Interview: Andy Wolfe, former CTO, ReplayTV

[The Hollywood studios] essentially wanted to control what anyone could record on TV. They wanted sole discretion over how long you could keep a show after you recorded it. They wanted to limit how many episodes of the same show you could record. They wanted to ban thirty-second skip buttons and to prevent fast forward from reaching a certain speed. — Andy Wolfe

Andy Wolfe was chief technology officer of Sonicblue, the company that acquired ReplayTV and shipped the first units with video-sharing capabilities. He reveals what the Hollywood studios and television networks were really after in their lawsuit against Sonicblue's ReplayTV: control of your television. He spoke by telephone with J.D. Lasica.

Are you familiar with the term file serve television?

We actually think we invented it. The Replay 4000 shipped in 2001. It was the first networked PVR. You can connect two of them together and you can file serve from one to the other. A year and a half before that, in 2000, we did the first file serve music with a device under two names. We put it out as a Real receiver under the Dell brand. And that was a thin client that you could put on your network and go back to your computer to serve MP3s. We built that as a theme across a number of our products.

We also did a dedicated product called the Rio central, named one of the best in show at CES 2002. Seeing that product was one of the reasons that led us to buy ReplayTV. They had a proposal to do this networked video in the home on the next platform. They didn’t have it designed yet, they didn’t have it developed, but they developed some core technology, and that was why we bought ReplayTV and that was the product we brought to market. Internally, to us, that was always the killer app.

I have 3 of them in my house — two 4000s and the 4500 – and they’re all networked and they all talk to each other.

Tell me the backstory of ReplayTV.

ReplayTV was a dotcom that had raised $150 to $160 million funded by Anthony Wood and Kleiner Perkins and others. And then the networks came in: NBC and AOL and lots of industry partners. In the end there were over 100 investors. They burned through all their capital. We came in, late 2000 at Christmastime, and they were close to bankruptcy, and they had stopped manufacturing products and they were gonna become an IT house. They asked Anthony Wood to come back in and sell the company. Anthony rallied the troops around the idea of building this network platform, client-server technology into the product. We actually had been working on the same idea. We worked with another company, Sensory Science, a Linux based Intel architecture box that played DVDs and was really a home server, but when ReplayTV came in we saw they were further along in software development and held patents on personal video recording, we bought both companies. By December 2002 Replay was up to a 35 to 40 percent share of the PVR retail market.

TiVo was the other big dog.

They both showed up together at CES in 1999. TiVo shipped a month or two before. In March 2000, TiVo got a big influx of cash by going public. Replay was scheduled to go public in April 2000, but the market suddenly soured and the IPO got canceled.

And file serve TV was one of Replay's big selling points.

That was the model, file serve TV, and it was inside and outside the home. So we designed the Replay platform to be able to serve from one to another; we designed it to be scalable with a thin client so you could record it on one Replay and access it on a thin client built into a TV, and we designed it using ichannels, an outside-the-house file serving technology. The idea was to build servers that would operate commercially and you could purchase content. It would be downloaded to the box in your house. And that’s where we got into fights with the industry. When we went out and proposed our own DoD servers around the ichannels, that’s when things started to get nasty.

We went to the content creators proposing to use their content, with them getting 65 percent of the revenue, and they basically said no. The same thing happened in the early days of the digital music era, none of the music labels wanted to license their music.

Why were the studios so opposed?

I might talk about it after the lawsuit dismissed. But a lot of it is, they just don’t have these rights. The truth is nobody has the rights for Internet distribution for a lot of this stuff. The contracts they have give them the rights for certain kinds of distribution, and they’ve also given away certain kinds of rights exclusively. For example, they’ve given HBO and Showtime exclusive rights for subscription services, so they can’t license their content to Internet subscription services. If you wanted to have a service where you pay 20 bucks a month and get 10 movies, nobody has the rights to do that because they’ve given HBO and Showtime the rights for subscription services. I haven’t seen the contracts. At least they’re worried that it would.

The other thing is that in a lot of cases, it’s not that they don’t have the rights to the show — they don’t have the rights to the music. Because when they signed the contracts, it was originally intended only for broadcast TV.

Eventually that’s going to change, they’ll realize there’s a business in licensing this concept. But the big issue for them is, they don’t want you to have stuff archived in your house. And that’s the model we are starting to get to. I could record everything I want to watch and decide later which I want to watch. I could record 50 shows a day and watch only one of them. The truth is, very soon you won’t have to throw away the 50, you can store them for a year. Or forever. And that scares the hell out of the studios.

If you’ve ever talked with them, they know you’re not gonna watch the commercials for recorded material, for anything that’s not live. But the other thing is, if you talk to people in the industry, they’ve got this buggy whip problem. For 50 years they’ve been taught that their brilliance is in programming, and programming is the art of deciding what shows get watched when. They get you to watch shows that come on after other shows, or to attract a particular demographic to increase the value of advertising dollars. If you sit down with TV execs, that’s what they think their skill is. Not creating quality television, somebody else does that. Their art is in programming, and that’s going away. And they haven’t learned how to make the adjustment. You’ve taken away their tool for manipulating the public or their tool for extracting the value of their assets. They haven’t learned how to make that adjustment.

Did you chiefly deal with the networks or the Hollywood studios?

It’s interesting. The New York people were always interested in new business models, and the L.A. people would tell them to shut up. You always expect New Yorkers to be the arrogant ones, but the people we were dealing with from the relatively younger companies like HBO, Showtime, and MTV always wanted to listen. It was the L.A. people from Disney and MGM — relying on a business model developed more than fifty years ago — who ultimately said, ‘You don’t understand our business, you’re a threat to our business,’ and they set about to crush us.”

I thought the main thrust of the lawsuit was skipping commercials and network sharing.

Not true. If that had been what the lawsuit was about, it would have been settled in a few weeks. That was a smokescreen. They knew those things could have been negotiated.
we believe to this day we would have won in court. The real issue involved a couple of things. They wanted to be able to control what you could record — the same thing involved in today's clash over the broadcast flag. They essentially wanted to control what anyone could record on TV. They wanted sole discretion over how long you could keep a show after you recorded it. They wanted to limit how many episodes of the same show you could record. They wanted to ban thirty-second skip buttons and to prevent fast forward from reaching a certain speed. They wanted to limit the total amount of the stuff you could record. They came up with a number, but I can’t tell you what it is.

Have we already surpassed it today?

On my laptop.

So those were all things that we felt were non-negotiable. They conflicted with existing usable models. Nobody makes videotapes that self-destruct, right? Or that lets you record The Simpsons only once a week. And we just didn’t want to get into regulating customer behavior in those ways, and we didn’t think there was any law that backed it up at all.

As far as sharing files with others, they had an argument. At least there was some level at which those things were legitimate. We were pretty clear that somebody could use Send Show in a way that wasn’t permissible. But that’s never been the basis for getting rid of something.

As for commercial skipping, it’s never been clear that they have any defensible position there – at least it’s a very obscure argument. Because they were saying that if they were suffering economic harm, it was by default unlawful. I’m not sure they even got close to making their case there.

But the other stuff was just — we want our way. That their business models were being threatened. We couldn’t imagine a judge sitting there making a ruling that nobody will ever be allowed to have hard disks recording video bigger than this many gigabytes.

In ReplayTV's Send Show, how did you settle on the figure of limiting it to 15 people?

There were two functions. There was streaming in your house, and that was limited to eight units, and there was Send Show, where you could email a show to somebody else, and that was limited to 15 people and only second-generation copies. If I sent it to you, you couldn’t send it to anyone else. By the time we made that final decision, we knew there was a possibility of a lawsuit. We wanted to make sure there weren’t people running commercial video distribution services on our platform. That wasn’t what it was for. It was for: Hey, I saw this episode of this great new show and I’m going to send it to my friend, or look, my kid was on the news today, I’m going to send a copy to Grandma.

Fifteen was just a number that prevented any commercial use.

An average customer sent one show per month. We never knew of any commercial applications built on top of our platform. Why bother with it, when with the Internet you can just send what you want?

That’s why we were amazed there was such rigamarole around this. We sold 60,000 of these things. ATI sells a million cards a year that lets you record shows and attach it to your email. They still do. Sony sued us, but they let you record stuff on your Vaio and burn it to DVD and email it to anyone. AOL lets you attach a show to Instant Messaging. It’s amazingly hypocritical, with these companies and their software that are out there and can do all this stuff on the PC, they took it for granted, but when we came up with this device that ordinary people could use, they panicked.

Is ReplayTV a computer?

It’s a computer but it’s all proprietary software. You couldn’t get into it and run your own software. It had a hard drive and processor and a PCI card.

How do you see the file-sharing wars playing out?

My theory has been, the only way to get away from the free-for-all piracy situation is to have these boxes where you have access to value-added services. Put all the back B shows up for sale, or have movies on demand, or have streaming music with the latest bands, and have that so it only works with boxes that have somebody’s logo on it. And then people will use the boxes with the restrictions as long as they’re not too unreasonable.

It’s easy to get stuff online. People aren’t going to want to swap shows in great numbers – they’re on these file sharing networks because they can’t find the stuff through legitimate means.

Some of the other stuff we went through was kind of silly. The networks have this concept that nobody ever records their programming. And yet, right in the middle of our discussion, the FX Network showed all 24 episodes of 24 from midnight to midnight, and we asked them, You think that many people sat through it? I mean, you guys program knowing that people will record this stuff, it’s your business model.

Were you dealing with the CEOs?

Never. The lawyers and the business development people. In the early days, we dealt with the presidents of HBO and Showtime and the small networks.

Where do you see all this going?

Three possibilities. One is that everything becomes centralized, and the cable companies are gonna win and they’ll offer services delivered to your set top box. A least common denominator thing. They own the content, they’ve got their own servers, and you subscribe to pay per view and get the shows from the cable provider. It's low quality, low service level, very few choices in terms of product, but they have the relationships in terms of licensing contracts. Your box will offer only a few features. We're starting to see some of that with Time Warner.

The second possibility is that Microsoft drives it, that everything lives on your PC as a client server, and there’s a cheap thin client box that connects your TV to the PC. They’re pulling strings in the industry to make that work.

The third possibility is a retail play. Now it’s likely that tier 2 or tier 3 guys like Apex or another company in China tries to copy what ReplayTV and TiVo have done. And they build boxes that have no rules, that are feature heavy. They don’t care what the networks say, they can change brands or owners. An Ethernet is a $3 connector these days. We did it the first time and it was hard because there were performance issues and software complexity, MPEG decoder chips were complicated and buggy, but all that stuff has been worked out. The guys who make DVD chips are now all making PVRs on a chip and now they can pressure TiVo and Replay to decide what goes in the product. But once there are 50 different brands out there … Apex became one of biggest DVD brands by getting the parts, slapping them together and selling them cheap.

So you don't see TiVo or a DVR-driven DirecTV as part of the equation?

DirecTV maybe. They’re in chaos because of the merger. They and Echostar will be part of the story. In the end they’re the same as the cable providers. They can’t cross any lines or else they start losing programming from the networks. So they’ve got a tough line to walk. They will generate some of the PVR volume, but they can’t do too many innovative features. They can’t do video on demand because they’re satellite, they don’t have enough bandwidth. All they can really do is in-home recording. Will they allow you connect together in a client-server environment and build your own libraries? There’s not much incentive for them to do that.

But they can download stuff to your box, right?

They can, and they might. There's no indication yet that they’re thinking that way. But they have copyright issues. They have a contract with the networks and for them to save stuff on the box may be in violation of their contract. There's a difference between me deciding to save stuff on the box and a satellite company. I have a fair use right to do that as a consumer. For them to do so, they need specific commercial rights to do that, they can’t save stuff on the box without permission.

TiVo is in a tough spot. They have the brand, but very quickly other people are sneaking up in terms of price. They had network executives on their board of directors. At certain levels we had an open relationship with them. They would tell us we’re stupid for picking fights with the studios, and we would tell them they’re stupid for spending $400 million on advertising. Because we didn’t have $400 million to spend on advertising we needed some other way to get the attention of customers, and we did that by putting in features that customers were demanding and no one else had.

Did the entertainment companies' lawsuit bankrupt Sonicblue?

Not even close. At the end, the lawsuit cost $1 million a month, and general expenditures were $7 million a month over that. What bankrupted the company was that we grew from an $18 million CE company to a $400 million CE company in 24 months and we never built the infrastructure to support it. We were leaving millions of dollars in the supply chain left and right. We had millions and millions of dollars in supply chain problems. Plus we were $56 million in debt from the purchase. But lawsuit was almost irrelevant, it represented only 3 percent of the expenses over a two-year period.

What do you make of the file sharing of TV shows on the Net?

Intentionally, we made certain that we didn’t have a clue about that stuff. I’ve never seen it, I’ve never tried it. We occasionally talked to people who were involved. But we didn’t know much of what was going on.

You look at what gets file shared, it’s not what’s on TV. It’s to rip off movies still in theater that are not yet on DVD. Occasionally I have some sympathy for these guys. You can’t say to them, hey, the day after something appears in the theater you should put it on DVD so that nobody steals it.

What about grassroots programming?

We did a marketing study and found that two things were in high demand: porn, and Bollywood, because Indian films are not widely distributed in the U.S.

That’s part of why this whole thing got a little threatening. We think that if there was a real service, that independent content would become an important part of that service. We didn’t think people would sign up for a service if it only had independent content. They’ll sign up for Harry Potter or Terminator 3. It’s the blockbusters that get people’s attention.

We got a call from churches who wanted to distribute their sermons on Sunday mornings by sending videos around. There are also surveillance applications. We found lots of people who were interested in building new things on top of this. We felt these other things would follow, but the entertainment had to drive it.

Will file-serve TV happen in our lifetime?

It’s gonna happen in the next 5-7 years. If one of the powerful guys gets out there first and gets momentum, it’ll be corporate. If they don’t, then the equipment will get spun off and it’ll happen anyway without any structure around it.

You mean some facet of this on-demand service, but you’re not talking about universal jukebox.

Well, what I’m saying is if it doesn’t happen legally, I think somebody will do it offshore as an internet service.

Interview conducted June 23, 2003