Silicon Valley Speaker Series: 25 Years of Innovation
Sept. 19, 2000
In recognition of Microsoft's 25th anniversary celebration, this month's panel discusses 25 years of industry innovations.

25 Years of Innovation

A roundtable discussion among Steve Capps, Heidi Roizen and Tim Bajarin, moderated by Dan Gillmor.

Mountain View, Calif., Sept. 19, 2000

PHIL GOLDMAN: Let me start off by introducing our moderator, Dan Gillmor. Dan is -- I have this long, long list. Dan asked me not to read the entire list, but I probably will -- this long list of credits. Dan is a columnist for the San Jose Mercury News, and he joined the paper in September of 1994. Before then he spent about six years with the Detroit Free Press, and was with the Kansas City Times, and several newspapers in Vermont, and I wonder how many there are in Vermont. And he's freelanced for a number of publications, including the New York Times, the Boston Globe, and the Economist. He's also a frequent contributor to NPR. He's a long-time industry observer, although we'll get to observe him today. And he brings a keen eye for technology and business trends to our discussion today.

So, please welcome Dan and our panelists Steve Capps from Microsoft Research on the end there, Heidi Roizen with SoftBank Venture Capital on the other side there, Tim Bajarin of Creative Strategies in the middle, and also James Joaquin of Ophoto in the middle. Thanks very much, we're looking forward to it.

(Applause.)

MR. GILLMOR: Thank you, and thanks for not reading everything. When I was asked to come to do this, I mentioned it to a colleague, and his response was, "Well, someone up there has a sense of humor." Actually, if you haven't heard some of the news of the day, I can give you an inside scoop on what my column for tomorrow is going to be about. The news that it will refer to is Joel Klein announcing that he's stepping down at the end of the month from the Justice Department.

(Laughter and applause.)

MR. : I'm assuming that he's going to some big law firm, right?

MS. ROIZEN: We're funding his startup.

MR. GILLMOR: There is no truth to the rumor that he's replacing -- (inaudible) --

The format of this discussion is basically, they talk, I listen; then you ask questions. And then I'll take some notes and ask a few questions myself.

This is really an illustrious panel, apart from your moderator. I'm really interested to hear what they -- (inaudible) -- and what's really gotten to them -- (inaudible) -- down deep about what -- (inaudible) -- technology has changed, and it really has been an astonishing 25 years, this last quarter-century. There's never been a rate of change that comes close to it in human history. So, rather than read long introductions, what I would like for you now is to start with Steve and come down is to ask each of you to describe briefly a little of their history, and then to answer, to the extent that they want to, one of three questions that -- (inaudible) -- one is, just looking back at what might be the most single important innovation of the last 25 years, how innovation has changed the way we work, and play, and entertain ourselves; and what innovations or potential innovations are going to have the biggest impact moving forward.

So, any or all of those take a crack at it, and we'll come straight down the row, and then see where it goes in a free form interaction. So, Steve.

MR. CAPPS: Sure. I guess I was lucky enough to work on the original Mac, that thing called "Finder" which people had often confused with the system, which was just a little program that used the shell icons around. Another guy and I worked, and we -- (inaudible) -- but anyways it was a really amazing thing because in a very short amount of time, when I think back about designing hardware or software, one of the things he did was set up a developer organization, and it was very cool that you could actually do that with such a small group of people.

Then, after that, I kind of cast around looking for something else to do, and I went and talked to PCs, and they all said, you have too many ideas. You've got to focus on one. I said, well, that's no fun. I would much rather work on five things. So I ended up going and calling up a friend of mine who was VP of Microsoft, and I called up and said, hey, what should I do here? And he said, well, come work for us. We like people with five ideas.

So, I've been kind of down stomping around there thinking about UIs of the future, and within six months of stepping away from UI I kind of said, I think we've got lots to go there. I think we stopped in 1984 on really having been that much great innovation on the UI. And I actually have -- (inaudible) -- if you do a little multimedia presentation. Since this is 25 years, tell me, just buzz the buzzer if I'm taking too much time.

This is a computer I programmed in high school, which is a little bit more than 25 years. First of all, the title, I think, is wonderful on the panel, it's called Control Data. And this is the real panel. But this is how you debugged it. You sat in the thing and you said, oh, the memory address register is doing this, and I'm amazed that I actually remember these acronyms and what they still mean. But that was my life when I was in high school. And, as you can imagine, that was actually in the '70s, it was kind of fun.

But I look at it today and I go, well, we kind of have the same thing, you know. But the icons just don't blink. So, anyway, so that's kind of my job is to say, we're doing a great job of shipping file system debuggers. And the user really doesn't need a file system debugger. And when you think about somebody buying a computer, a Mac, a Windows, anything within five minutes, they have to understand a hierarchical file system.

MR. : May I interrupt? That's the most efficient PowerPoint demonstration I've ever witnessed.

(Applause.)

MR. CAPPS: I know you'd like to hear me ramble on about ten more minutes about hierarchical file systems, but I'll just hand the mike over to JJ here. But what I do is like be a thorn in everybody's side saying, we shouldn't be doing what we're doing.

MR. : We're going to come back and ask you what we should be doing.

MR. CAPPS: Sure. Anyway, and I won't answer one of your questions because there's not going to be a quiz about them.

MR. JOAQUIN: Okay. Well, short and sweet from Steve. Long and sweet. My name is Jim Joaquin. I had the pleasure of working with this gentleman at Apple for four years on the Newton project. I was at Apple for six years. I'm actually too young for this 25-year panel. I feel like Harry Potter, somehow I got snuck into the tournament.

But I started programming on minicomputers, and then workstations, and then personal computers since I was a youngster. And at Apple went from managing the engineering team to product marketing, which really meant that you were an extroverted engineer in the Apple days, and from there kind of crossed into the dark side of management in managing teams, and now running the company.

I started my first software company in college with four other guys called Clearview Software, which got bought by Claris, if there are any old Mac folks here, you remember a product called Smart Forms, it was called Stupid Forms at Apple. That was my first commercial software product. And most recently I was co-founder of a company called When.com, which is a Web-based calendar service, and sold that company to America Online last year. It's now the calendar that's in AOL 5.0. So we've got millions of folks using that. From there, I started my current company called Ophoto, an online photography company.

And in terms of answering the questions, I think one trend that I personally have been focused on is looking at how we as humans use applications and use software. I've been building software and working for software companies for a long time, and the big innovation and the big shift that I've seen is that the software has gone from the desktop out to the network, out to the server. And I think where that works the best is where it's transparent to the user.

So, you could argue that there are great innovations like TCP/IP and the HTTP protocol and all this great stuff, but for the average customer I think it's amazing when an application like email -- probably a good example for this campus -- becomes server based, and you have Hotmail, and you have this notion of being able to access your email from any computer that's connected. And your files are no longer trapped on your desktop. And that was the idea with When.com, that we could take a calendar, and we said, you know, we think the calendar-connected application can go from the desktop to the server because certain classes of applications get so much more powerful when you connect them to a network because they connect you to other people, so you can add community to your application, and they can also connect you to content.

And the focus for us with the When calendar was this idea that your calendar could tell you when your favorite events were happening. So we brought in about three-quarters of a million events updated daily so you could find out when your favorite concerts were happening on any given day on your calendar. Your favorite sports games, symphonies, TV shows, et cetera. And that really transformed what you could do with a desktop calendar. And it really wasn't a simple transformation to what happens when you put it out on the server, and there's a lot of obviously technical changes to how you build server-based applications. I think that's been the other innovation for Silicon Valley is, you look at; do we have the right code, the platforms that we develop on, the hardware that we use. This notion of server-based applications is a really big change.

What I'm currently doing with Ophoto is tied to that, there's been a singular evolution in the world of photography, which has been the advent of the digital camera, and that's really changing how people take and share and photograph and celebrate the human experience with photography. And my co-workers and I think that there's a better way to do that with this notion of actually having your photographs live up in the cloud on the server, and you can access them any time, anywhere, and you can also share them with other people, and you can really extend your reach through photography, but it comes back to this idea of applications moving from the desktop out to the server. So, I'll throw that out there as kind of my favorite innovation topic.

MR. BAJARIN: My name is Tim Bajarin, I'm president of Creative Strategies. And actually, Phil asked that question 25 years ago, and I had to actually calculate backwards 25 years. I was actually right out of college, I went into semiconductor R & D. And in '81 I went to Creative Strategies, which in those days was a traditional market research firm. Over the last 10-15 years, though, as I became more involved with the company, I began taking it more and more to the strategic planning roots that it started with. And so now I spend probably 85 percent of my time at the strategic planning level up at the top with most of the big companies, or a lot of big companies, a lot of projects, in fact, with Microsoft, where we try to help our customers figure out what their customers want 12 to 24 months from now.

We do a lot of big picture thinking of 5, 10, and 15 years out, but to be honest with you, the design cycle is so short now that you've really got to be thinking about 6 to 12, 18 months, not much farther than that. So, our role is at the strategic planning level to help companies figure out what their customers are going to want in the short-term future, which is realistically 12 to 18 months.

I have a prejudice up there, but honestly the most important innovation in the last 25 years is the microprocessor itself. I mean, we're here because of the microprocessor. But if I look at the technologies that really began changing the way we -- especially the way we work, it would be things that are related to communications, the cell phones, some of the pagers that came out. And one particular one that I had a big involvement with, in fact I actually met some of these guys at Apple in those days, in 1983 I wrote the report on printers. And in that printer report, I made a statement. I had had the privilege to be shown the first laser engines that could be used on desktops back in '81-'82. And I don't know if any of you guys go back to that time frame, but a laser printer back in 1981 and 1982 started at $50,000 and it was a half a room. It wasn't a desktop. Anyway, when I did this report on printers I said in that, I said, I can imagine someday that if we get this printer into smaller devices, that we could some day publish on the desktop. And this was way before the desktop publishing phenomenon. And when Apple was struggling with concepts in 1984-'85, they asked me to come in and do some of the original thinking on some of the marketing with that. And actually think desktop publishing is one of the technologies that's had probably one of the most important impacts in the process of business communication going forward.

And, interestingly, at that same time, within about a month or so of finishing the project we were doing on desktop publishing, John Sculley came to me and asked me to take a look at presentations, and could presentations have the same kind of impact, desktop presentations have the same kind of impact on communications as desktop publishing. And, at the time, we said that we thought that it could potentially have a dramatic impact on the way we communicate, but not as much as desktop publishing. And it was ironic because Apple, of course, could have capitalized on it, the company that really capitalized on desktop presentations was Microsoft with PowerPoint, of which now you see PowerPoint presentations being used in not just business, but in churches, in social groups, in political campaigns. It's had just about as much as an impact.

And the other thing that I'm most fascinated by, and I'll get into the future stuff as we spend more time, I get to see stuff that's going on in labs in just about all the big companies worldwide, so I have a good feeling of what's going on, and we'll talk about that in a minute. But I think that ultimately the Internet itself was able to capture that process of communication so that you were able to use voice, telephony, presentations, desktop publishing, a lot of these things, with the Internet now becoming the network that's connecting them all together is a culmination of a lot of the stuff that took place over the last 25 years, and has become the next really significant publishing.

MS. ROIZEN: Well, Dan made a mistake when he asked me where I was 25 years ago, and I was a junior in high school. And the thing I think that's relevant about that is, when I was a sophomore in high school and turning 16 my dad asked me what I wanted for my birthday. And, of course, I wanted a car. But I didn't get one. He got me a calculator. And I remember that because it was like $300, and it could add, subtract, multiply and divide. I remember thinking, "I wonder what kind of car I could get for this calculator." But it's really amusing to think 25 years ago that was this really cool thing that really went out. Think about where we've come in 25 years.

I also want to preface this by saying, I'm awed by being in the presence of great technologists. I'm not a technologist at all. I was a creative writing major at college. I just was luckily related to a programmer, and so that gave me my start in technology, because he wrote some really great code because he wanted to start a company. So we started the CP/M company in 1983. Usually I say that and people are like, what's CP/M. It looks a lot like DOS, which I still see on a daily basis. It's remarkable to me that you still actively see DOS. But, of course, you do.

Anyway, started a software company, ran it for 14 years, was one of the first mass developers, did products like Click Art, and did all sorts of products, sold the company, and then after 14 years of dissin' at Apple about why couldn't they get developer relations right, they said, well, if you're so smart, you come run it for a while. And I joined Apple on the same day as Gil Amelio, which is very interesting. And I left six weeks after Steve Jobs came back on the scene. And so I lived through a very interesting period at Apple, which afforded me many opportunities for mind expansion, career growth, and character building.

And I kind of went off on my own, mentoring some companies, working on some boards of some companies. And then I had -- one of the questions is, what has been a technology development that really changed your life, and I had this kind of life-changing moment that even to me as a non-technologist, it was almost embarrassing to admit it was a life-changing moment. But I got a cable modem at home, and it was one of those homes where all of a sudden the Internet went from being something you had to go get, that was out there, took many minutes, that was slow, and that was a problem, to something that was always on and instantly responsive. And I had one of those moments that said, oh, my God, we're at the beginning of a whole new world of things that will impact all of us as just plain human beings, because here is this amazing window to the world that I've never been able to experience before.

So that's partially why I decided to go back to working and become a venture capitalist, because I thought this is really great, but I don't want to run companies anymore, I'd rather just have other people run companies, and give them advice once and a while. And so venture capital seemed like a better job than being a CEO. And that's when I joined SoftBank.

I think the thing I would say about that is, that's the good news. I think the bad news, and I'm just delighted to hear what Steve is working on, is because it's gotten increasingly more difficult to use technology, and it's amazing to me that, you know, okay, I'm not a technologist, but I've sort of spent a long time in this industry, I remember being able to configure my own machines and do stuff, and I don't know if I'm just getting stupid or lazy, but good Lord, when something breaks on my system right now, try to figure out whether it's -- if your cable modem is not working, is it the head end, is it the Ethernet hub, is it the connection, is it the... -- I just don't know what's wrong with these things when they break down.

The process of backing up, the process of adding a new device, you know, even on the Mac platform it's not trivial, and certainly on the Windows platform these things are very complicated. I would say my other, I would say, both the great opportunity and the great pet peeve of mine is the integration of devices. And, you know, I use as my example, I think a lot of us in the technology field tend to live on the bleeding edge, we're always down at Fry's buying the latest things that we then have to integrate into our system, and it doesn't talk to any of our other things. And I always swear, you know, someday I'm going to have a major car accident because I'm going to be dialing my phone, my phone doesn't actually know what my address book is, and I'm going to have an accident because I'm trying to input into my GPS system an address using something -- you know, the interface that reminds me of the Addresso label maker from 1963. And at the same time, I know that address is resident on my desktop and in my calendar it says I'm going to that place right now, why can't my car just know that and tell me how to get there.

So, I think that, again, it's a huge issue that technology is so complicated for us to use. But, at the same time, it's a huge opportunity because right now we have all these disparate devices, and it isn't going to -- it is in our very foreseeable future that we're going to be able to integrate all of those devices and really get some power out of them.

MR. : Twenty-five years ago, I had dropped out of college rather than flunk out to play music for a living. And it actually, before we move on here, I'm thinking of the user interface, technologic user interface that works really well, and it's called the keyboard, and the midi standard that connects all kinds of devices simply and transparently. And I keep wishing for something like that for the rest of our technological lives. I haven't seen it.

But, Steve, you really did raise a great topic at the start, and since Heidi also wants the answer.

MR. CAPPS: There's not a UI answer. I mean, to ban from a computer screen, thinking is one of them, which is what you just said, preparing to boot, whatever it says, it should be instant on. There's just a lot of concepts that users don't really care about that -- like I like to say, to paraphrase the old saying, when you have a file system in your hand, everything looks like a folder. And everybody is pretending like, what James just talked about, having to expose to the user the difference between here and there meaning up in the clouds here. Because there was an article I read that, you know, 100-gigabyte drives are not going to be that rare very soon from now. And there's always going to be a disparity between what you can afford to put locally versus what you can afford out on the ether. And so that's one piece of technology you are going to have to expose to the user, but the last thing you want to do is hit the drive button to go over to your network drive, you know, on a Mac that's how they do it. I just think it's taking, you know, 1970s technology and carrying it forward to the new thing.

And then when you get into things like peer-to-peer, which I think, you know, Jim was talking about putting things on the server, peer-to-peer is something that's also equally compelling. Especially with photos where you do have the local drive, why not just make a private community, because I really don't want people lurking at my kid's photos, a private community where we can just automatically synchronize the machines. But we, of course, would never call it that. We would just say, there's a photo album, and it just automatically happens. It should just be all turnkey.

So, to try to wrap this up, people think of photos as photos, and email as email, and documents are documents. So, fine, documents can still exist in folders. Chances are, if you're Windows users, it should be in a My Documents, and you shouldn't be able to drop it some random place, or you shouldn't be losing them in some random place. And, there's just some easy concepts we could learn on how the existing world can take care of it for new users, existing users love a: and c:, fine, let them have it.

But then let's just get at those with the concept of... pick your data type, music, photos, documents, programs, and junk, which would be your junk drawer, which is what I would call the Explorer or the Finder. You go there when you really can't find it, you rumble around there and you'll find it. And, inherent in photos should be sharing, you shouldn't have to think of another mind-set. You say, yes, there's photos, these are the ones I want to give mom. Inherent in music, you know, even though I'm married to a copyright lawyer, I shouldn't say this, you know, you might want to think about sharing music, especially streaming. Why not just make it, whatever I'm listening to, my friends, if I'm IM'ing with them, they can hear what I'm listening to. It should just be all-automatic. We shouldn't expose all this stuff as "upload file," or "send file."

So, I think we've got the technology, we've just got to be a little bit ruthless with ourselves and just say, let's reduce the functionality for existing Windows or Mac users, and go for a nice simplified world. And, you know, WebSpeed is a great exemplar of that, it's something that's very simple, plug it in and go. Computers could use that. And, sure, it's like a Russian doll, if there's a DOS window that she mentioned that we always still see, except I guess in Whistler there won't be, I'm not sure. There's the DOS window you see, and then maybe they'll be a Windows window you see, but then eventually you get into a really nice kind of seamless online experience that has to talk to your computer. And that's where I think we should just incrementally go. And I think the world would be a lot happier with that.

I don't know if that answers your question, but that's the rough gist of where I've been concentrating.

MR. GILLMOR: How would your other devices relate to that world?

MR. CAPPS: Well, again, you should be able to say, in your contact list, you should be able to say, this goes on my phone, and if this happens, you shouldn't have to do anything besides say that. And, you might be able to say select all, it puts all these on my phone. If you only have like 10 friends like I do, but a lot of people they have a lot of contacts. They have their business contacts, all that stuff. You can get really fancy. A lot of people, and I'll throw in XML here somehow, you know, but a lot of people will want to say, well, I've got my contacts, my system's email will know that I'm on a business trip right now, so it will upload automatically to my business contacts to my phone. Fine, if you want to be that anal, that can happen. But I think the average user just wants their phone list to be the same everywhere. And they don't have that many friends to worry about.

(Laughter.)

MR. CAPPS: It's like what's the average number of buddies, it's like fewer than ten, right, real buddies in your buddy list.

MR. : That's people you're willing to have interrupt you?

MR. CAPPS: Yes, exactly.

MR. : That's an interesting definition of buddies.

MR. CAPPS: There you go.

(Laughter.)

MR. CAPPS: There's another example. Yes, some people you want to let interrupt you at all times, and other people you don't want, you know, kind of causally, and we don't really capture that very well with the interface.

I think to answer one of your questions -- and reinforce something that Tim said, communication I think is definitely -- I think we've got the paper-centric computer done. And Office is a great UI for that. If you really want to crank out a number of bullet lists with Roman numeral indents with XYZ style sheets, Office is going to do that best. And that makes very pretty paper. But when you're sitting there IM'ing and you don't really want to have a conversation more than five minutes, much less put it on a piece of paper, it's a whole different gestalt, and we really should design a computer that supports that now. And just, I like to say, let's take the existing desktop, treat it as Chernobyl, put it in concrete, and just let it live. It will be like DOS or CP/M, it's fine.

MR. : And Windows.

MR. CAPPS: It's fine. You run it in a box, Windows in a box is a great idea.

MR. : And what's your take on what to do with all of this?

MR. : I agree with everything Steve just said. We have fought some of these battles together in the past. I think, you know, machines need to get more connected, so as we get more persistent Internet connectivity, then a lot of these things become more possible. I would like to see computers not only get simpler, but get smarter about remembering me. I use a laptop, and I would love to get rid of this laptop. I shuffle it from home to work, from work to home, and it's really just because I stored a lot of knowledge about me on this laptop, and I wish I could just transfer that knowledge, so that no matter what computer I'm using, I could identify myself, and then all of my documents, preferences, my address book, et cetera, all show up. Wouldn't it be great if the address book that Steve wants to use for photo sharing is the same address book that goes to the cell phone, it's the same address book that works with your email program. If you take the Microsoft family of products today, there's probably five different address book codebases out there and file formats, et cetera.

I know I'm not supposed to say the Newton word, but one interesting concept way back in our Newton days was that the Newton operating system had this notion of suites, and application data was kind of published out into a database that was system-wide that other applications could talk to. So the address book in the Newton could be used by 17 different applications. And that simple concept hasn't really shown up to the desktop platforms yet. And I think there's a lot of technology work that can be done to make things talk together and to be more persistent and more connected.

MR. GILLMOR: I was at a meeting in San Francisco yesterday with some people who were heavily into what's being called peer-to-peer, the definition of which seems to be growing, by the way, and a lot of this was coming up with those people, and I'm wondering -- (inaudible) -- Tim, you're looking out at stuff that's coming along, so tell us some secrets that you've found out from those people in your life, what are they doing?

MR. BAJARIN: Actually, there's a lot of exciting things, and we'll get into that. But I want to just reflect on what these guys were saying first. I totally agree, especially with James, I do the same thing. I have a laptop, and I basically went to a single laptop. That has all my preferences, and you have all of this stuff, my junk, my stuff I guess they call it. But I have everything that I need on there. And, fundamentally, I'm always taking it back and forth. What I really want is technology where the user interface is incredibly easy and, more importantly, anticipates what I'm going to do, and do it in such a way that that data is always connected, and everything is always up to date, even though that word you were talking about synchronized, needs to go away. The fact is that technologically, that's exactly what we have to do.

But I have, unfortunately, some bad news, because it's only going to get worse before it gets better. When I go on the road, and this happens a great deal with a lot of people, I carry in my case my Palm Pilot, my cell phone, my two-way pager. Do any of you carry at least three devices? Some of you have maybe gotten it down to two, right? One of the publications I write for is Global Computing, and I asked my readers about three months ago to get back to me on what would their perfect mobile computing device be? And I didn't give them any parameters. What I thought they'd get back to me is on some kind of mobile or ultralight, some great ultralight, you know, that was really light instead of these seven-pound monsters they carry. Out of the 50 responses I got back, 35 of them came back complaining and giving me ideas for what they call their combination device, which basically married the cell phone, the pager, and the wireless Internet connection, and the PDA and the pager and the two-way pager into a single device, so you're only carrying one device with you.

And I actually think that that's going to be the next major hardware battlefield in the next three to five years. And you're going to see one to two key announcements in the next month that impact adding the cell phone voice capability to this, which is just going to set off a whole slew of things that are going to start looking at these devices as multipurpose devices. But as soon as you do that, instead of this being my calendar and scheduler as a single device, and I start adding in more features, which, in theory, is the best thing you could have for this from a platform standpoint, the product gets worse, not better, because now I have multiple devices, with multiple databases, and multiple things to try to interconnect.

So, the one thing I will say that is exciting, because I spend about 60 percent of my time on mobile issues with my clients, is that I do think we're going to see this new battlefield going forward on this particular space. It's going to be important. But at the same time, that means we've got a tremendous amount of work to be able to deal with the UI and the synchronization question to make it work.

One of the reasons WAP phones are not going to take off is because of the UI. Have you ever tried to program a WAP phone? Any of you? It is really tough. Even if you have Tegic or one of the simple input devices. So, ultimately, some of the work that's going to have to be done is, I need voice input. I need to have character text recognition capabilities. I have a whole slew of things that bring these things together to make it easier. But it's not just easier to input, it's easier to set up. As an analyst, I get sent all kinds of products, and it's amazing to me, every time I sit down to them, they have a different UI. They have a different loading system. And, Dan, you get the same thing, and Mike and those guys, look, we get all these marvelous products, and every one of them is a completely rethink about how you put them all together. And I think there's an incredible amount of work that's going to do at the UI level, at the synchronization level, at the design level, at the device level.

And what concerns me is, at this point, those three groups don't talk to each other. Whether they're in the company or not or individually, but what we try to do then is, we throw up committees of which we now take three to five years to figure out those problems. And that's got to stop, because your traditional consumer is just not going to jump in on this.

So, just think in reaction to -- in response to that, I actually think that these devices coming together as one is the next major platform in the next five years, three to five years, and they come in two form factors, this and the "Web pad" devices, things that are going on, including at Microsoft, on that slate, what do they call it, the tablet, that is actually much more important than any of you guys can imagine as a platform or a device going forward. They're not anywhere near where they should be, but I think that's going to be important going forward.

And then, five years out, the next thing that I'm watching come out of the labs is projects that are related to electronic paper. I'm absolutely convinced that over the next five years to the next 15 years, we're going to perfect that electronic paper concept. And when that gets perfected, that is going to have the next major revolution in the case of bringing digital stuff to a display, hand-held format that replicates paper in the way we know it now, but is completely modular, completely wirelessly enabled or data-driven, and it completely changes the whole concept of newspapers, books, and anything that today is in print. Does it mean that the print part is going to go away? No, I don't think so. But I think this is going to create a whole new level of electronic digital devices that will hopefully spur a whole other level of creativity.

MR. : I just noticed someone in the front here with a Palm or a pocket PC but with a stowaway keyboard, a folding keyboard, which I think is personally the coolest --

MR. : I do, too.

MR. : -- new device of the last year of anything I've seen. Of course, my favorite.

MS. ROIZEN: Then I'm going to be a curmudgeon and say, it's the coolest device we can come up with is something that relies on arcane, you know, 80-year-old methodology that take a child months to learn how to actually use effectively, that's pretty pathetic.

(Applause.)

MS. ROIZEN: Yes. No, I'm saying for the average user. I actually love that thing, too, I think it's really cool. But, I mean, my point is, and what Tim was saying is, the first thing we ought to remember is, none of us are the target market. We all just put up with this stuff, and its punishment -- how many of you have little kids? I have little kids, and my little kids -- first of all, my older one is learning to type, she's seven. I don't think my younger one is going to have to learn how to type. I think it's really an interesting point, because Via Voice already ain't bad. And by the time my five-year-old needs to learn how to really use the computer, that's probably going to go away.

My older one came up to me a couple of days ago and said, I need Internet access. I'm not on email. It's a social problem for me. She's seven. So the world that they're living in and their expectations are very different. I think the natural battle that we have in this industry is sort of the one platform, I will provide you the end-to-end solution, which when they work they work really well, and they start whole markets, versus the I am a way to solve the discrete issue, and I'm going to try to solve that.

Now, what's interesting is when you're a venture capitalist, you see everything as discrete issues, because generally three-person startups don't try to solve the whole problem. They try to pick off a piece. And I think what happens then is, you get into these funny markets where, and I'll take CRM for an example, I'll take the whole digital goods management space. You want to put a digital goods solution together today, it's like going out and buying a car, but you've got to buy your carburetor in one place, you've got to buy your tires in one place, which actually may be a good thing nowadays. You've got to buy everything in disparate places, put them together, and chances are it won't work the first five times you try it. Everybody has got different business models. It's really a mess, because the alternative to that is one person controlling the whole platform, and nobody wants that either.

So I think a little bit we're all going to live in this kind of quirky model where I think we could all agree, you know -- would anyone disagree with the statement, anything that does not need to be instantiated as a physical product should not be instantiated as a physical product, right? We all want to buy our movies, and our audio, and any form of entertainment, reading material, any intellectual property -- good, there's no reason to stick it on a floppy and ship it somewhere, right? But yet do we all buy it that way? No. Even the most sophisticated systems are still very cumbersome and difficult. And part of that is because we're living in the transition phase where nobody has been able to affect an end-to-end solution.

Part of it is, I think we're living in that. The other thing, I think is, we have a natural tendency to want to just make our products more and more functional and more and more complex, because it's what our customers ask us for. Customers don't call you up and say, I didn't buy your product because it looked too complicated. Customers call you up and say, I want sync to my this, or I want you to add an extra widget. I know this, when we used to be in the word processing business, we had this product called WriteNow, and WriteNow was this clearly elegant, really simple word processor. Well, customers didn't call up and say, please keep it simple. They called up and they said, I want style sheets, I want a spell checker, I want a thesaurus, I want auto-hyphenation. And before you know it, we've built this massive thing.

And so I think the problem with that is, then you actually decrease the number of new users you're able to bring into the fold. And I think, yes, we're going to be in a little bit of a challenging situation. On the one hand, I can look at what Tim just said, and I can say, you're right, I mean, you know, like we need Batman utility belts to walk around town and carry all this stuff we're supposed to carry around.

On the other hand, do I really want a device that I'm going to have to go to graduate school to figure out because it's got all this stuff integrated, and it's really complicated. And again, something about my job is, we don't have to present the future, we just have to recognize it when other people bring it in. And so you have this little thing on called the Squirt Brain, it hasn't been announced yet, so don't tell them I talked to you about it, but it's one of our investments. And I think this is a fun example of something that just only does one little thing. What it does is, it records my physical activity every day. So it just records how far I walk, how many steps did I take. It interpolates from the data it knows about me, how many calories did I burn. If I wear a heart rate monitor, this actually sets up a little personal wireless LAN for those of us that wear a Precor or one of those things, it keeps track of my heart rate at the same time.

So then I can go look on a Web site, and the only thing I have to do is put this in a cradle that's not even on a computer, just on a phone line. Calls up, downloads all my data to my Web site, puts all these cute little charts and graphs up. And, okay, you have to be kind of, again, kind of anal to want to use something like this. But I think the idea is, again, until we start saying -- I don't want this on my Palm, right. I don't want to put my Palm on my hip when I walk around. I think there are certain applications of technology for which it's entirely appropriate to have a completely different device. And I think there are going to be some really cool devices on the horizon that only do one or two things, but do them really well in a very consistent manner that requires no set up and no education to use them.

And I think it would be just really interesting to watch how these worlds collide, because you definitely have the -- I want Windows wearable, for lack of a better way to say it, versus the I'll be happy to buy just the little things I need for $20, and I don't want any of this other stuff.

And I think it's a really interesting time. I have no idea what the solution is going to be.

MR. GILLMOR: Let me take an audience survey on something. Heidi said that she doesn't think her youngest is going to need to learn how to type. How many of you read books? How many of you believe that books can be written by dictating into a machine? Okay, so you think that the books of the future are going to be written by people dictating into a machine?

MS. ROIZEN: Some people do. I didn't say all books. Some people do.

MR. GILLMOR: Okay.

(Inaudible.)

MR. : Let me point out though, the thing that we forget is that any of the technologies we bring out goes through an adoption cycle. My first actual project that I worked on was on the IBM PC for IBM, which was an amazing first project for a guy just out of college. And one of the first big things I did, I got to write what turned out to be a landmark document inside the company, was answering the question, what should we do with the PC Jr. when it came out on the market.

MR. : Shoot it.

MR. : But you can't go to IBM and say, shoot it. That doesn't work there. And so I had to go in, and I wrote what in essence was the document that explained what I called the consumer adoption cycle. This is like 20 years ago or 15 years ago. And that has not changed. The consumer adoption cycle basically starts at the top, and the technology fundamentally gets in the early adopter stage, then it gets flushed through the business cycle, then eventually it gets down to the consumer cycle. Well, what was interesting on the consumer question, when I did this report in '85 I think it was, I said that I did not believe that the PC would have a chance to get into the home for at least 10 years based on this cycle. And if you look back historically, PCs didn't actually start making any inroads into the traditional consumer markets until '94-'95, just about the ten year cycle. So when you talk about these devices, the kids that are in school today, and those that run around their house aren't going to be adapting or adopting the input that we have today.

If we had great voice recognition, it wouldn't be an issue. But the education system is going to push that tech space, or that kind of input. In the case of a disability, and I'm very familiar with what from some work we've done, if you're forced in that particular part of that input, you will mentally go in that direction. But in the case of the way the adoption cycle works, and this is why when you were saying about your five-year-old, I'm less optimistic that she's going to not be able to do that. My grandkids someday might have that happen. But the generation that's even now, they're gong to be highly influenced by the input that we have on the tables today, even our youngest ones. And what you have to really be thinking about is what that adoption curve is going to be that is somewhat cultural, and educationally driven as opposed to just technology driven.

Now, if you are going to get introduced to a brand new culture, a la let's say China, or somewhere in Africa, and you start them off with voice recognition, they may not ever use a keyboard.

MR. : I like that. There's a good segue here on something to be a little more future looking. One of the pieces that I've had that's yet unproven is that -- consistency in UI is bunk, basically, for kids -- that I've seen, since you're talking about kids, I've seen my kids use the mouse completely upside down, and she has no problem. Where as an adult today does have problems. And I actually think we designed the entire GUI from Mac on up or on down, depending on which end you want to look at, from we were trying to like bring computers to our parents, or our friends. And nobody today that I know of, I look at Apple, I look at Microsoft designing the computer for kids. And I actually think it's going to be a really interesting venue, going back to just taking your existing GUI and just freeze it, that's fine for all of us old farts, but for kids I think it's going to blur into entertainment. It's not going to be about cranking out book reports. That's work. It's going to be much more about sharing, much more about having fun. And it's going to be very much more like TV.

So, one of the things we're going to do a focus group on soon is just playing around with what happens if every single time you hit command "r" or reply or new in email you got a completely different program, just a complete different, you can only get toolbars this time, or you might get Pokemon characters the next time, you might -- you know, just think of it as stationery on steroids. And for the classic UI designer, that's the worst possible thing you could do. And, of course, Control-Z and Control-X and Control-V would still be your cut, copy, paste, but you might not have a toolbar. You might just have menus. And I really think kids can adapt to that.

Go look at from one video game to the next, and that's kind of the gist of what I'm saying. If the main reason you turn on that computer is to have fun, is to communicate, the last thing you want to do is have a battleship-grade toolbar. And I do think there's something there that somebody should pursue. Now, it may be just a bunch of -- I don't know if it's a good idea or not, I certainly think it's worth pursuing, but I think it's kind of the counter to the somewhat nerdy, all things on one device, and it kind of ties into what you ere saying, which is, yes, maybe you would just want -- you only do music and instant messaging, why can't you find, because of the Web, because of the net, find a UI that just does that for you and is fun. And there's no reason -- I think there's a lot shrink-wrap mentality still in the system today. That's why you add features, because you only had one slot on the shelf, and you had to put as many features as you could on that shelf.

MR. GILLMOR: Let's take audience questions. You had one?

QUESTION: (Inaudible.)

MR. : We agree, totally agree.

QUESTION: (Inaudible.)

MS. ROIZEN: But I think your point is --

QUESTION: (Inaudible.)

MR. : The problem with voice technology -- (inaudible) -- that it's noisy.

MR. BAJARIN:

MS. ROIZEN: But I think you have to be situationally aware. And I think you're right, there's always going to be multiple input methodologies. One of the reasons I like to type is I'm a 90-word-a-minute typist. My husband hates to type. He just got ViaVoice, although I didn't realize, I walked in and he was going, "Open the goddamned window." He was talking to it. He said, I'm trying to train this to open the window. I said, I don't think goddamn is part of -- (inaudible) -- and then he was saying, Rocco, come here, which is the name of our dog. I don't know, you put sentences in and then chat, I don't know what. But the poor dog was -- (inaudible) -- I'm here, I'm here.

I do think you've just got to be situationally appropriate. And different uses are going to gravitate to different places. I always think, I would get on there and think, I have access to all sorts of technologies, there are many I don't use. So, when I -- I barely ever carry a cell phone, because I agree that it's annoying, and I don't like being interrupted. And people learn to adapt to me because I get like 100 to 200 personal emails a day. I get no voice mail, because everybody knows I'm totally unresponsive on voice mail, don't bother. If you call my voice mail message actually has this five-minute thing, a preamble, intended to get you to not use the voice mail. And so people get trained away from that.

And so I do think you're going to find people using different ones, or as we were talking about before, a lot of teenagers using multiple ones simultaneously.

MR. : I think you've got to remember, you may not be old enough, but 10-15 years ago you would see people walk into phone booths to use a cellular phone. Now you see them with an earpiece jabbering to themselves, and it's just a sociological thing. I agree with you -- actually, a computer lab full of kids, I wouldn't care if they're all chattering at once. But, you can't imagine lying in bed with your mondo laptop talking to it while you're wife is trying to sleep. So you're right, it is situational, but I also think it's a sociological change, it will happen, just as we witnessed with cell phones. It's a matter of timing, though. Adoption cycle and cultural change, it's not short, it takes time.

MS. ROIZEN : So all of these changes from the last 25 -- I'm fascinated to see how technology changes society, things like how we work has changed, how we adapt. I mean, we've talked a lot about unlearning keyboards, it's exhausting as a person to adapt to that. So from your perspectives, can you talk about how you think that society can adapt to all the rapid changing technology, because I can pick up a new device tomorrow and in another year that's $1000 out of my pocket, and it's obsolete. And from a job standpoint, I could spend six months, a year, two years learning a new technical skill, think it's great, and in another six months somebody from college is going to know 12 times as much as me. So there's a very large societal impact on how that happens. So I'm curious to see how you think we should adapt, we will adapt. There's the question.

MR. : Technological obsolescence maybe one of the engines of growth in your industry, keep that in mind.

MR. : I'd throw a comment in, I think the rate of innovation that we saw with the hardware and the software of the personal computing platform is slowing, or it's shifting, out to the network, out to content and community. I know a lot of individuals that don't live in Silicon Valley that buy a personal computer for one reason, and one reason only, and that's to run AOL. And they get their AOL screen name, and they use email, and they use a buddy list, and they go to chat rooms, and they take advantage of all this community that enables them to interact with other people, but there's no more reason to force them to buy upgrades or new processors, or more memory.

They're happy with what they have, and they're focused on discovering some of the new things they can reach over the Internet, and reach out to other people through some of these community tools, instant messaging, I think, is a great example of something that just took off like wildfire and it really reaches an enormous amount of average people, and it's almost the democratization of technology, where they don't have to learn cryptic operating systems or interfaces, they can focus on something very, very simple, and they can actually spend less money and have a great experience. So I think that's going to change the rules of how all of us that work in the industry build new stuff.

MS. ROIZEN: I think also, though, there's a great fear of obsolescence. There's a great fear on the part of people that I'm going to buy something and I'm not going to get my value out of it. So I think part of it is that we're changing our business model, and so with cell phones you didn't have to pay $500 for the phone, you got the phone for free and you signed up for a monthly service. And the PC is moving that way, software is moving from shrinkwrap to ASP, and I think you're going to get people comfortable with models where the obsolescence is taken away from their control and put over in the hands of the business that can manage that. And I think that will go a long way towards solving this.

MR. : Let me raise a question there, though. How quickly will people get tired of being dinged for $20 or $30 a month by any number of people who want that $20-30 a month?

MS. ROIZEN: It depends on how important it is to the people who have that. Try taking away somebody's AOL for like a day, and see whether they'd be willing to pay money.

MR. : How many of those things are there, is my question, over time. And that's where everyone wants to be.

MR. : That's becomes the issue -- is there a pain threshold, or is there some compelling issue? I'm not a venture capitalist, but I've dealt with them and worked with them for a lot of years, and the first thing they want, besides looking at differentiation, is are what you giving me something truly compelling, so that the customer, or the individual is going to want that, and that's the critical factor there. And you're right, as a consumer, just putting my consumer hat on, I have probably seven or eight services that I have signed up for various reasons, because I'm in the industry, right. Not as a traditional consumer. As an individual consumer, I look at that and, you know, I need Internet access, that's about it. And it's kind of unfortunate, and yes, probably correct, is that the majority of the stuff I expect to come from the Internet is free. And that's not going to shift much.

MS. ROIZEN: I think there's one issue, everybody has come up with the $20 to $30 a month model, because there's lots of things we're used to buying in $20 to $30 increments. I mean that's one of the issues with the music industry, we all agree that the problem is it's eleven cuts, but I don't want to buy eleven cuts, I want to buy one or two. But, they have no model for that. There's not a really good micro-payments infrastructure end-to-end solution yet, so you can't charge people 50 cents for the cut. Just like I have a subscription to cable TV, and I have Direct TV, and they each cost me $30 or $40 a month. I only really want two channels off of each of them. Why can't I just pay $4 for the 4 channels I want? I will, I just can't right now. So I do think what's going to happen is customers are going to -- I think the system is going to evolve so we can be more precise about what we want delivered to us.

MR. : So real 'a la carte" consumers, over time?

MR. : It's just that it's going to take time, because the traditional businesses have got to figure out how to adapt to it. You talk to these entertainment industry leaders on this issue of doing one cut at a time and selling it, they go bonkers. I mean, their eyeballs like fall out, and they're having a heart attack, because they say, wait, we can't do that. And it's going to take a lot for them to figure out that if you don't do it you're not going to survive. But, they're not at that point yet.

MR. : That raises lots of other issues.

MR. : I just wanted to go to the one example of payments, we're talking about 25 years, 25 years ago if you wanted to put a second phone in your house you had to pay extra money for it, and most people didn't do it.

MR. : Not a second line, a second...

MR. : A second handset. You know, you paid per handset, it was like a buck or something per month. And people would say, I'll go talk in the kitchen, you know. So once it got to the point where it was a flat fee -- when you aggregate things, though, imagine the phone service of the future that did do this, and delivered photos to your device, and all that kind of stuff for one fee, maybe with some kickers if you really cranked them up people would pay for it, because there is a -- (inaudible) -- value, not just an incremental value.

MR. : Actually, there's another good example of that, we're talking about pain level. One of the clients I work with is a company called Surveyor. And they have created what they call the "baby cam." And fundamentally it's the ability, you can't see it on here, and I doubt I can get in here, but I can basically go to a URL site where the Web cam is, if my kid was in preschool, and actually check on the kid. Well, we did focus groups on that. You get the parent connection from their desktop at work, or from a Palm Pilot being able to check on the kid, periodically during the day, we figured that maybe if we charged $9.95 they would like that. It turned out their threshold was as high as $29.95, because it was a pain issue, it was something they really wanted to monitor, and that was something that was very high on the list. But, you throw some other thing in, like news and weather and stuff that we just figure is a traditional part that I can get anywhere, it's not going to happen.

MR. : Is there somebody over here?

MS. : Yes, in the context of the last 25 years I remember being told that there was going to be a paperless office, everything was going to be electronic. You were talking about, as far as how we were going to interact with the information going back and forth, and now the biggest application, or the biggest company running on the Internet is a company that sells paper books. People go out there and order up a paper book that's now delivered to their house, so they can open up a paper book and start reading it. And I think about some of the things you're talking about here, as far as the method of delivery of the information, and I'm thinking that it's not really so much the method as the information itself, and that content.

Heidi was mentioning, for example, Ford building the car, and taking the pieces together and assembling it. And what you're buying is Ford's brand name on assembled pieces from everybody else. You may want a Ford, or a Toyota, or a Hyundai, everybody is going to want something different. I guess my comment, or question to you all is, who are the Fords and the Hyundai's in this whole situation coming up, and also do you believe that the next wave of digital content is larger than the last wave of the Internet?

MR. : The last question I think, if you got the stuff that -- (inaudible) -- and I were talking about, and I made fun of XML, but some XML like thing where there is a free interchange of information, I definitely think that amount of content will put the HTML-based content to shame in terms of quantity, and in terms of interop. So I'll answer that one little piece of your question that I do believe that will happen. It's not like -- (inaudible) -- but information does want to be free.

So if we can get that working, then the need of having a brand, like Ford, like Microsoft, like Apple, where you trust that all the pieces will interop will be less. So you could go to Cobalt and buy your server, and go to Sun and buy your -- go to one company to get one part and one company to get another part, knowing that the parts do interconnect. But, I think humans are human, and we will always want to buy the "Built Ford Tough" kind of things. So you will actually reduce functionality to get one brand. That's why Apple still exists, right?

MR. : It's not just brand, it's ease of putting it together, right? How many of us want to sit down and go figure out what all these components are, even if they're there, and let us mix and match and figure out what that is. Ultimately I'd just as soon let somebody else do it. And I think when you get to the consumer level, that's a really big issue. They're not going to put up with a lot of this stuff.

MR. : We're already there. I mean, the Gateways and Dells and Compaqs today are doing exactly what Ford does, inside those computers they're sourcing from lots of different vendors, lots of different components. So it's a question of who can deliver you the best customer experience, who creates the brand that you trust, and a compelling price point, whether they price it per month, or whether it's something that you buy. And the free market will dictate kind of who evolves there, I think.

MS. ROIZEN: I think even using your example of Amazon, for example, and I'm still buying a physical good that's still being shipped to me, in part that's because there's a certain pleasure of having a physical book. There is still a discrepancy between consuming that in paper, versus consuming it in a digital form. And I'd say, still I tend to read most of my books in book form, in paper, dead tree form. On the other hand, I've given up most of my magazine subscriptions, I've given up most of my newspaper subscriptions, because of searchability and immediacy. So I get it online, because it's more relevant to me to have it online, right.

And I think it's really interesting when you look at the e-book market, I think it's really interesting that Jumpstart TV Guide bought Rocket Books Software, because when you think about it, it's really rather remarkable that TV Guide is the number one selling publication in the country, and TV Guide's information is something that I inherently something I want to index and search and reference, and that's much more easily done in electronic form, and yet where I want to do that is not in front my computer. So it really makes a lot of sense. But, it's just getting there.

I just think that the other problem we have and it's analogous to that situation is where we want to consume this stuff, particularly digital goods, is not where we currently can get it. And I always say, how many of you currently have stereos hooked up to the Internet? And it's recognizing we're not the target market, and even all of us don't have stereos hooked up to the Internet. So still I can't exactly get what I want using this, but I think it will get to that.

MR. : That's interesting that GemStar is the company that patented the display of television schedules in rows and columns on a screen, no kidding. The intellectual property issue is something that if we have time for I'd like to get back to. But, let's go to the audience.

MR. : How many of you think that the current economic model underpinning the Internet is sustainable in the long term when advertising doesn't work on the net? I mean, how is the Internet going to survive if a lot of companies are finding it really difficult to make money from it?

MR. : I probably should answer that. For our business we have no idea. Well, that's not true, we have some ideas, we don't know. We don't know if it's survival, we don't know if our business model works out more than a few more years. We certainly hope so, but the evidence is not very clear on this. Furthermore, all intellectual property in my opinion is jeopardized in the current form, with things like the Napster, Gnutella, FreeNet model. And it's jeopardized at one level, I suspect that we all need to find new business models, those of us who are in intellectual property. Talk about a changeover, that's going to happen.

MS. ROIZEN: I think venture capitalists get asked that question a lot, particularly because of that problem. I think that the very short answer is what happened is we got in this weird situation where the capital markets rewarded shareholders in advance of rewarding customers, in advance of satisfying customers. And so you got a situation where you didn't have to have a proven a business model in order to make a ka-jillion dollars. And all the way up the public market supported that for various reasons -- (inaudible) -- that was a flawed model to begin with. And people made a lot of money on that model.

But ultimately, stock markets tend to go back to PE ratios, they tend to go back to company's share prices being worth the discounted present value of the future cash flow stream. And therefore, companies have to have models that make money. And everybody got away from that, and they started looking at things like -- (inaudible) -- and things that inherently didn't make money, you had to attach something else to that. So we're in this incredible transition phase right now where old models, and I think the media model, what happens when advertising isn't inextricably tied to the content you're consuming?

All that free content you get on television is there because Ford pays to have you watch that 60-second spot, well, when you can edit it out what happens to the business models of Ford and of ABC for making that happen? I think we're in this incredible transition. But, ultimately corporations still need to make money. Ultimately, when we look at businesses now we much more look at not what is it buzz word compliant, or eyeball business, we say, let's take the trends, let look at how much is it going to cost us to make this thing, who's going to want to buy it. Is it a painkiller, or is it a vitamin? It's another one of those things, does it solve a real need or is it nice to have? It's easier if it solves a real need to get people to put the dollars down for it. And ultimately in the end can we make money off of this thing. I think there the Internet is still vastly untapped.

I still think there are many Internet businesses that failed because they weren't businesses to begin with, or no matter how fast they grew they couldn't grow into the valuation that was created for them. And then the third reason is some of these new paradigms consume money for a couple of years before they get to a point where they can sustain and grow. And that's not just the Internet, I mean, there's a lot of businesses where it took many years of losing money before you could make money on those businesses. And the market got impatient. At first it was overly aggressive, and then it got impatient, and a lot of those things went away.

I'm sure that's been a very squirrelly answer to a very straight question.

MR. : A couple of things to throw in there, on my second Internet business -- I think, to your question, the Internet has a lot of business models. It's really just a channel, so there's a whole range of models out there. But, I think we've seen, we've lived through a real gold rush where we saw a lot of businesses that were essentially what I think of as features, not really businesses crop up. And a lot of people profited from that, in some cases, but we've seen the public market economics change, that really changed the rules requiring people to run a business. I think there's a lot of Internet companies waking up realizing that they are a for-profit business, and they never really had that phrase before. And so you will see some changes.

I think if you take an old economy example, you know, not everybody can be a Wal-Mart. So I think you're going to see a lot of businesses that are very sustainable and profitable, if they're run as a cottage industry. There's a lot of ideas out there that could be very efficiently run with six people, and be profitable, but never go public, and should never take venture capital. They wouldn't necessarily fit the venture ideal model. But, some will and some won't so I think you'll see some consolidations of big players, some people that figure it out, and some that don't. And you'll see a lot of successful small businesses that stay small.

MR. : One last point, in Slate I just happened to cross a couple of articles by Nathan Myhrvold that he wrote maybe three years ago, that were exactly about this. And they were pretty right on the mark. And going back to the TV model, there was a great thing in the New York Times Sunday Magazine maybe a month ago about, is TV dead? And it talks about these issues exactly, about how it's really going to change, and we are in a cusp of a big change where the 30-second spot is going to go away, and you're going to be watching a commercial, basically. When you think you're watching Friends, you're going to be watching a commercial.

MR. : Heidi, you'll be happy to know Donald Dormond agrees with you about single devices doing simple things, he also wants them very cheap. By the same token, when I'm driving with my cell phone set up for headset so that I don't have to mess with it, and the address is in my Palm, sure I want my Palm to tell my cell phone what the phone number is. However, if I put the phone number in my Palm, or my notes in my Palm, and I'm on my phone, how do I read my Palm.

So to the question, companies prefer proprietary technologies so that they have an edge in the marketplace. How do we mesh that with the need to have transparent data formats and communication protocols? In terms of the telephone network what blew that wide open was something called the Carter phone decision, which allowed you to attach whatever instrument you wanted to the phone network, as long as it didn't hurt it. Do we need that kind of legal intervention to force the market?

MS. ROIZEN: Well, I think this is the challenge that a lot of us in the industry face, is that for many things, open platforms need standards, and standards are created either by momentum, just the market gravitates towards something because it happens to solve a problem better than anyone else, or they're dictated. There's sort of no good way to do it in a rapidly evolving market, where unfortunately the problem is -- (inaudible) -- so fast that you can even -- you know, a standard that's going to work today isn't even going to work for broadband. There's just so many moving parts.

So I think the trick here is, and I think it goes back to the question you want to bring up, I think IT issues are going to become pretty important. I think some people are being very proactive about how they articulate something, so that when the standard does evolve they happen to have the patent for it. And they can go back and tax the economy for that, and derive profit from that. I think it's going to be very tricky. I just think, to me, when we look at companies to invest in, and they have secret sauce, you know, they have some patent or something, that's less attractive than when we look at a team and we say, these people know how to execute, because a lot of times it's just consistent execution that wins the game, and not the intellectual secret sauce.

MR. : That and a lot of the guys who have had management experience, like in your case, you've already done this when you went to get money for Ofoto, I have to believe that's one of the first things they look for was your experience in it, as opposed to intellectual property, specifically.

MR. : That and the dartboard.

MR. : There's a great quote from -- (inaudible) -- and he says, it's not the idea it's the execution. He says, you want a billion dollar idea, make an OS that's small, stable, and free. And it's true, it's not really the IT, not really the greatest idea, it's just how you can get it out.

MR. : I'm curious, rather than focusing on things that will become prevalent and central to everything we do, isn't it the likelihood that more and more, as more things are invented, that if we have six UIs today that people use, and we invent five new ones, that means in three years we'll be supporting 10, one may have died, but boy, OS2 is still out there, a lot of things are still out there.

MR. : Who's still supporting OS/2?

MR. : There are people still supporting it.

MR. : The Amiga.

MR. : Not even IBM.

MR. : One of my questions is, when you go to a person's home very often you'll see different ways that people have organized in their home, I'm always amazed at this. Some people have, they can find everything, because it's all out, other people can find everything because it's all organized. Why wouldn't it be the same in people's use of computers, that they would have hundreds of different ways of doing something rather than a "standard" way of doing it with the user interface?

MS. ROIZEN: I think this goes back to one of the core dilemmas of anyone who runs a -- (inaudible) -- sometimes a critical mass of things will develop with some momentum, and you have to create that critical mass. And I think the reason is because maybe you want to do something one way, but the rest of the market decided that they could only build -- (inaudible) -- if they all gravitated around somebody else's way of doing it. So I do think, and I think going back to your question, one of the most ostensible business models you see is something that has a huge network effect applied to it, that the entity becomes more valuable the more people subscribe to that entity's way of doing things, and then it's very hard to break them away to something else.

So I think the reality of the market is, maybe people want to do things ten different ways, but an industry can't crop up to support ten different ways to do it. And so someone will either bribe them with money, or create the better mousetrap that will make people go there, or some phenomenon will occur. I don't know if you've read -- there's this great book called the tipping point, it talks about how little things --

MR. : The author is Malcolm Gladwell.

MS. ROIZEN: And it's very interesting that some things actually become standards for almost no discernable reason, other than a few people who chose it along the path. I think that's the problem with having an infinite number -- (inaudible) --

MR. : In a way what you just asked was a pretty good sense of the personal computer, which on some level is this all things to all people device, which is one reason I think it's going to last a long, long time, despite all this other stuff that's coming up.

MS. ROIZEN: But, having been -- you know, we made a word processor for Mac, it's very well regarded, people absolutely love it, it has a cult following. I still have people, Roger Maccabe (sp) needed a copy of it three days ago, he stores all his music files in WriteNow format, and he can't clean his hard drive. He's like, where do I get a copy of WriteNow. So one person I know still uses it. I emailed him a copy, and he's off and running with it.

But, why did WriteNow ultimately fail? Why did it fail? It failed for two reasons. The number one reason it failed is because people wanted to buy a whole suite of applications that were all the same. And the second reason it failed is because they wanted to share files with all the other people who were using word processing. And so once the market went more than 60 percent Microsoft Word, everybody needed to share files in Word. And so what happened is even though the product itself was in many people's minds superior in performance for what I needed, my need to integrate with everybody else made it not as useful to me as using the standard.

MR. : I'd be really interested in hearing each of you comment on hearing the development over the last 25 years of handheld computing, and pen-based computing. Some of you have some deep insights into that area. I'm sure all of you have some things to say.

MR. : I think historically the work that these guys did on Newton is going to go down as very, very important, and pioneering. I had the privilege of being involved a little with that, as well. They showed it to a few of us analysts really early on. And while none of us bought the hand recognition concept, we did buy the concept of the PDA. And if you look at the Palm Pilot, and a lot of the things that are out today, it's just a reincarnation of that, done a little bit differently, obviously with a different emphasis.

And I remember, because Jeff Hawkins and I, when he was working on this in the early days, we had a lot of discussions about the thinking behind the Palm, as opposed to -- and by the way, the lessons he learned from when he was at Grid, followed by the lessons he learned, and we all learned by what happened with Newton, but I do think historically, even though it was a tough time for them, that will go down as a pioneering effort.

Going forward, this question of creating one device that's multiple devices, for multiple uses, that's very interesting. I totally agree with Heidi, you are going to get various devices that are application specific, I think you're going to see a lot of creative things there. But, at the same time, we're going to end up with a device that's going to be much more practical in the sense of expansion, and multiple uses. And that's why I actually argue, I just did a column recently on this subject for one of my magazines I write for, that I actually think that the Springboard slot is the equivalent of the PCI slot for the PC.

And that you have to really think hard about -- what Jeff did in the early days with the original Palm platform was actually right. He shouldn't have made it expandable at that time, because he had to create a new concept and get people to buy in, within its simplicity, which is that it does four or five things really good. But, when he went and created the Handspring product, he was right on the money, because he knew that the market -- he had already helped get that market to one point. The Handspring piece, with the Springboard slot, takes it to the next level. And I think what you're going to see is a tremendous force, that's going to force the hands of Palm and Sony to deal with this "PCI slot" question, to give these things more functionality. And if that does that, these devices do become, I think, more interesting and more powerful.

But, I do believe that we're going to see them come together with communications, paging, wireless connectivity, et cetera. And while I don't necessarily want to carry it on my belt, as you pointed out, if I can carry less things on my body which I'm doing now, and take it down to one, that's going to be a big plus. So I do think that the handheld has a real good future, but I think it's going to go through tremendous iteration over the next five years.

MR. : One thing to tie into that is -- I ruefully comment when we talk about Newton is-- we went out and talked to developers and retailers of Newton, and the last thing they wanted to do was carry hardware-based software, something that you plugged in a card, because the cost of goods was so high, what happens when you get stuck with 500 copies of Fodor's Chicago guide, when the cost of goods is $10 for each one. But, the Internet has now enabled that. And I think it will be interesting to see if the Handspring takes off on that, if you can get rid of that shelf space issue, I keep bringing up the shelf space subject, because I really think it's impacted the personal computing industry a lot more than we really a lot more than we really realize. That if people can just go to a Handspring site and say, I want that device and get it, and you can kind of get that efficiency of producing the goods, then you might get that plug in the GPS thing, I want to plug in this.

But, it's just funny to watch this, because our studies with Newton, and JJ would know this more than I would, that people did not want to be swapping cards, and retailers didn't want to buy cards, and developers didn't want to have that barrier to have such a high cost of goods. But, all three of those things, I think, now because of the Internet now may not be true.

MR. : Well, it's a data-driven question, and getting information, but like he said, one of the problems I would have is I can't make this a GPS system without adding either the GPS chip into here, or a modular point. And if you talk to the guys that are involved with the design of this, one of the actual problems is do you do it from an external capacity through the spring board slot or something, or do I put a DSP chip here, a video chip, do I put a GPS chip, or whatever. And you start putting the multiple chips on here, it doesn't just impact cost, it impacts size, power draw, et cetera. But, eventually what I want to do is in the GPS model, I want to have the GPS function, and then I want to download the maps. Don't want to have to carry multiple cards with maps simultaneously.

MR. : How quickly will that also impact stability, because I know people with Palms who have loaded all sorts of stuff on them, and their Palm handhelds crash, and they have to reload their software, and it sounds really familiar somehow. It's something I don't want to do.

MS. ROIZEN: I'm not carrying anything. I don't want to carry something, and I mean, it's really funny, because I know we all lug our laptops around all the time. And where do you use your laptop? You use it on the plane, you use it in the hotel, you use it in the office, and you use it at home. Well, if you had -- machines, if they were just ubiquitous and all your stuff was on the net, you wouldn't need to carry that stuff around anywhere. I always think it's interesting, my husband travels overseas he does a lot of medical relief work, he doesn't carry a laptop, because the connectivity issues are overwhelming. But, there are Internet cafes everywhere he goes. I mean, he's in the deepest darkest jungle, and I'm emailing you from an Internet caf é on AOL. It's just amazing, and it's kind of funny that in some places because they didn't have -- they don't have laptops everywhere they have gone with this other stuff.

Again, I think we have to think outside. And I'm sure it's what you do all day long. I get really accustomed to carry my stuff, so let me carry more stuff, and we forget to think about rolling it back one level and saying, I've got this to carry, I don't want to carry that.

MR. : Who likes the zero pound computer. Actually, my favorite one is the one on the Starship Enterprise.

MR. : I think one thing to add to your question, just looking back to the Newton experience, we did a great job of building a handheld platform, but we did it at a time when the market really wanted a handheld organizer. That's really all the market was ready for. They wanted Sharp Wizard one step further. And I remember in 1993 we touted that as the year of wireless, and Newton was the first with all of these wireless add on devices, and a PCMCIA slot, when no one knew what that acronym stood for, and we were a great R & D lab for the industry at the time.

But, I think wireless is going to be a key driver. And I think a lot of the stuff that we talk about looking forward is really going from a handheld organizer, which I think the Palm platform did a fantastic job of, to going to this IP enabled or network connected device that blurs between the organizer and a computer. And it's that blur that there's a lot of customer education that needs to happen, there's a lot of solutions that have to be built around that to give customers a reason to want that. Getting maps is one good reason, that's a common thing. The folks at Yahoo find that the Yahoo map servers get pinged everyday around lunchtime, because everyone is printing the map to go to the lunch meeting. So look at the common things, problems customer need to solve, and that will drive a lot of that. But, I think wireless will be a key enabler.

MR. : And again, communication. I think the voice activation to this, and I know you were talking about the problem with that, but you'll more than likely have a microphone to that, as well, as one of the models where you're still being able to take notes simultaneously.

MR. : One thing I think just to add to the wireless, stuff like Bluetooth I think is going to be really cool. One of the things that we were smart enough to kick overboard in the early Newton thinking was, when you walked into a room all your Newtons would say, yo, yo, how's it going, and that way you could have scribble notes between each other, and the machines just did it automatically.

MR. : Like share viruses.

MR. : And we went out and we tried to talk to people about collaboration, all this stuff, and luckily the marketing guys came back and said, nobody gets it. Nobody got it. Now, the time now, because I think the time is right now, that with Bluetooth when you walk in, you know, what we're talking about is taking a 10-gigabyte PowerPoint file, turning it into a bunch of dead trees, and handing it out to people, and then talking over the slides. When you think about that it's kind of bizarre. Whereas, if you could use Bluetooth and just send it to everybody's machine, whatever it is, and maybe all you're doing is sending it to somebody's server somewhere so they can refer to it later, is a much nicer model. And I think there's a lot of cool applications here that we just don't understand yet.

MR. : And I think it's a mistake to think that these kind of devices are going to replace your "desktop computer" whatever that may be, because the constraints we're talking about, size, portability, weight, wireless connectivity and the bandwidth associated with that, to build a great solution within those constraints, you can't do everything. And a lot of products have failed trying to do everything. So this idea of having a device that's an adjunct to a bigger, more powerful device that's got the big LCD and the sub-woofer, I think makes a lot more sense than to try and put everything into one box.

MR. : I was wondering, maybe it's an East Coast thing, and we don't talk about it out here. The Oxygen project in MIT. The thinking at least seems to answer most of these questions brought up here, maybe not answer, but address them. Do you guys have comments?

MR. : The Oxygen project I heard of was actually at the advertising agency -- (inaudible) -- maybe it was in conjunction.

MR. : No, it's sort of a free form project, basically I believe it has a couple of different departments involved, but --

MS. ROIZEN: I don't know that I can easily give you a sound bite, I mean, you're pretty much describing it, it's that everything talks to everything else. It's the most promiscuous thing that's every been discussed since the beginning of computing.

MR. : And the concept also is that your hardware is more or less a black box and everything else is software.

MS. ROIZEN: Right. You can read Michael Dertouzous' book, which is out there, I can't remember the name of it, and he's got another book coming out in the next few months, and actually, this next book is pretty much about Oxygen. You can go to the MIT Web site and get information, also.

MR. : There are projects like this going on in a lot of places. I was at the Sony computer science lab in Tokyo in July, and they were working on a pervasive system that sounded fairly similar to that. I haven't seen the MIT thing, though.

MR. : I think the key will be focus and narrowing of scope. A good example, that's somewhat related to this, is cellular phones. If you look at how the U.S. screwed up with digital -- competing digital phone standards, and then you look at Europe where they had this one thing called GSM, and you got one phone, and one phone number, and everywhere you went your phone worked, and your phone rang, they solved a very, very specific problem, but they did it around a standard, and they beat us to the punch there. So maybe the same thing will happen with different points of technology in our personal computing.

MR. : So far we have talked about technology, I'm just wondering what's the next generation of power sources, because, for example, in may case, when I travel I don't even bother to bring my laptop, because it takes like -- not only that it takes five minutes to boot up, but after I boot up a quarter of my juice is gone. So what do you think about a next generation of that power sources?

MS. ROIZEN: The first thing is, I wish there was a plug in the airplane. Some of them have it. There are some technologies, again, some leading edge technologies that are reducing the power required to use things. For example, wide band will send a signal with a much smaller power supply, and with greater efficacy, so that you end up needing less power. We are seeing some -- I think this is one of the things where as venture capitalists you don't see a lot of people come in and go, I've invented a new battery. We imagine that most of that is happening -- (inaudible) -- and they're going to come back and have something.

MR. : But, I do think the -- (inaudible) -- the Strong-arm is a great architecture -- (inaudible) -- was what we used to call it, Intel has a chip that runs super, super hot, and Transmeta did some great work there. There was some paper floating around last week on some new work, I can't remember, where some guy, I think it was the University of Rochester, has this way of shutting down the chip partially. So I think that's where the breakthrough is going to be, not that there's going to be nuclear fission in a bottle.

MR. : Look where the big power consumption is in the devices, and then change those things.

MR. : Conservation is our next big power source, if we're smart.

MR. : The advantage of what you have there with a keyboard is it fits in coach on the thing in front of you. When they lay back, they won't kill you.

MR. : Not on United.

MS. : We have time for just one or two more questions. So, if anyone has --

QUESTION: I heard Jaron Lanier talk a few months ago and he said that he doesn't think artificial intelligence will ever really reach the level of virtual reality, not because we don't know how to make the hardware, but because the software will never be good enough. And I was wondering if you could comment on whether hardware or software is more advanced at the moment, and if that's going to change at all?

MR. : There's two parts of software for that, the raw mimicking the human brain, so when you type in, I want to get to CDs from here, how do I do it, and it doesn't take you to Amazon.com in the browser. But then there's also the application of what users really think, and how users think. And, it may be -- we were talking earlier about the socialization around telephones, it may be a learned skill. I know Gates certainly feels this. We built prototypes a couple of years ago, where you could actually type in things like that. And we did user tests on them, and users just never even began to think a computer could do that. You could actually type in, what's the weather in Miami, and users just didn't get it. And when you tried to get the thing pertinent to them, like give them the task of how many emails did you send yesterday, they'd go to yesterday and they'd count them, when they could have just typed it in.

And it's an interesting thing. We talk about narrowing and providing specific solutions. Is Amazon showing artificial intelligence when they suggest books? You know, from a Turing test, yes, they are. But it does solve a good thing, you buy Harry Potter 1, 2 and 3, well maybe you want Harry Potter 4. I mean, I think we all could write that AI program. But it is artificial intelligence, right? So I do think there will be applications like Tell Me I think is exploring some interesting stuff there. But it's not really AI, but you don't really care.

MR. : (Inaudible.)

MR. : So, I do think it's going to happen. I don't think -- in the mid-'80s when all those startups that were -- what were they knowledge-based or something. You know, it kind of comes in waves. The expert systems, all this stuff. And they were touting themselves as the spreadsheet for ideas, and stuff like that. So, it's going to happen, we just probably won't know it when it does.

MS. ROIZEN: I think the other thing is, it's really hard to design software in anticipation of hardware breakthroughs. I mean, first of all, it doesn't make economic sense to do that, because what if it's late. And it's really hard to test. What I have sort of felt watching this industry a long time is, it sort of leapfrogs. And every time like something is coming out -- like, I remember, and it's interesting how quickly, as soon as you come up with a new hardware breakthrough, it's just completely just saturated. And I remember thinking, what are we going to use 600 megs of storage for when the CD came out. And then there was multimedia, and then you didn't have enough. What are we going to use all that broadband, why would anyone need better than 28.8. And it's kind of interesting because what happens is, people have a tendency to invest in things that can do all of that.

So, I do think part of it is an iterative process, and we are getting to the point -- I will say today there are interesting -- you know, my husband came home yesterday and said, I need a G4. I said, well, why do you need a G4? And he said, because ViaVoice is faster. And he's willing to go out and spend thousands of dollars on a new machine because it will speed up Via Voice because he has to kind of talk unnaturally slowly. If we just got a little faster processor, maybe a lot faster, all of a sudden that solves that problem. And as soon as you get a faster processor, we'll put more features in to make something happen. So, I think it's just a matter of waiting it out.

MR. : Let's take one more, and then we'll wrap up.

QUESTION: My intelligent agents are my friends. If there's something I need to read about or find out about, I'll get a link in an email. Most of the browsing I do, people send to me. So what Amazon is doing is exactly right. I don't need a computer to do that for me, I have a bunch of friends who do that for me.

MR. : This is a question I ask people, how many of your -- would you let your friends be your Mastercard. And actually most of my purchases, in fact 90 percent of my purchases I would love to share with my friends, because of exactly what you say. And there's no AI there. It's just connecting the parts together, and getting it to where I feel that it's not tied to my Passport and anybody in the world can see it. And I think there's just a lot of low-key things.

Let's go back to the TellMe example. Caller ID would do 90 percent of what TellMe does. It should just know what my phone number is and give me my stuff, and recognize next, or skip, or something. And you don't really need the AI voice stuff. Yes, the AI voice stuff is nice, but I think there are a lot of low-tech solutions for this that will pave the way for the higher tech ones to happen.

MS. ROIZEN: (Inaudible.)

MR. : What's critical to that is, the design goal has to go -- be coming from that viewpoint as opposed to the way we design things today, which is, we can do it.

Read More: