Opening the Portal Into Your Television
An Interview With Carl Malamud

by Bruce Stewart

Carl Malamud is known for projects such as Internet Talk Radio,, and putting the SEC's EDGAR database on the Net. He is one of the founders of the Internet Multicasting Service, and is currently working on NetTopBox -- an attempt to route around the heavily-patented, multi-billion dollar Electronic Program Guide industry and build a distributed, open media guide. Carl will present a session on NetTopBox at O'Reilly's upcoming Open Source Convention.

We recently spoke to Carl about his current work and the need for an open program guide.

Bruce Stewart: What is the Internet Multicasting Service?

Carl Malamud: We're a non-profit research corporation, known for a series of projects on the Internet we've lumped together under the rubric of "public works." We've been around since 1993 when we started the first radio station on the Net. We're probably best known for putting the SEC's EDGAR database online, running it for two years, and then getting the SEC to take over the service.

You can see some of our past projects on our site.

Stewart: Since you founded Internet Talk Radio in the early days of the Net, have you been keeping up on the current controversy surrounding streaming radio stations on the Web, and if so what are your opinions about it?

Malamud: By controversy, you mean the CARP copyright review that set obscenely high rates for streaming music?

Stewart: Yes.

Malamud: Bleah. I went and saw the copyright office in 1993 because BMI and ASCAP wouldn't count Internet Talk Radio as a public radio station, even though we were dues-paying members of the public radio satellite system. They said they'd help, but they ended up studying the issue. After 9 years of deliberation, I'm afraid the copyright office didn't get it quite right.

I suspect one of the radio stations with deep pockets or a law school close by will take the CARP rates into court and that we haven't heard the last of this.

By the way, when we were running our radio station, we didn't really have the option of playing music since we couldn't get licenses from BMI and ASCAP. So we concentrated on developing our own content, and I must say I'm glad we did that. We got to work with the National Press Club, the Kennedy Center, the U.S. Congress, and a bunch of other very official places, and we also did a lot of live broadcasts followed by archived audio/video/images at places like the IETF and Interop.

So CARP may not be the end of Internet Radio, but it might force some groups to start thinking of how they can create new content instead of simply spinning platters you can already hear at much higher fidelity with a $20 CD player.

Stewart: Tell us a little about the NetTopBox project.

Malamud: Electronic Program Guides are a little-known area, but a pretty crucial one as consumer media starts to meet the Internet. You wouldn't think that the idea of a directory of programs on the tube would be novel, but the Patent Office, after a careful search of their own files, determined that no patents had been issued and this was therefore new and then proceeded to award a few hundred patents.

That means the gateway into your television is now the private property of a couple of big media companies. And, if you look at TV Guide®, you can see that innovation isn't exactly flourishing in this area.

The NetTopBox project is an effort to open the portal into your television. We think that's important whether or not you believe in convergence. If you think TV-land is different from PC-land, you still need to be able to control your consumer devices from your network (or live forever in a world of a million remotes).

If you think media is going to show up on your laptop, you should be even more concerned about patents that limit what you can do to find media you like more effectively.

Stewart: Which companies have received the majority of these patents?

Malamud: Gemstar is the price-performance leader in this field, though there are plenty of other bogus patents out there. For example, CDDB got a patent for taking a hash of a CD-A file header and sending it to a server for lookup ... go figure.

Stewart: What motivated you to start working on NetTopBox?

Malamud: It seemed like an area that could use some help. We got done helping the U.S. government with their databases, so we figured we could lend a hand in this area.

Stewart: What is your timeline for achieving an open source electronic television program guide?

Malamud: ASAP. Seriously, we're looking at a 2-to-3-year project. We're not convinced that we're the ones building an "open source electronic television program guide" ... we don't want to be Gemstar and there's plenty of companies, coalitions, and communities out there trying to do better than them.

I think our contributions will be more strategic. We'll pick a few fights and try to win them rather than trying to be Google meets Gemstar. We'll explain a bit more of our strategy at OSCON.

Stewart: Can you give a little background on Gemstar? Are they the ones trying to get a lock on this type of information?

Malamud: It's more than Gemstar. It's a mindset at the Patent Office, in the venture capital community, in big media companies, and in startups ... they're all trying to build walls around our public parks and charge admission.

Companies like Gemstar deserve all the market cap they can get from Wall Street, but that market cap should be from value-added information, not by running the only toll bridge allowed to cross the swamp into cable land.

Stewart: What have been the biggest challenges so far in this project?

Malamud: Figuring out where to start and how to do so without incurring seven-digit legal expenses.

Stewart: What do you think of products like TiVo and ReplayTV?

Malamud: Love them! They changed the way we talk to our television.

Bruce Stewart is a freelance technology writer and editor.

Return to the O'Reilly Network.