« Graffiti as Conversation | Main | IN Network »

March 01, 2005


Orchestrating the End of the Internet?

"Anyone who wonders how the Internet will die will find one possible scenario in the recent decision by the Internet2 consortium to bring Hollywood into the design process for our next-generation Internet.

Hollywood is on a roll. In a fraction of the time that it took the music industry to emasculate Napster, the Motion Picture Association of America has managed to shut down the highest profile file-sharing sites (Suprnova and LokiTorrent) and begun to sue its own share of college students. More importantly, the MPAA recently persuaded Congress to legislate something their fellow lobbyists in the music industry never managed to achieve: a copyright control device in every player. By this July, every DVD player and TiVo box will sniff for a "broadcast flag" that prevents it from copying digital TV broadcasts. This hardware intervention effectively destroys even the possibility of fair use, since artists and educators cannot transform, parody, or criticize what they cannot record. [via Rhizome]

The Electronic Frontier Foundation is mounting a noble campaign to grandfather a compliant tuner before the legislation takes effect [1]--but in the meantime the MPAA has set its sights on its next acquisition: the ultra-high bandwidth Internet2, which runs on the 10 gigabit per second Abilene backbone:

"We've been working with Internet2 for a while to explore ways we can take advantage of delivering content at these extremely high speeds, and basically manage illegitimate content distribution at the same time," said Chris Russell, the MPAA's vice president of Internet standards and technology. "Those would go hand in hand." [2]

To judge from the statements of Internet2 bigwigs, their technologists have already capitulated before the battle has even begun:

"This wraps together the broad interest we have in working with our members and potential members on advanced content delivery," said Internet2 Vice President Gary Bachula. "Obviously we're interested in making sure that's legal and safe." [2]

The presentations I've seen to date from the Internet2 consortium, from music classes taught by "master" conductors [3] to biometric and authentication applications for "managing identity" [4], suggest that Internet2 is a broadcast organization in network clothing. While it's doubtful that everyone at work on Internet2 shares this vision, the consortium's choice to "collaborate" with the MPAA could give media conglomerates a chokehold on the 21st-century Internet.

The stated goal of this collaboration--to investigate new business models for streaming movies--sounds reasonable until you read that Internet2 is already capable of transmitting a DVD movie from Switzerland to Tokyo in under 5 seconds. (Cut to Jack Valenti choking on a bagel as he reads this in the morning paper. [5])

No Hollywood exec is going to sanction a business model that lets Joe User download a movie onto a hard drive faster than the time it takes to launch his Web browser. Forget streaming video on demand. Hell, that isn't even enough time to watch a BMW ad.

The technology behind Internet2 *breaks* anything remotely resembling a broadcast business model, which is why the MPAA will do its best to disarm the technology by installing Digital Rights Management directly in its routers to stop interesting content from ever getting into the pipeline.

Now, the idea of "intelligent routers" may sound appealing to the average Congressperson, but the technologists of Internet2 should know better. Internet 1 was able to adapt so quickly to new uses--from email to the Web to IM--because its routers are fundamentally *dumb*. As engineer David Reed and others argued in the late 1970s [6], an indiscriminate "end-to-end" network would allow its users to hook up ever faster and more capable computers to its endpoints, without locking out uses that the network's architects could not have foreseen. Broadway was built for horse-drawn carriages, but since then its level pavement and wide footprint has accommodated Model Ts and Toyotas--precisely because its architecture was not optimized for carriages. Even companies like Disney and Microsoft have publicly recognized the importance of e2e to technological innovation. [7]

Yet David Reed already smelled a threat to the e2e paradigm back in 2000, citing among other threats Hollywood's interest in streaming movies. In "The End of the End-To-End Argument?," Reed imagined uses that could not be foreseen by intelligent routers, including "collaborative creative spaces":

"With broadband networks we are reaching the point where 'pickup' creation is possible--where a group of people can create and work in a 'shared workspace' that lets them communicate and interact in a rich environment where each participant can observe and use the work of others, just as if they were in the same physical space." [8]

Reed's description of emergent collaborations bubbling across the network like so many games of pickup basketball is a world apart from the stuffy master classes of the Internet2 consortium. But it reads a lot like Internet2's stepsister, the MARCEL network of Access Grid communities [9]. If the "official" Internet2 consortium is a symphony orchestra in tails, the MARCEL network is a makeshift performance troupe. Internet2 has 200 university and corporate sponsors; MARCEL has a motley crew of artsy scientists, network performers, and Jitter jocks. Internet2 uses stable high-bandwidth videoconferencing for the privileged participants and netcast for everyone else; MARCEL uses the rickety Access Grid platform, which permits all users to participate at the same level.

As MARCEL's Don Foresta has suggested, "efficient use of network resources" will be the argument marshalled by the media conglomerates against creative re-purposing of Internet2, just as the phrase was used justify the commercialization of the airwaves even if it contradicted the physics of electromagnetics. [10] (In Italy fascist apologists vindicated Mussolini by boasting that the trains ran on time.) Again, Reed saw this coming:

"The architects who would make the network intelligent are structuring the network as if the dominant rich media communications will be fixed bandwidth, isochronous streams, either broadcast from a central 'television station' or point-to-point between a pair of end users. These isochronous streams are implicitly (by the design of the network's 'smart' architecture) granted privileges that less isochronous streams are denied--priority for network resources." [8]

Privileges and networks don't make good bedfellows. For all its talk of community and access, Internet2 seems to be offering a backwards-thinking hierarchic model of culture, a sort of Great Performances meets Reality TV. To be sure, MARCEL has experimented with broadcast models as well, featuring gigs by luminaries such as fractal mathematician Benoit Mandelbrot and Max/MSP inventor Miller Puckette. But these admirable cameos don't reveal MARCEL's true potential; that happens when three students from different continents suddenly realize they are in the same Access Grid "room," and begin trading Max patches or holding pen-and-paper sketches up to the videocamera. In these quotidian, pickup collaborations--as in the beguiling video-composite performances Net Touch and Net Hope organized by Tim Jackson's Synthops lab in Toronto [11]--high-bandwidth networks prove they can be even *more* reciprocal than low-bandwidth networks. [12].

While MARCEL has for some time seemed a promising platform for the interchange of ideas and networked art, only recently have I come to realize that it can also serve a valuable tactical function. Like the EFF's efforts to make room for legitimate uses of digital TV recordings, MARCEL's creative community can develop and showcase remixable network performances--both for their own sake as well as to provide empirical evidence for future court cases to defend the value of end-to-end networks. [13] In so doing its members can promote the vision of a vibrant future for the Internet--one that lets us all play onstage instead of admiring the players from the balcony."

Jon Ippolito


[1] http://eff.org/broadcastflag/


[3] http://www.nws.edu/NWS_internet2.asp?pg=NWS_internet2.asp

[4] http://www.campus-technology.com/print.asp?ID=10405





[9] http://newmedia.umaine.edu/marcel/

[10] http://www.newamerica.net/Download_Docs/pdfs/Doc_File_143_1.pdf

[11] http://www.rcc.ryerson.ca/synthops/process.htm

[12] Theorist-gadfly Jean Baudrillard pointed out that reciprocality was the key feature missing from Hans Magnus Enzensberger's definition of emancipatory media.

[13] Cyberlaw guru Lawrence Lessig laments that a lack of empirical evidence doomed his argument in Eldred v. Ashcroft.

Posted by jo at March 1, 2005 09:12 AM


"Subject: RHIZOME_RARE: Internet2: Orchestrating the End of the Internet?
Date: Fri, 4 Mar 2005 15:18:44 -0800
From: "Philip Galanter"
Reply-To: "Philip Galanter"
To: rare@rhizome.org

(Please consider for Rare/Digest to correct factual errors.)

A few days ago Jon Ippolito posted a sort of manifesto positing Internet2 as a threat to the kind of internet artists and academics would like to continue to use.

I know Jon a bit from my MARCEL involvement and elsewhere. Jon's a really smart guy with a keen gift for deft rhetoric, and I am sure he means well.

Unfortunately Jon's post invokes several basic misunderstandings of the related technologies. These confusions are no mere technical quibbles. They are fundamental to the central thesis that somehow the Internet2 effort may bring about the death of the internet.

This couldn't be more wrong.

I'll let the basic facts, as corrected in the following, speak for themselves.

>.By this July, every DVD player and TiVo box will sniff for a "broadcast flag" that prevents it from copying digital TV broadcasts.This hardware intervention effectively destroys even the possibility of fair use, since artists and educators cannot transform, parody, or criticize what they cannot record.* The broadcast flag does not prevent making recordings for time shifting or other personal "fair use".

This is simply not true. There are hairs to be split, but basically (1) the broadcast flag only applies to over-the-air broadcasts (not cable, satellite, or internet streaming), and (2) it will not prevent copying for fair use. For example, you will still be able to record over-the-air broadcast TV shows at home for later use.

The broadcast flag system *will* prevent large scale redistribution, i.e. massive piracy. But this has always been illegal...even in the era of videotape.


>>The technology behind Internet2 *breaks* anything remotely resembling a broadcast business model, which is why the MPAA will do its best to disarm the technology by installing Digital Rights Management directly in its routers to stop interesting content from ever getting into the pipeline.

Again, this is simply not true. Router level digital rights management is not being considered by any of the internet standards bodies. It's not even over the horizon. However, the current worldwide internet upgrade from IPv4 to IPv6 *does* make multicast an intrinsic part of the protocol rather than an add-on. And multicast is *exactly* the technology a broadcast model needs.

But multicast also benefits "the little guy" because in principle independent artists will no longer have to pay for increased server capacity as their audience grows. The shared network, rather than the server, will distribute the stream to as many viewers as are interested.

So if anything, "broadcast" related technical changes in Internet2 (and eventually other networks) will serve as a democratizing equalizer.

And by the way, IPv6 multicast has *no* built-in Digital Rights Management. None. And routers under IPv6 remain "dumb" contrary to implications otherwise.

(As a footnote, multicast is also the enabling protocol technology that makes the Access Grid, MARCEL's current platform of choice, possible.)


>>>For all its talk of community and access, Internet2 seems to be offering a backwards-thinking hierarchic model of culture, a sort of Great Performances meets Reality TV.

Again...not true. Reasonable people can disagree when it comes to matters of esthetic taste, but contrary to Jon's central thesis Internet2 technology remains both content and application agnostic.

Elsewhere he mentions "privileged" isochronous channels. But isochronous channels don't, and can't, even exist under either IPv4 or IPv6 or on either Internet2 or "internet1".

The ability to quickly create improvised collaborative groups was recognized as being among the highest application priorities in the
earliest pre-planning of Internet2. Application level efforts such as the Internet2 Commons, VRVS, and indeed the very Access Grid technology that MARCEL depends on, are some of the fruit of this early vision.

Today on Internet2 non-hiearchical social interaction isn't speculation...it's already well established standard practice.

And when it comes to Internet2 *content* people are free to do what they will. If one finds the current crop of artistic efforts to be wanting the best, and entirely invited, response is to go out and create something better.

To sum up, there is simply no factual basis for any Internet2 vrs MARCEL conflict.

And I personally look forward to working further with both!"--Philip Galanter

Posted by: Jo at March 5, 2005 12:15 PM

Date: Tue, 8 Mar 2005 23:32:23 -0500
From: "Jon Ippolito"
Subject: RE: Internet2: Orchestrating the End of the Internet?

Thanks for this thoughtful response, Phil. It's hard for me to disagree with someone who is so plugged into Internet2--and says such nice things about me. But I'll do my best, with the help of Google and the EFF's Seth Schoen.

>> (1) the broadcast flag only applies to over-he-air broadcasts (not >>cable, satellite, or internet streaming

That's basically true--although Seth tells me the jury is still out on whether the flag applies to cable. In any case, Hollywood didn't need to lobby for a broadcast flag on cable or satellite broadcasts because those providers already can (and generally do) slap on even stricter DRM. The FCC approved this stricter form of exclusion--zith minor limitations--in their recent "Plug and Play" proceeding.


>> (2) [the broadcast flag] will not prevent >>copying for fair use. For example, you will still be able to record over-the-air broadcast TV shows at home for later use.

Yes, if you use software or hardware that meets Hollywood's specification. While in theory such a device might let consumers do everything they're legally allowed to do, in reality the MPAA has no
incentive to encourage devices that let consumers do something Hollywood won't profit by. Jack Valenti isn't really interested in letting you freely record digital TV with your current DVD burner, email a clip from a President's press conference to your mom, or create a high-definition video installation based on an archive of short TV clips.


To be sure, Jack doesn't have the direct authority to approve or disallow new PC tuner cards and DVD players. But before the broadcast flag, Hollywood had no recognized legal interest in appealing an FCC decision. Now that there's an FCC ruling designed specifically to protect their own interests, the movie studios are hiring expensive lawyers to do just that.

In the short term, some consumers may not notice the immediate effects of the flag. (Though I'm guessing Johnny Consumer will be ticked off
when he learns that the $500 video player he bought in 2005 won't play a recording Aunt Betty made with a DTV receiver she bought in 2006.) More important for the long run, however, will be the flag's effect on innovation in video software and hardware. To build a legal device to interoperate with the broadcast flag will require it to be "untamperable." That rules out any technological innovation that requires tinkering or experimenting with an existing apparatus. In particular, it rules out open source software such as GNU Radio, because open source projects *require* others to be able to tinker with them. Do we really need another market where open source developers are told they can't compete because it would it would wreck the business plans of entrenched commercial interests?


While analog recording devices are not constrained by the broadcast flag, the MPAA has been trying other schemes to "plug the analog hole." A particularly ludicrous proposal, documented in the MPAA's "Content Protection Status Report" filed in the US Senate in 2002 and echoed by two TV execs at last week's DVB World, calls for embedding anti-copy chips into every analog-to-digital converter manufactured. As you probably know better than I, those wee little converters are everywhere--in digital scanners and camcorders, but also in thermometers, seismographs, computer mice, mobile phones, and light-meters. That Hollywood would presume to constrain technological development in everything from health care to scientific research testifies to its fanatic obsession with controlling technologies that are incompatible with its business model.


>>The broadcast flag system *will* prevent large scale redistribution,i.e. massive piracy.

But will it? Many observers have noted that the broadcast flag's DRM has all the toughness of a wet paper bag. It's just unencrypted bits in a
stream, and the spec is publicly and lawfully accessible. The MPAA even sidestepped the question of effectiveness in their official FAQ. The ease with which it can be subverted makes me worry that its introduction will spur illegal reuses of digital TV while locking out legal ones.


By the way, you don't need to be a Bram Cohen to get around this DRM. Sure, you won't find a flag-free player manufactured after this July. But if you're keen on committing massive copyright infringement, just plunk down $150 for a tuner card before the deadline. Then you can spew pirated Alias episodes afterward to your heart's content.

>>The ability to quickly create improvised collaborative groups was recognized as being among the highest application priorities in the earliest pre-planning of Internet2. Application level efforts such as the Internet2 Commons, VRVS, and indeed the very Access Grid technology that MARCEL depends on, are some of the fruit of this early vision.

I'm glad to hear you think there's plenty of room on Internet2 for pickup collaborations outside of the broadcast model. Knowing you, you've probably participated in some interesting events on Internet2. So forgive me if the consortium's public face--which I've seen in Ann Doyle's presentations and Internet2 Web sites--doesn't reinforce the vision of open and improvisatory collaboration described above. Some of the networked performances I've seen associated with Internet2 sound innovative, but they take for granted a clear distinction between performers and audience. Likewise, I want to be part of the Internet2 Commons--but not if I have to shell out a couple grand to join it. And an Open Student Network is a great idea, but not if the end result is a television channel for Connie Chungs-in-training.

But as you suggest, much of the problem may lie with the choreographers rather than with the engineers. Perhaps if MARCEL and Internet2 folks
brainstormed together, they might come up with less hierarchic models of high-bandwidth culture.

>>Router level digital rights management is not being considered by any of the internet standards bodies. It's not even over the horizon.

That's good to hear, and you're in a much better position than I to know what Internet2 chieftains are contemplating. I suppose my concern is less with what Internet2 is now than with what it could easily become if the MPAA starts getting its claws into it.

You see, I don't know how to square your reassurances with the comments by MPAA and Internet2 VPs in the News.com story I posted earlier. When Chris Russell talks about working with Internet2 to "manage illegitimate content," how is he going to do that without sniffers inside the network that tell him what's being traded or who fed it into Internet2?

Today, you and I can plug off-the-shelf PowerBooks into Ethernet cables at our university offices and communicate via Internet2 without
Hollywood's blessing. I was optimistic that this privilege might someday belong to a much wider cross-section of people. But now the pessimist in
me is thinking that even spoiled academics like us may be denied that freedom if the MPAA gets its way. As the folks at Public Knowledge commented in regard to the FCC's "Plug and Play" proceeding:

"One of the key issues in this proceeding is the extent to which content companies and content-delivery services can leverage the Commission's
goals of promoting digital television, cable compatibility, and competition in the navigation-device market into sweeping regulations whose principal effect is not to serve these goals, or even to prevent piracy of digital television. Instead, the real purpose of these proposals is to restore to content companies, to the extent possible, the degree of control over video they exercised prior to the invention of the videocassette recorder."


I hope that folks like you with some influence in the consortium will take this threat seriously enough to be mindful of it. Thanks for helping to air this debate in public.

>>And I personally look forward to working further with both!

I'd love to start a working group on Internet2/Access Grid devoted to questions of access and control. As long as admission is free :)


Posted by: Jo at March 12, 2005 11:00 AM

Date: Wed, 9 Mar 2005 20:23:02 -0500
From: Philip Galanter
Subject: Re: [MARCEL-members] Re: Internet2: Orchestrating the End of the

Thanks back to you Jon for furthering the discussion of some of the tough issues raised by the new networked communication technologies.

To be clear, my intent in my first response was to address the question posed in the subject line. That is, I wanted to make the point that Internet2 is not orchestrating the end of the internet, and that in fact they are extending and enhancing the very virtues that you and many others hold to be valuable. I tried to do this by ccorrecting a number of technical misunderstandings that seemed to indict Internet2 as a villain, when in fact the opposite is the case.

So I hope it won't be too disappointing if I don't respond to your second post in a point by point manner. It seems to me that most of the concern there is really more about the MPAA and the broadcast flag than Internet2.

Internet2 is indeed talking to the MPAA, but they are talking to literally hundreds of organizations and interest groups. Some of those groups hold opposing views and differing visions of the future. It is in everyone's interest that Internet2 provide a forum for as broad a discussion of advanced networks as possible.

And I don't want to be put in the position of defending the broadcast flag. I can see issues and interests on both sides, and find myself
somewhere in the middle. But I'll toss in a few thoughts nevertheless.

First, it's important to remember that more than one market force is at play here. Yes the MPAA (and RIAA) wants to protect the property rights of those who create and market media. But the consumer electronics industry doesn't want to see the end of home recording. The carrier companies (cable, satellite, ISPs, etc) don't either. And consumer groups still have a voice. (And so does our democratically elected government.)

I'm convinced that when all is said and done the typical consumer will still be able to record at home for all the fair use reasons currently available to them. The MPAA has said that even they want home recording to be preserved. Will there be transitional problems? Will old equipment become obsolete? Sure...as always. Ask anyone who went with Beta rather than VHS. Or audiophiles who thought the Elcassette would lead them to sonic nirvana. Such is the nature of

Next, regarding hackers and the ability to innovate and experiment with broadcast media. The broadcast flag, to my best understanding, has to allow for not only hardware recording devices, but also computers used as home entertainment centers. Can you imagine Microsoft not demanding this? And to keep the competition fair third
party software vendors will have to have some way to create products as well.

As a programmer what this says to me is that operating systems will have to provide a software layer that will allow playing/recording/skipping/looping video media while preventing (or attempting to prevent) massive piracy. Those software hooks will
have to be available to any programmer...even kids and hackers...because ultimately they will be impossible to hide anyway.

Perhaps someone else will come up with an example, but under such a scenario I can't imagine functionality that is short of piracy and
yet unavailable to random programmers. I'll admit that there is some speculation in the above...but this is all a work-in-progress and there is speculation on all sides...even on the EFF site.

Getting back to Internet2. A few quick points.

"Pick-up collaboration" on Internet2 is indeed live and well. But guess what? Artists didn't invent it. Scientists are leading the way there. They are also the ones who invented the World Wide Web. Nevertheless, both are available to artists as open platforms for creativity. Have at it!

And yes, the Internet2 Commons has a fee attached to it, but you have to understand what you are getting. Standard videoconferencing (with Polycoms and Tandbergs and so on) is limited to 3 or 4 sites at a time. If you want to include, say, a dozen locations you need a device called an MCU. Along with the MCU hardware cost there are also maintenance costs and administrative hassles. For many schools buying and supporting their own MCU's is prohibitively expensive. And contracting for external MCU services is really expensive too.

For many schools the Internet2 Commons provides very useful functionality. Rather than tax every Internet2 member they decided to fund the effort by only charging the schools that want to use it.
Compared to the commercial alternatives the I2 Commons fees are a really good deal.

There are, of course, other ways to videoconference. iChat on the Mac is cool...as long as everyone else is using a Mac and you only
need to connect to a couple other people. The Access Grid is great, but it requires multicast (perhaps via a unicast gateway) and isn't exactly plug and play or commonly used.

For connecting random sites nothing is as ubiquitous as good old H.323 and H.320. Check out last years megaconference. *372* sites on every continent but Antarctica connected via video and voice.


Regarding putting low level DRM into routers. All I can suggest is looking into what it would really take to get such a protocol, or *any* new extension, into IPv6. At most Internet2 could sponsor a proposal...not that I think they ever would. And then there would be an *international* standards process to contend with. I don't care
what the MPAA may or may not want...it just ain't going to happen.

Finally, regarding the better documented Internet2 performing arts events. You have to remember that many of these events are designed
for a certain kind of setting. More often than not the setting is a large conference for an audience of several hundred university technicians and administrators. Such a setting invites a rather standard "concert" type presentation...and comfortable mainstream content.

But this is hardly built into the network!

And the master class thing may not be your cup of tea, but in large parts of the country distance education, and access to the talented people that tend to migrate to urban centers like NYC, is a
significant breakthrough.

There are all kinds of other options waiting to be explored. Way back in 1999 NYU's first use of Internet2 involved small performances, intimate improvisations, and other artistic "pick up"
experiments with theater students at MIT. More recently NYU Professor and performance artist Barbara Rose Haum did a very nice piece with collaborators at the University of Kansas.

Personally, when it comes to MARCEL I am less interested in more academic theory. What I'd love to see MARCEL spawn is more actual art. And I am sure that as soon as an Alan Kaprow for the network age wants to reinvent what we mean by "art" and "performance" Internet2 will be there for them.


Posted by: Jo at March 12, 2005 11:46 AM