Message: 4
Date: 8.1.93
From: <nicholas@media.mit.edu>
To: <lr@wired.com>

Set-Top Box As Electronic Toll Booth: Why We Need Open-Architecture TV

Is Bill Gates using John Sculley's speeches to guide his alliances? The makers of computer hardware and software evince uncanny synchronism in their lusting toward the cable industry. This is not surprising when we consider that ESPN has more than 60 million subscribers. Microsoft, Silicon Graphics, Intel, IBM, Apple, and HP have all entered major agreements with the cable industry.

The object of this ferment is the set-top box, currently little more than a plug adapter but destined to be much more. At the rate things are going, we may soon have as many types of set-top boxes as we now have infrared remotes. Such a smorgasbord of incompatible systems is a horrible thought.

The passion for this box stems from its potential function as, among other things, a gateway through which the "provider" of that box and its interface can become a gatekeeper of sorts, charging onerous fees for information as it passes through the gate and into your home. While this sounds like a dandy business, it is unclear if it's in the public's best interest. Worse, a set-top box itself is short-sighted and the wrong cynosure. We should broaden our vision and set our sights instead on open-architecture television (OATV).

Smart Boxes Are Not Enough
What's going wrong? It's simple. Even the most conservative broadcast engineers agree that the difference between a television and computer will be limited, eventually, to peripherals and to the room in which it is found. Nonetheless, this vision is being compromised by monopolistic tendencies and by an incremental improvement of a box to control 1,000 programs, 999 of which we don't want to watch. Nonetheless, OATV - where computers and television become one - has so far been seriously out-boxed in the first round.

The word "box" carries all the wrong connotations, but here's the theory. Our insatiable appetites for bandwidth (see my column in WIRED 1.3) put cable television in the lead position as a provider of information and entertainment services on demand. Cable services today include set-top boxes because only a fraction of TV receivers are cable-ready. Given the acceptance of this box, the idea is to aggrandize it with additional functions - give it long pants, so to speak. But this cannot be the right approach.

The Under-Set Pizza Box
When Sun Microsystems introduced the SparcStation 1 in 1989, its chassis had the form of a pizza box, which launched a trend for under-monitor electronics containing all the elements of an "open system" (credited to Sun as well). Map this simple change of thinking into a television receiver, and imagine an under-set configuration with more computer-like modularity and an expandable chassis. In such a world, the future of television is more clearly seen as data broadcast. At first, most of the data will be video, but eventually there will be other services, including data about the data (as I suggested in the last issue, the medium is the model). An open systems approach is likely to foster the most creative energies for new services and be a vehicle for the most rapid change and evolution.

I have argued that number of scan lines, frame rate, aspect ratio, pixel shape, and "interlace" versus "progressive" scan rates were non-issues and should be variables, not religions or laws. During a recent congressional hearing, Representative Ed Markey patiently listened for over an hour to arguments about why interlace and square pixels would help job growth in America. Give me (and him) a break. Of course interlace has little place in the future, and perhaps someone will come up with an interesting application for non-square pixels (hard to imagine), but to suggest that either of these be legislated is simply silly. Let the under-set computing engine worry about that, not Congress or the FCC.

OATV Can Be Topless
It is time to make the leap to an OATV world, not limit our vision to an expanded and proprietary cable set-top box. In an OATV world, the monitor itself can be an option. The bits may be video, or they may not be. They may be audio or data destined for online services or in-home printing of a personalized newspaper. It's as if one must yell at the concatenated players of the FCC's HDTV bake-off - the so-called Grande Alliance - "It is not just about TV."

The recent alliance of Microsoft, Intel, and General Instruments (GI) to develop a set-top box and TV operating system is very revealing. Add GI's predominance in the cable industry to the enormous power and computer savvy of Intel and Microsoft, and the result is a formidable consortium, if not a cartel. Why would this truly "grande" alliance focus on the set-top box? Surely the members must see the bigger picture. (Perhaps the problem is that Microsoft thinks MS-DOS is an open system.)

Why Open Systems Are Important
Open systems are not just about being well documented. They have other properties deeply rooted in the simplicity of an extensible standard, most or all of which could and should be in the public domain. This is important to the growth of new services, third-party equipment, and the kind of international sharing that makes the Internet such a phenomenon.

Why not learn from Wang, Data General, and Prime? What those once high-flying companies had in common was a total disregard for open systems. Open systems exercise the entrepreneurial part of our economy and call into question proprietary systems and broadly mandated monopolies. In an open system we compete with our imagination, not with a lock and key. The result is not only a large number of successful companies, but a wide variety of choice for the consumer and an ever more nimble commercial sector, one that can change and grow.

This may not work for automobile manufacturers but it does for the computer industry and it can work for television. The reason is simple: None of us give a damn about the box; we care about programming. Just as software and system services drive the computer industry, programming and intelligent browsing aides will drive the television industry. Ask yourself: Under which scenario will we see new media and the most innovative content - one featuring an enlarged set-top box, or one featuring open architecture television?

Next Issue: Modern Multimedia

[Back to the Index of WIRED Articles | Back to Nicholas Negroponte's Home Page | Back to Media Lab Home Page]
[Previous | Next]

[Copyright 1993, WIRED Ventures Ltd. All Rights Reserved. Issue 1.04 August 1993.]