come to embrace open standards. And other companies have moved in this direction, as well. Today's version of Final Cut Pro is also largely open standards and open source. What you pay for, essentially, is the Graphical User Interface. This is one of many examples of a software product based on open source software and open standards, with a proprietary container. Another example is Apple's DVD Studio Pro. For this software, you literally pay for the GUI. Everything else is open source and open standards. And Apple does not object if you don't pay for it.
What is still lacking is the aggregation of these rich media tools into a complete, easy-to-use package. A first effort toward this kind of this bundling is the CD "dyne:bolic." This disk is a complete, open source, Linux-based, laptop video production and distribution suite. It comes with the following software installed: MPEG4IP (live Internet streaming and capturing clips in QuickTime compatible format); FFMPEG (transcoding and streaming in Flash, WMF or Real format); Cinelerra and LiVES (edit and publish video clips); FreeJ (VJ livesets); Audacity and ReZound (edit audio); and Gimp (image manipulation software). Unfortunately, "dyne:bolic" is not as user-friendly as one would like. But it is an important step in the right direction.
The tools suites are appearing. But more than anything else, we need education. People have to learn that they can produce video comparable to professional broadcast quality using these inexpensive, open source tools. What can be more important to the future of democracy than giving citizens the ability to better communicate with each other?
Live (Via WiFi) From Bryant Park
On January 14, 2003, we met in New York's Bryant Park for the first successful live broadcast uploaded over a public wireless network for transmission over cable TV. I was joined by my collaborators Kenyatta Cheese and Marty Lucas, the latter of Manhattan Neighborhood Network, a non-profit public access channel in New York that is distributed by TimeWarner Cable. We wanted a proof of concept that established basic procedures for the broadcasting of video in real-time to a cable or satellite TV network from remote locations. Public wireless nodes provide enough bandwidth to carry IP video streams at a sufficiently high quality. These clips are perfectly acceptable for TV transmission.[3]
Moreover, we wanted to do this using a laptop, open source software, a consumer-grade camera, and an easily available broadband Internet connection -- preferably WiFi.
The motivation for this exercise was to demonstrate that classical TV production equipment, requiring tens of thousands of dollars and a specialized infrastructure, is becoming a thing of the past. It is now possible to use a laptop-Internet system for on-the-fly transmissions from remote locations for distribution over a television network. The hardware and software we used was deliberately chosen because it is within the skill level of even a moderately technical person.
So on a freezing winter afternoon we went to Bryant Park with two laptops (one as a backup in case of battery failure, because it was so cold outside), two digital video camcorders, and a couple of professional quality microphones. We established a wireless connection through a local, public WiFi network maintained by the non-profit NYC Wireless, and broadcast from that spot to a computer at MNN studios. The video and audio was captured by the camcorder and fed into the laptop, where it was encoded as MPEG4/AAC streams, then sent out as a unicast stream via the WiFi connection. At MNN they played the stream through a scan converter -- which converts the stream on a computer into a video signal -- then broadcast it live on the air. Instead of sending the video/audio to a replication server, a client computer with a static-IP address received the stream from across the Internet and played the media out to a video switcher and onto the cable channel/satellite broadcast.
The show consisted of the three of us interviewing each other about the laptop-WiFi broadcast process: we can do it! It was totally self-referential delirium, which of course is what television is all about.
Only months later, CNN used the same basic technique to broadcast live reports from Iraq during the war. But what is most encouraging about this technology is how it can lead to new forms of distribution that bypass centralized broadcasting entirely, allowing for the creation and distribution of video programming from within a Peer-to-Peer (P2P) network.
P2P-TV and Effective Video Blogging
When this emerging production potential meets the network distribution paradigm, it reaches its most profound level. Many people have observed in the early use of Peer-To-Peer (P2P) technology, such as Napster, the power that comes from the sharing of digital resources across a distributed network. When so many computers are joined together, it creates a huge repository of material.
This potential attracted the attention of open source developers, and they went in two different directions. The first was to create decentralized P2P networks in which the main priority is to protect the privacy of participants (such as Gnuttella, Livewire, etc).
The other direction was to coalesce P2P networks so they can more effectively distribute popular resources. One of the best examples of this approach is BitTorrent. The biggest problem for any P2P network is the curse of popularity. The more popular a file becomes, the more bandwidth required to provide it. Because more people want that file from you, you need more bandwidth to serve all those requests. For this reason, only high bandwidth operations could engage in massive P2P distribution of files that are suddenly in great demand.
BitTorrent was developed to prevent the bottleneck that happens when timely new video clips become popular. This is especially important for original content that exists only at one or two locations on the Internet. If only a few people have a file when it gets attention, the file becomes difficult to access. BitTorrent addresses this problem through the active sharing of network resources: when each new person starts to download a file, her computer automatically becomes a server of the same file, able to supply other requests from within the BitTorrent network. There is no waiting for the file to completely download before the computer can begin serving. So the more popular a file is, the more upstream bandwidth it immediately acquires.
This capability is extremely useful for video. Audio files, of course, are smaller. The network distribution of video requires better logic and sharing of bandwidth. BitTorrent treats bandwidth in a way that makes it much easier for individual households to serve video. Typically, home networks use much more bandwidth for downloading than uploading. Most home DSL and cable networks are architected to handle a large amount of downloads, and they assume you will send up very little. BitTorrent, and similarly designed P2P networks, coalesce all of the upstream capacity for the households in the network, creating an aggregate that is not only large, but efficient. Without a protocol like BitTorrent, it becomes far less practical to serve video files to more than one downloader at a time. But with this capacity, the serving of independently produced video to large audiences from regular broadband household networks can become a popular practice.
Web logs are another example of how people are shifting from passive media spectators to active media producers. Now that a rich media layer is being added to blogs -- with the appearance of video blogs -- it seems that a viable alternative to centralized TV networks is emerging. For example, consider what might happen through the joining together of video blogs, Real Simple Syndication (RSS), and BitTorrent. This is a very powerful combination.
RSS is a mechanism for the indexing of content on blogs. It also enables the automated entry of content from one blog into another. This very basic approach to syndication, which is already being used (in one of several versions) by nearly 1 million web sites, makes it possible for blog content posted on one site to circulate across the internet in an instant. RSS allows for the proliferation of meta data so blog content can be indexed not only by other blogs, but also by search engines like Google or Yahoo!
Now that a rich media layer is being added to blogs -- with the appearance of video blogs -- it seems that a viable alternative to centralized TV networks is emerging.
|
Once it becomes a common practice for independent video to be posted on video blogs, RSS will facilitate the widespread distribution of video across the Internet. Unfortunately, we are still far from effective meta-tagging of time-based media. You still cannot search through a video clip online for the type of content it has, or its position on the file. You may want to find all the clips on the Internet that mention your name, for example, and the moment when your name comes up in the clip. But while you can do this kind of search with text on web pages, it is still impossible with rich media. But for the time being, video tools in combination with blogs and RSS is a viable substitute.
Blogs and TV mesh well because they are both so self-referential. Just as most blogging seems to be about blogging, most TV is about TV. TV is also a huge cut and paste engine that reuses the same material over and over -- just as blogs do.
I am collaborating with a group of people to build an IT infrastructure and the conceptual criteria that will support broad participation in video production and distribution using blogging technology. Our blog for the project is at http://www.unmediated.org. We also have a video blog, DV Guide, with examples of past video shows, as well as material collected for future shows. Visit us at
http://dv.open4all.info.
DVGuide needs artists, editors and others who are working in digital video, crossing borders and dealing with questions of globalized living who would benefit from the advantages of a raw media, reciprocal, BitTorrent network. If you would like to contribute to DVGuide as an editor, producer, or otherwise, on an ad-hoc basis or by some other arrangement, please make contact. Ideal content for DVGuide would be open-source, raw, captured media, fully produced documentary, samples, or other locally produced yet globally relevant material. If you are interested in allowing artists to hack, render, tweak, mix, and show your media, then you may want to consider contributing.
The Ultimate Set Top Device
We are not far from the day when all the technological pieces discussed above can be put together into a stand alone, easy to use, hardware unit that can be plugged into a set top box or game console that is connected to the Internet. Such a device would include hardware MPEG encoding, P2P BitTorrent technology, and a user friendly interface. It would allow you to play the videos you downloaded from the Internet on your television screen. It would also have a "video in" plug for a digital camera, and a start/stop button. Simply by plugging the camera in, you could become a video producer. After recording a video piece, your clip could upload automatically to your blog using BitTorrent distribution. This would allow other viewers to play clips over the Internet of you and other participants in your DVGuide-like network, through a box that is connected to your TV set.
As is often pointed out, the Internet is an end-to-end network with all of the intelligence concentrated at the periphery, while the network itself is deliberately dumb. Television is the opposite: the end devices are stupid, while the network itself is sophisticated. Control is exercised within the TV network itself. A device that connects to set top boxes which combines the technology of the Internet with the effective distribution of TV can give us the best of both worlds.
Footnotes
[1] Drazen Pantic, "Everybody Will Be TV," 1999, archived at http://www.open4all.info/wiki/drazen/Everybody_Will_be_TV
[2] http://www.un.org/Pubs/chronicle/1999/issue2/0299p29.htm
[3] Although "broadcast-quality" video is loosely defined as 640 x 480 pixels at 30 frames per second, most viewers will accept the look of 320 x 240 video doubled in size. We also decided that a minimum of 8 fps was "acceptable" for watching short video clips of 30 seconds or less, although 15 fps -- the minimum rate at which the human eye sees fluid motion -- is preferable. For audio, we chose 16 bit stereo sound, encoded at a minimum of 32 kbps using a standards-complaint audio codec (AAC).
|
|
