 Good afternoon, I'm Lisa Smith and I'm from the Information Resources Group in Education Services and welcome to today's marketing forum. I'd like to present Gerald McCoskey and he's here from New Science Associates and what he's going to talk about today is what they call continuous information environments which includes a whole lot of new and changing and emerging technologies and he's going to talk about some of the information that they provide and some of those technologies specifically. Check. Great. Thank you. I'd like to thank all of you for making time in your schedules to be here. It's fun to talk about these things and I hope over the next hour or so to help clear up for you what on earth we mean by continuous information environments. I managed to forget on my way out of my office a box about this thick with samples of our research so I apologize if you'll leave your name on a list or leave a business card or whatever we'll make sure we get you either a copy of the text of the speech or some samples of research or whatever it is you'd like to have. I'm going to go moderately fast. If I go too fast, slow me down. Feel free to ask questions. Go off on a tangent. I have enough material here to go two hours but I'm going to skip some and just put a couple up very quickly. Very rough idea of the agenda. Just what is new science? You do subscribe to several of our research services right now. Go into some problems in the business environment and the thing we call continuous information environments. Finally some conclusions at the end. New science is a six-year-old unfriendly spin-off from the Gartner Group. That's the easiest way to peg us sort of in the marketplace and where the founders come from and so forth. We now have about 350 client organizations, seven different research services. I started one of those seven, the third one chronologically speaking in January of 1989. That one's called Intelligent Document Management. And just this May, I launched the Continuous Information Environment Service. If you'd like to know about the other research services, let's just do that offline afterward. The relationship that a company has with us is similar to the Gartner-IDC relationship. There would be written research, which we write all the time. We happen to have three kinds of it. We have research highlights. I have only a couple examples here. I'm sorry. We have what we call research highlights that are usually three articles, four pages each, twice a month. Lots of different kinds of topics. We have what we call strategic directives, which are more like white papers, three or four times a year per service. And then we have what we call ZAP analysis. And I don't have an example of that along, but you can imagine that some industry event occurs that we think is pretty important. So we drop everything right up in analysis and then we do a fax broadcast with it so that all of our clients get to see Apple IBM Alliance analysis from their favorite analyst house the next day, that kind of stuff. We also have unlimited untimed inquiry service, so people can call us up and ask all kinds of different questions. And right now we're moving into electronic delivery. I'm very happy to say that Lotus Notes is our most advanced electronic product. So we actually have some clients like Texaco, which replicate our database of research onto their site. And then they broadcast this to whoever wants it within their organization. So that's a part of our research service as well. Very happy to be having that. So let me talk about continuous information environment specifically. I'll spend no time on our other services at all. Part of the, there's two sources for this service. One of them is innovation in the market, which is the typical source of new service ideas. There's lots of seats up front. You're welcome to come forward. And I won't be bothered. There's innovation. And then there's a second seed for this particular service. And the seed was really frustration. There's a lot of problems that people face when they're trying to deal with our current information environment. And I'll call it the information environment rather than the computing environment because this service is about voice and data. It's about both the marriage of the two elements, the marriage of the two types of devices and the things we do with them. The kinds of problems are I get up from my desk and I can't take my personal information manager with me. I'm sort of isolated or cut off. Or if I buy a wizard or something like that, it generally has an incompatible set of things compared to the calendar or Rolodex that I have on my desktop. Or if you have a customer service operation, maybe you have people that are dropping off your inbound 800 queue. So you need some kind of a smart call distributor that begins telling people, well, we expect you're going to have to wait for four minutes rather than, you know, if you hold on interminably, we'll get to you eventually. Or maybe you have this problem where salespeople are spending more time doing their paperwork than actually doing their work of pitching face to face and being in front of people. Or maybe nobody's ever where you can get a hold of them and you can't schedule a meeting, you can't do anything like that. Or perhaps now that we finally put people on email, you start getting this unbelievable volume of messages. And it's very hard to cope with a number of messages and try to figure out which are important, which are not, which you can read later on. So some of the solutions people do, they buy a sharp wizard, they install a smart ACD, they take laptops or sell you the phones and put those into the hands of their sales forces. They install voicemail, so on and so forth. These are all different kinds of solutions. And these solutions to the problems are driven by real needs. I mean, these are real problems that people have when they buy a voicemail system or they decide to put laptops into the hands of the sales force. But very often, these are point solutions. They're sort of band-aids on a particular problem. And typically, they're a static solution, but something that really ought to be more flexible, ought to be architected. And ought to be moving towards some vision of where things are headed. But frankly, most people when they build these kinds of solutions are very focused on trying to solve the problem, get it working, do it at a price that's worthwhile, and move on to the next particular problem. Let me take two minutes just to build a business context for looking at these technologies. I'm doing a business context because more and more there's a correlation between technology strategy and business strategy. I'm borrowing from the Jim Beam commercials a little bit here, the print ads, this is my 45 second business strategic management summary slide. And it really moves from stick to your knitting with information as a cost way back in the 50s through a wave of diversification and globalization and M&A through the go-go 70s and 80s into some sense of recovery and trying to build ideas like strategic alignment or business re-engineering, the my camera sort of stuff, business redesign, very big hot word right now. And really moving back toward this idea of sticking to your knitting and moving companies away from being conglomerates and back into their own original lines of business. Yet, now there's a slight difference instead of information being a cost where, gosh, I guess we've got a cut checks and process orders so we'll build the general ledger system. Now it's, our information is our key asset. If you ask John Reed at Citicorp, what Citibank is, he'll say it's an information company, not necessarily a bank. All of these days I imagine these conversations take other kinds of twists. So there's this real idea, in fact, what happened through some of the 60s and 70s was that there were some strategic systems that came out. There were Saber and Cosmos and American hospital supplies, terminals in the hospitals, and all these really big strategic systems. So the CEOs turned to their newly minted CIOs and said, okay, you build us one of those. And it was sort of like bang, bang control theory. Information systems went from being a cost to being the complete focus of the next strategy of the business which was unrealistic. And what's happening now is this idea of strategic alignment where you get more coherence and more collaboration in the business strategy and technology strategy. And then the business tactics and technology tactics, which means, and this is one of the reasons why I'm saying this, I'm gonna present a whole lot of ideas mostly technologically oriented. But this idea of strategic alignment means they should only be used where they're appropriate to a particular business. They shouldn't be used just for technology's sake. So grant me the leave to talk about a lot of different things and present a pretty broad scope and sort of a vision of where we think things are going. But it's real important to go back to the business and understand what they're trying to do. So what are they trying to do? These are some of the things that people are really worried about today, the current business context. Increasing globalization, global competition, barriers falling everywhere, frontiers falling, decentralization, de-tearing, de-staffing, whatever euphemism you care to use for rifts. An emphasis on group work more and more, not just individual work. Information, again, regarded as an asset. And information used for rifle shot marketing, much more. No longer is it the day when the young and Rubicon media guru walks in and says, you need to spend a half billion dollars. And this is where you're gonna spend it. Now you have the internal marketing department saying, well, we're gonna do this particular ad at this time after this program, because it gives us this response on sales. We're gonna do these print ads, these radio ads, etc., etc., being able to target a particular audience in a particular place even. And maybe even doing phone calls out to them with an automated dialer of some sort to solicit business in different ways. There's really a lot of different components here that let you build better solutions. Big emphasis on quality, big emphasis on process redesign, big emphasis also, as if one can have this many priorities at once on time as a competitive weapon. And I just want to talk about this very briefly. The idea being that information systems used to be mostly batch, which meant an event occurred. Somehow you collected the data, transmitted it, gathered it, massaged it, did things to it, and finally you could cut a decision. And that would flow back, hopefully, if you had a good business. It would flow back into your basic process so that the next event that occurred, the next purchase that was made, was informed by what you'd learned from that first cycle. And what's happening is that this cycle is really cutting down and down and down with things like laser scanners in the supermarkets. You can now decrement inventory immediately when something is picked up. The ridiculous example of this, I was at a big soda manufacturer recently. And the ridiculous example is that when you uncork a soda pop can, there's a transmitter in the cap that tells you that somebody just opened and consumed your product. And we were envisioning, you know how they have network monitoring systems with the whole map of the US? It was like, the ballgame just let out in San Francisco, you know? Things like that. And on the right I have two pie charts here of what the world probably looks like if you have a field force or a sales force, let's say, and what you'd like to have it look like. The big white piece in the first pie chart is paperwork. It's a lot of redundant time, a lot of effort taken out of their day. What you'd really like to do is have people either cold calling or in face-to-face meetings as much as possible on the assumption that that's where actually the work gets done, that's where sales get increased. So this idea of harnessing time and therefore information systems is driving a lot of these kinds of technologies. So how do you harness time? You can answer questions more rapidly or process orders more rapidly. You can repair down to equipment more rapidly. You can even sometimes prevent the equipment from going down. Some Xerox photocopiers now have an embedded expert system called RIC. And this expert system is plugged into a modem. So when sensors inside the machine detect the machine's about to fail, the machine places a call to the service rep who shows up hopefully in time to prevent failure. Copiers, escalators, elevators, electronic mail and voice mail are all things that everybody notices when they go down. They're all sort of in this, they're of a common logical type. Also maybe putting a contract in front of a prospect faster. JP Morgan uses an airline leasing program that they developed where the system with some very simple tools goes out onto an online database, checks some various leasing options given what they're trying to, to the package they're trying to put together, comes back, formats a document and creates a boilerplate contract customized to that situation in about three hours that used to take about 10 days. And then what they do is they iterate that contract. And so every time they think about it again, check a few new options that really let them get much further into the contract. And hopefully also they don't iterate the same exact length of time that they used to take to do the contract. But they get it, or the proposal, but they get it in a lot sooner and hopefully win the business. And there's even ways that you can actually save your client time by doing some of their process for them. So before I jump into the technology stuff, this is end of brief business digression. Before I jump into the technology stuff, let me tell you where we think we are now. This chart, I like to call the Pig in a Python chart for, I guess, obvious reasons. In fact, I was presenting this to a fellow in Europe once and he said, no, no, no, no, you mean the elephant in a boa constrictor chart? And I was like, what, what? And he said, the little prince in the middle of there, there's an anecdote about that, so I said, fine, there's deep literary illusions here if you want them. But that's not what my presentation is about. What we mean here is that over time, this is, I'm taking a position here, I'm saying the worst case scenario is that this graph is actually not true, that the chart really looks like that. But over time systems simply get more complex and people will not actually be able to harness them and use them more appropriately. We believe actually that there's a period where we will get to a period where things are actually easier to use, that the tool gets out from in between us and our work and allows us to do our work and enables us and assists us. And I think that's pretty much what everybody in the software industry is trying to do. What many people also in the hardware industry are trying to do with these new devices like pocket PCs, like pen-based computers, like whatever else. And we are not over that hump yet. And the things that make up cruising over this hump, trying to build some of the complexity of using the system into the system itself. So that for instance, if I'm on a wireless network, if I have some kind of a portable platform that has either a radio or an infrared link, and I go from here to the next building, I'm on a different local area network. And typically with most network protocols, if you unplug a machine, you need a network administrator to sit down and say, machine is gone, and change the routing tables. Or if somebody plugs in a new machine, they need to publish that new machine to the network. Well, you don't want to have to do that kind of stuff. You want the intelligence to be built into the network so that when I leave and go somewhere else, all the routing tables and so forth are updated. And I don't, I as an end user, don't have to worry about that kind of stuff. Similarly, if I'm writing a document in WordPerfect and somebody else uses Word, and I want to send them the document. If possible, I don't want to have to worry about format conversion and all that. I'd like the network to take care of that transparently for me. So I'll come back to this a lot. What we call the phase right now that we're in is the federal highway project's face of information systems, more specifically of data processing systems. By this we mean, this is the time, like in the 60s and 70s in the United States, where the interstate highways are being built. PCs are being sold, lands are being connected. The proportion of lands that is connected is growing slowly and steadily year after year. We're putting in gateways, bridges, routers, all the things that are really analogous to this, this idea of an information highway in the United States. What we're worried about as new science, because we try to look at emerging technology, we have a charted look only at emerging technologies and try to help people assimilate them. We're worried about companies that look into this, this federal highway project's phase, develop what they call their IS architecture, decide they're going to be UNIX compliant, which means POSIX and X-terminals and TCPIP on the network and this risk architecture and so forth. And then they say, fine, we're done. Then they go back and they decide to implement and they build their sort of data network. We're worried about things like, what if later on down the road you wanted to fold your voice network on top of this? Have you actually locked yourself in to certain choices by building the strategic architecture and sticking to some vendors and being principled and architectured and layered and all these things? Have you perhaps locked yourself away so that things you might want to do in five years are more difficult to do at that point in time? So we try to throw things into the pot that will help people think about these sorts of questions. Everybody tells us that things are getting lighter, smaller, faster and cheaper. We hear that chips are cheaper and RAM seems to be getting cheaper and so forth and so on. I have a chart I didn't bring with me where I just plotted the 8086 family, the X86 family and the 680XO family over time year by year and it's geometric. If you just plot the raw MIPS ratings, the top MIPS ratings and I realize that MIPS is not an excellent measure of performance but it is exponential and you look at that and then you ask an audience in 1981, 82, how much did you pay to put a workstation on somebody's desk? And they go, well, three to $5,000 and what could that person do on their desk? Well, they could do word processing, they could do a spreadsheet and if they were really advanced maybe they could do email or file sharing or something like that and then you ask them, what does it cost today to put a workstation on somebody's desk? And it's three to $5,000. And what can they do? They can do word processing, they can do spreadsheets, they can do more sophisticated email, they have a mouse and GUI and they have all these sort of peachy new things but that really does not explain where all those extra MIPS are going. And there's a real sense of frustration on the part of people that are spending all this money because now it's not enough to put an 8086 on somebody's desk. They've got to have that 386, 486. There's not enough payoff coming out the back end for where all these extra cycles are going. And also, our complexity is still increasing. We still have a lot of duplication where there's a network, there's a land manager who keeps track of all the names of people that have PCs and are plugged into the land. There's a voicemail administrator, there's a phone system administrator, there's a security administrator for those people that need to get on the mainframe. And every one of these people has a list of the people in your company, accompanied by a few more attributes, that's all. And when somebody changes departments or leaves the company or whatever, you have to cascade physically a series of messages around to make sure that these changes are made, et cetera, et cetera. I'm just giving you one sort of trivial example of the duplication. As I said, the cost of your average system is not decreasing. We're also always looking for the silver bullet. There's a real tendency, and I don't know if this is because of our society or what's going on, but there's a real tendency to look for the next 123, the next page maker, the next single entity that's actually going to make our life easier and solve all our problems and cure our woes and build a big market. And I would postulate that there isn't really a silver bullet, that there's a lot of really good stuff out there already, and what we need to do is use it more intelligently and do quite a bit of integration, do quite a bit of fitting together of the pieces that exist now. And I will talk a lot more about that. The kinds of integration that are necessary are really predicated on the idea that our current information environment is fragmented, it's broken, there's a lot of what I call seams, you can call them gaps, you can call them anything you want, but there's really a lot of places where people are not able to use their information environment as usefully as they would like to. So for example, if I change places, if I get up from my desk and I go down the hall, or if I go outdoors, or if I change modes, if I turn from my computer to my telephone, usually I have to take the telephone number off the email message, rekey it into the phone just to dial out, and the number of people that have auto dial, which is not rocket science, is amazingly small. And this is because we tend to look at the phone network as this separate thing, this separate entity. We don't spend a lot of time doing integration, the integration that is done between computers and phones these days is mostly in a customer service operation in the back office where a call comes in, you take the caller ID number, you look it up in the database, you try to bring up their data so that when you route that call to an agent, they actually not only get the call, but they find out who the person is. And then we've learned over time not to scare that person by saying, hello, Mr. Jones, we actually let them, we greet them normally, we let them tell us who they are, but by then we don't have to wait 15 seconds for a transaction, that's really where this has been driven lately. And there's also some gaps in the context between work and personal life, and I'm not going to go into that very much at all. In line also with the idea of progress and integration is I wanna start with this idea of linear progressions, of evolution. This is sort of the Darwinian view of what's going on. First we had the personal computer, then we had legables, if you remember the sewing machine-sized things, the K-Pros and the very first few of those, portables, then we moved to laptops, notebooks, palm tops, and now we have some of these new pen-based computers. There's a real propensity to look at these things as a linear evolution, the pocket or the HP95LX being sort of end points of this evolution. And I say end points because there's not a lot smaller you can get a keyboard. And still be useful in any way. So we really are getting to some point where there's a physical limit on the keyboard as an interface. Similarly on the telephone, we've moved from radio telephones to cellular, to mobile phones, to PCNs, personal communication networks, personal communication services, always smaller, lighter, somewhat cheaper service. Well, a little bit later on I'm gonna talk about what happens when these things start crossing each other, when you get some hybridization of functions. I'll come back to that. So the goals of the CIE service are to try to eliminate the seams. Sorry about the wackiness here. To try to eliminate those seams, to identify them. I had a real interesting conversation with a fellow who we were talking about seamlessness as a goal. And I was sort of talking about it. And he said, well, hold on, look at your shirt. I mean, you're wearing a shirt and your shirt has seams. And the reason your shirt has seams is that it's a lot easier to build a shirt that has seams in it. It's real hard to build shirts in all different sizes if you don't want to have these lumpy things. So maybe what we want is not a seamless environment, but a well-seemed, carefully architected, or a seam-full environment. Which is a subtle difference, but it does point you toward layers and architectures and APIs and good definition of the environment you're trying to work around in. And I like that a lot. We're also working on how to hide complexity. Goals like having a single virtual address for a particular worker where eventually, let's say 10 years down the road, you'll try to call me or you'll try to send me a document or you'll try to fax me something or you'll try to send me a file, and you'll really have one alias for me. You'll have a single address that locates me as a worker as I walk around. I'm very likely to be carrying some kind of device, or at least when I get back to my desk, I'll be able to find that message in the appropriate format. And hopefully there's some idea of location independence and media independence here, where it doesn't matter whether this is a multimedia document, an image embedded in a text file, what have you. And there's another goal of the service, and this is almost political, but it's about user empowerment versus control. And let me give you a couple of examples here. If I put a handheld terminal in some route delivery person's hands, it's really easy to timestamp every keystroke, and I could do tailor analysis on this person's job and make their life miserable instantaneously. The more portable platforms and transaction systems we have and the more data is going through the network and things are cascading through, the easier it is to sit down and start doing that kind of work. It's not a good idea. That's how you alienate workers, that's how you turn people's lives miserable. Similarly, think about what you think when somebody's wearing a pager. And I'm sorry to those of you who wear pagers. But there's this tendency to think, technology anchor, this is an albatross, this is you are on call all the time, you have to respond to this thing. And it took the pager companies an unbelievable length of time to figure out that they should have one that shakes instead of beeping like wild, so that when you were in the theater, you weren't embarrassed and had to rush out, things like that. So let's picture then that we have a personal telephone. Let's say we're gonna give our workers a cordless phone that they can carry around their office, which should happen five years from now. Well, to an uneducated user, the easiest thing for them to think about this new cordless phone is that it's a two-way high bandwidth pager. It's an on call device. Why should I carry the phone? Because then the boss is always gonna know where he'll be able to hit a button, ring my phone, and what do you mean you didn't answer the phone? You were carrying a pocket phone, I gave you one. Okay, that's the worst side of the control element of using these kinds of technologies that I'm gonna present to you here. The other alternative is sort of different. Let's say that this meeting is being held five years from now. And because of the kind of guy I am, I would have a pocket phone and I would also maybe have a pen-based computer. My clipboard would actually be a pen-based computer and I'd be running calendaring software on my pen-based system. And on the calendar it says that at three o'clock I'm supposed to give a speech today. So it has a system clock. Assuming I don't change anything on my calendar, three o'clock rolls around, it should immediately assume I'm in a meeting. So the system should switch to what I call meeting rules. And meeting rules are kind of simple. Meeting rules means that my system sends a message over a wireless network to whatever my message hub is, wherever messages come into and are sorted out to get to me. And it tells that PBX file server, email server, whatever it is, it tells it, don't bother Jerry. Send all his messages to voicemail. This is perfectly normal, perfectly acceptable, social thing to do these days. Some people get grumpy about being lost in email, but this is not rocket science. This is a simple, simple transaction. It's a pointer in the location database. I could do a little bit better. I could say, send all my messages to my assistant who knows how to do triage on these calls and will ring my phone if it's really important. Or I could go one step better and I could say, I could look in my own little directory here. I would have a directory listing, which I use for email, for voice, for whatever other medium. I could put a check mark by the three people that I've been trying to get a hold of but haven't been able to get a hold of in the last couple weeks. And I could just, then when the system switches over, the message would include those three people's numbers and when my PBX sees the inbound call, many of which have these numbers coming in with them, it could match it up. If it's one of those people, my phone actually does ring. So I could then put up a very simple filter to make myself available to the people that I wanna be available to, unavailable to the rest and begin to stagger out my work. I could have effectively a software do not disturb button or a limited do not disturb kind of button. Things of that nature. And this is one kind of design for empowerment where this kind of a setup and it's just the kind of setup that I particularly would like but yours will differ. I'm almost guaranteeing it. That you have different ideas about how you want your environment to be shaped and what kind of tool to have in your hand, whether it's an earphone with a boom, whether it's a no phone at all and no device at all. You just want to carry your file of facts and have somebody print your calendars. Many, many different ways that people wanna do this. But these sorts of technologies allow us to have this kind of freedom. But the difference between design for control and design for empowerment is sometimes very subtle and it's real important to look at it and try to build it in. The new technologies I'm going to be talking about have lots of open up lots of new market opportunities. And I've just scattered a series of technologies on here. The VRU on the left for instance is a voice response unit, lead tracking systems. So for instance, in a Salesforce automation environment you might have lead tracking software that helps you manage whom you need to call next, what you need to do. You might have auto dialing capabilities from that software that lets you ring the phone out. You might have computer integrated telephony systems for inbound numbers that do what I was talking about earlier and you might have a voice response unit to handle simple inquiries so that people, let's voice responses hit one for this, two for that, three to be sent to Nepal. But I want to also say, and I don't have time to talk about the different vertical markets, but the horizontal market is equally important and if you talk to a lot of the pen system vendors for instance, they'll talk heavily about vertical markets and they're very interested in. I imagine that every insurance IS person who develops things for their car collision estimation department has been approached already. Somebody's talked to this guy and said, yeah, look, we can do this on a pen system and look how it'll work and here's a demo of how it works and so forth. But don't ignore the general office market, the broad horizontal market, because a lot of the things we're talking about here do enable people to do their work more effectively to begin to ignore the medium and the tools and get further along. Now the box I forgot with all the samples of research also had in it a series of copies of this which is the scope diagram of the continuous information environment service. I apologize for that and we can get you copies later. I will keep it on the side so if you'd like to see it again, I can bring it back up. There's a lot of words on it as well but let me explain how to read this now. The two small circles in the middle are the hardware and software that people will be using a decade from now and I don't care whether the hardware has a keyboard on it, whether it has a pen interface, whether it has a touch screen or speech recognition or whether it's completely a telephone like the cellular phone down on the right or completely a computer like the laptop on the left or if it's some hybrid mixture in between it's mostly a computer because it does computing kind of things but it happens to have a microphone and a speaker and it happens to have a wireless link that's either a cellular voice or cellular data so by all intents and purposes it's also a telephone. It's just not a complete telephone, things like that. So there's sort of gray areas in between and companies will be producing products of all different sorts of natures and form factors. Then the circle above is advanced interfaces ranging from personal productivity stuff like personal information managers out toward group work. Out toward the ability of not just facilitating individual productivity but getting people to work better together and this is very critical. We believe that you can work like crazy and build the ultimate personal productivity tool but that's not enough and trying to understand this ether that connects us and the medium that we have and the new tools that we have at our disposal and how they change the way we work is critically important and I will come back to that. The big circle on the outside, the big oval is what we're calling a continuous information architecture. This is the infrastructure that needs to be put in place and needs to be transparent to the end users. I as an end user, as I said before, shouldn't have to worry about whether I'm attached or not attached to the LAN. I should have things like deferred input and output that keeps track of and if I break my connection at an inauspicious moment it has graceful failure and it says, oh, didn't manage to send that message so we'll send it next time we have a connection. Things of that nature. I shouldn't have to worry about the fact that well, a decade ago, 15 years ago we had the telephone and we had the telex and some fax. Well now all of a sudden we have the telephone and voicemail, tremendous volume of voicemail then we have facsimile has replaced the telex almost completely and we have electronic mail and full format file transfer and in some cases video conferencing and every time you add a new channel there's a new place to go look for your messages. I mean, I call from on the road and check the voicemail, ask somebody to read me my email, check in the box to see if a fax came in. We have a system at work where if a fax comes in they send me an email message, that's very nice. That's new science integration. I promise you we try to advise better than we do. But the whole idea here is to try to create a single virtual mailbox that both allows you to collect all these different message types, allows you to filter some of them out, show me only the messages I got in the last four hours, the last 24, whatever you want, show me only my voice messages, show me only the faxes, do something with it and then also some smarts in the mailbox and that means things like beyond mail, things that let you set up a couple simple rules and bounce messages. We're not talking here about deep embedded AI, we're not talking here about something that automates very difficult tasks. Our own opinion is that that's really very difficult to do and extremely difficult to reconcile with users to let them understand what is capable and what is not possible and how to actually design their own environment. So what we're trying to do is what are the simplest set of rules, what are the simplest set of features that provide 80% of the power for 20% of the effort and that's a common theme in this particular service is sort of application of the Pareto principle. How do you have some kind of a vision or a goal that's ambitious, fit it to your business and then use common sense and realism and budgets and depreciation cycles to add features and pick vendors now that get you toward that vision in the longer term but you need to have a vision otherwise you just don't get anywhere. And then just as an example on the left we have information access and conversion. This means that there's already a lot of good information out there in your company's databases. There really is. It's just that sometimes it's hidden by SQL which is a foreign language or it's hidden by local area networks and their gateways and the fact that you have to hop and skip to get into these databases or it's hidden by the mainframe security system. So there's some products out there like channel computing's force and trees, info alliance from software publishing corp, really not a lot of products in this category but these products try to make your life easier and you're trying to collect information either once or over and over again on a schedule so that you don't actually have to execute these particular steps. This cuts across also into things like HP New Wave with its agent task language, things of that nature. I'll come back to this diagram. I just wanted to present to you the scope, the general contents of the service. What I'm going to do actually, I'm gonna use this as a roadmap. I'm gonna start with some hardware, move up to the software and then go over to transparent messaging and back around the outside circle. I'm gonna try to do it pretty quickly because we don't have that much time and I'd like to leave some time for questions. Ultra portable systems are gonna come in many different form factors. They're gonna be like a phone, like a computer, with a pen, without a pen. I've already talked somewhat about this. Some of the critical issues you are probably extremely aware of. I'm not going to go into very many of them. The whole idea of you want this thing to be on for a while, you'd like to have some kind of a wireless link if you could and if it didn't cost too much money, et cetera, et cetera. Let me talk some more about just the software and let me take more time actually and go into this idea of advanced interfaces and what's going on. There's two different ways of looking at emerging new things like pens. And I'm gonna contrast two views here. One view I call the linear progression or the Darwinian view, which is you have this island hopping kind of strategy. First we have the command line interface, then next island was the windows, icons, mouse and pull down menus interface. And the next island is the pens. And finally, and each one of these new islands sort of obviates or absolutes the previous island. And then finally we're gonna get to the real computer, which is this continuous speaker independent speech recognition, where I just, it's like the Apple knowledge navigator video where I speak and it knows what I want and it goes out and it finds the information, comes back, massages it and presents it to me. The other point of view is what I call the market share view. And this says that no particular new interface technology completely absolutes an old one. It just replaces some of the tasks that the old one used to do and hopefully fits in as part of a market pie as a market share. And originally we had command line interfaces so that was 100% of the pie. Then mouse and windows came through into the marketplace so it took a large piece of the pie. What's happening now is the same thing with pens. They're gonna take a piece and eventually speech recognition which now has a very small sliver will take a somewhat larger share. But our perspective on this is that speech is actually a niche application. That the pen should not be ignored because it's very likely to take a large share of the time and energy that we spend interacting with machines. That it's a novel device. There's a lot of things you can do with it that are really very powerful. Also it's a social device. There's a tremendous social fit that pens have. The example I like to give is it would be perfectly acceptable if before this meeting or if during this meeting if this were less formal we were all sitting on a table. I would take my clipboard pen machine and lean back and answer my email or annotate somebody's document and check off changes they have to make. Whatever else like freestyle from Wang or like some of the pen systems that are coming out now. It would not be very good. It would not be very socially acceptable. I was sitting around in the room saying, okay file open, what do I call it? Foo.bar, it's not very social. Unfortunately also if your job involves interacting with a machine all day long or issuing commands all day long, your voice wears out. I mean you have about two hours worth of talking voice, constant talking voice in you and then that wears out. So we're not, I mean it's very likely, it's even probable that we'll have a machine that has a keyboard for power entry for people that know how to type. Maybe a specialized recognition module that when they want to dictate will recognize some of their voice and power type for them which will probably be trained to that particular user and maybe also some kind of a limited speech module where I wanna switch from draw mode to type mode and I would just say draw type. And maybe that would help switch layers in a modal interface, things of that nature. But not likely that the manipulation will be done with our voice. The pin is actually extremely powerful. It's powerful because it's transparent, it's natural, we know a lot about it. We've been trained with it. You've heard this, if you've heard any pin presentation already, you've heard that they teach us from grade zero how to use crayons, slate and pins and all this other good stuff. If you go to a Walder school, you actually get to go through a progression. And we have this vocabulary with them of things like proofreader's marks. If I draw a line through a word and put a pigtail on it, pretty much everyone in the room knows what I wanna do with it. If I draw a line through the whole document and put a pigtail on it. And you know to go update your resume. This is very, this is a powerful vocabulary. I put a carrot between two words. A little window will pop up, it'll ask me what I wanna write in, I write it, it recognizes it, I hit done, it goes back into the document, flowed into the words as if I had typed it in with a word processor. So there's a lot that you can grab and use that's really extremely powerful. And you can even sort of start pushing beyond and the example I give here is of cascading forms where instead of having five part forms, you could actually have a form that at the very beginning is very small on the screen. You're a real estate appraiser. You walk up and all the thing says is this residential or commercial? You check on commercial and then the form starts to unfold on you but only the options that are viable given which path you've taken through this pretty dynamic form are actually available. And once you're done checking off the boxes, filling in the words, your whole transaction is in, you can either plug the machine into a network and upload the information or whatever else you need to do but there's really a series of more powerful things. You can imagine perhaps also that the pen doesn't just have to be a stylus. You could have the pen as an instrument. So let's say I'm an architect and I hear about this new property that's opened up for a bid. Daimler-Benz has bought property on Potsdam or Pots in Berlin. Ooh, I'd like to bid on that. So you go to the site, you take your instrument, you put it down several places around the site. You maybe press the barrel button or you press something on your screen. It then records angle, position, maybe it's even a geopositional system. And what you have then is a wireframe of that particular piece of property. You could then take a CAD system and begin doing what ifs on what I would build on that particular property. So you link a couple of different pieces of technology but if you start thinking of this stylus as maybe not just a tablet writer but as an instrument, you could also maybe see that it's a test probe for a hardware engineer and you can run different kinds of hardware to do different kinds of measurement and so forth. Very briefly, I don't have much time to go into pen system details. Couple of different form factors hitting the market. Most systems look like the one on the lower right. Eight and a half by 11, clipboard shape, inch and a half thick, four to six pounds heavy. Some have backlit displays, some don't, some have a hard drive, some have no moving parts, they just have flash memory. Series of very minor differences really between those kind of platforms. And then Momenta, for instance, has come up with a machine that out of the box is a complete laptop PC. It'll run MS-DOS, it'll run Windows for Pen. It has a keyboard, which many of the other units actually cost extra, has an integral fax modem and a tethered pen. So there's sort of several different options in ways that you can use the system. And I don't really have enough time to go more deeply into that. So if you're interested, pull me back to it. Let me now move out from the pen interface itself up one layer, just a personal productivity very quickly and then out to messaging and group work because that's where I would like to spend some time. I've talked already somewhat about user agents and PIMs and mailbox management, so I'm actually even gonna skip that and go out toward group work. Sorry about the complexity of this chart. You can ignore the right hand column completely if you'd like to because all that's in the right hand column is examples of companies that have products that fit the particular tasks that are in the middle. The way to read this chart, well, this chart is in response to the usual group work chart which is a four square matrix that says same time, same place, different time, different place. You're probably familiar with that. Well, that didn't seem to do it. I mean, that didn't seem to differentiate between group work tasks well enough. And the tools on the right are placed here. They don't always just fall into one category. There's some tools like notes that perform tasks that are available in several different categories. But the way to read this chart is to come in from the outside to effectively come in from on top and start thinking if I, you're sort of increasing intimacy or closeness to someone else that you're trying to work with. So if all I have in common with somebody is the same network, then I can do things like electronic mail filtering beyond mail. I can do conferencing and scheduling software, things of that nature. If I go down a level and we're actually on the same project together, then we might do project management software. And by this, I don't mean something that gives one person a per chart and a CPM chart and lets one person know where you are. By this, I mean I'm finished writing a piece of code. I put it away in the shared document library. Somewhere there's a dependency map that knows who else needs to know that I'm done. Messages are automatically cascaded out to those groups that I'm done or I'm behind schedule. It's actually group work, like an Ipsy sort of integrated project support environments with some added value. Further down, let's say we're in the same task group. This is where things like workflow software come in, which a lot of people hold is synonymous with groupware. We think workflow software is one stratum in a broader model of groupware. And the idea here is we're in the same task group so we can probably figure out that the junior clerk usually gets the document first. They do this, X, Y, and Z, things to it. They then approve or decline or maybe they don't have authority. You figure out some kind of routing through the department. That's what workflow software kind of is. Or you could create shared document databases where we're all creating documents that we wanna see at some point where the value of having an ability to search into an aggregation of these documents is very high to the particular work group. These are just some examples of tool types or tasks that you could do at that level. Let's say we have more in common. We're not just in the same task group, but we're trying to create the same document. So here you could use document tracking systems. And notice that I'm not on at the same time or the same place yet. So here I'm just doing sequential document tracking, which means things like marking up a document. Somebody creates it. They send it to me. Asynchronously, I mark it up. I forward it to someone else. It gets back to the author. They do the changes. Go down one level. Let's say we're on at the same time. And then you have shared image work where when I do my marks on the document, you see it on your screen. And maybe we have a phone conversation going at the same time. Or maybe we're doing video conferencing and we have the document on something like an overhead and it's being transmitted. And we can all talk about it. Maybe it's an engineering drawing and we can make changes to it. Finally, if you're on at the same time, same place, same document, you have things like a computer aided meeting room, things like the Xerox CoLab experiments and there's been a series of other experiments. In fact, Elmer, the last thing I have out there is just a piece of furniture. Steelcase designed some furniture specifically for computer aided meeting rooms. They make it easy to embed a CPU and put a screen in the keyboard and then hook the different things together. So even furniture is a technology. And then at the very core of this diagram, we've got personal information managers and personal productivity tools where you no longer share anything with anybody else, but this is the heart of your work. This is where you do all of your work and this is here for a couple of different reasons. Reason one is that if you take a tool like Agenda and you hook it up to the email network, you have simple message filtering. You have a groupware kind of tool. You have groupware feature because you have smart parsing and the ability to look at information with the Agenda tool. So it almost makes this thing loop together. Second reason is that one of the tragedies of the groupware market right now is to use any of the tools on the right hand side, you've got to get out of your basic productivity tools. This is not universally true, but for most of them, if you want to use the tool, you've got to either create the document in it or translate the document into it or somehow get into this other environment. As opposed perhaps to I'm sitting in my favorite word processor, I decide I want to do shared markup. So I invoke a shared markup capability which lays a few more commands on top and allows me then to make a mark and then you'll see it, which normally wouldn't be happening. Where there's sort of an extra, there's an enhancement of the tool that I use all the time that gives it group work flavor. This is, we believe, one way of making systems more transparent, more easy to use. That's your two minute tour of the groupware chart. Sorry for the speed. Around the outside, these are the general topics. There's the idea of unified messaging. The three around the bottom that I refer to sort of as a smart infrastructure and information access and conversion. Let me take them in sequence. This is just a milestone chart. Messaging, we're looking at as hopefully working toward a situation where people are indifferent as to what medium the message is coming in or going out over. For example, let's say I come back to my desk and on my screen it says Jane tried to call me. There's a voicemail message and it knows it was Jane because it looked up the number and looked in the database. And so there's a message there. So I select that particular line and I hit enter and it automatically auto dials Jane because it captured the number. Why can't it dial the number back out? So then I get Jane on the phone and we're talking and it turns out that in the middle of this conversation I realize that I just drew a graphic that's very relevant or I made a table that I'd like to show her so I look for it on my own machine and then I say share or some other command and this document is sent to Jane over the medium and then she can see it and then we can do group markup or whatever else. Or maybe there's a video clip that I'd like to send. Things of that nature. And then we could bring down the document connection. We could keep the phone connection. We could maybe switch to something else. But the idea here is that neither of us should have to worry about all the different addresses and nodes and formats and all those sorts of things. Then I've actually already talked about message unification. I've not talked much about interactive voice response or electronic data interchange but I really don't have time right now and in the interest of leaving questions at the end I will skip that. On outbound messages what putting some of the smarts into the network means is I have a document that I've created in some format. Let's say word perfect and I want to send it to some person. So I should be able to look them up on my global address list which should not only know something about people and where they are but also about what device they have and what their favorite piece of software is. Which means then that I don't really wind up worrying about the middle two columns. The network as much as possible resolves these issues and decides okay we need to filter it out into a new format we need to send it over this email gateway or my destination, my recipient doesn't have a PC but here's our nearest fax number. So my document is converted into fax format sent to my fax with the number and sent forward. Things of that nature. I can talk if you're interested about wireless communication about intelligent buildings about some voice and data topics about format filters and converters but I'm not going to because we're close to the end and I wanna say a few things in conclusion. So and then give you time to take me back to any topic. There's a few underlying trends that you've probably been hearing that I just wanna make more explicit. One of them is that the phone and the computer are climbing on board. In the decade we're not so much likely to have a desktop phone or a desktop PC as we are to have something that comes with us. We might also have a desktop PC that happens to be our preference but if we wanted to we could have something that we could cart around with us. We become a normal part of our work day. The phones and the computers are also merging and I know that voice and data integration has been a holy grail of sorts and there's been a lot of visible failures in the market but slowly and surely you see some of the protocols and some of the abilities. You see multimedia workstations like the next machine where these things are really treated in a much more transparent fashion. There's a lot of evidence to say this is still happening. There's a lot of evidence also that says that companies that use these things smartly that do a lot of smart integration get tremendous benefits out the other end. And finally I wanna say talk about the hybridization of functions. Go back to the linear or evolutionary or Darwinian version of what's going on and cross that with another thought that it's not just evolutionary and it's not just Darwinian but we are messing around. It's genetic engineering and software is the ultimate genetic code. So it used to be I had to go out in the marketplace and buy a fax and buy a modem. Well now I can walk out there and I can buy a fax modem and a couple of them actually even work. Somebody took some firmware from one machine added to the other added a little more hardware capability but for 110% of the price of one of those devices I can buy one device that does both interchangeably. Same thing is happening let's say on network printers. I can go to Canon or Ricoh and buy a printer that's also a fax server, a low volume scanner or a copier and maybe even a file server maybe even has some file storage locally. The Xerox docutech machine is also a hybrid machine has a lot of different capabilities. And in software this is happening too. You see companies that started out in one particular niche and are now broadening out. You see WordPerfect a while ago was running some print ads where there's a guy standing up saying I just made a decision in five minutes that used to take me three days and what he's saying is that in what used to be a word processing company and is now pitching itself as an office automation system he did a group where a kind of task he routed a document and got approval on it and got it back. So every company in the marketplace once it saturates and begins to understand its own market cuts into others and therefore begins to do some of this hybridization also either by generating the code or by licensing it. What all of this gives us what all of these different techniques and technologies give us are a richer variety of options. The chart on the left just says that there's a client or a customer out there somewhere opposite you and you are a company. In a decade, well two decades ago all that would happen would be the person would call you up and some human would answer the phone and answer the question and get back. But some of those questions were very simple and some of them were actually very complicated. Some of them were what are your store hours? What is your mailing address? And some of them were much more complicated. We need to negotiate a contract. So as these different technologies come in they take little bites out of this space and some of these technologies are more on the phone market some of them are in the computer marketplace. So for instance, audio text which is a simple ability to play back messages when people call in that lets you broadcast your store hours when you call the movies they tell you what their movie times are things of that nature. It takes a bite out of this type of series of transactions you have with people either within or outside your company. What it winds up giving you first of all has more ways to get information and get good information but also at the very bottom you have the choice to use people when you really want to not because you have to. And for instance, I was talking to an ACD vendor and he was mentioning that LL Bean doesn't want to replace their human operators. I mean a part of the call to LL Bean a part of their service value is that the caller sort of feels like that person is sitting on the shore of Maine watching the waves break and the gulls fly overhead and there's a warm voice at the other end and there's a certain amount of companionship in the call and they see that as an important part of their service. So continuous information environment means continuous across places continuous across document types, media types. It also means trying to focus on information and not just on the devices and the software and so forth. And I want to go back to actually make for the first time a point about types of integration that we imply and there's three kinds of integration that we're talking about here. One type is integration between new devices the funky new stuff that's coming out and the current infrastructure because if your new device can't pass messages or share files or somehow talk with the current infrastructure it's really doomed to a short career. There's a real need for integration with what already is out there. The second kind of integration is between the computer and the phone and this I gave you the example of the pocket phone that I would have five years from now and my ability to have that phone not bother me at some point if I don't want to because there's software running somewhere that knows about what I'm doing at the time. And then finally the integration between an individual's work and other individual's work. It's the idea that it's not just individual productivity that matters but it's group work. So moderately general conclusions changes the only constant. There's no silver bullet. Think about integration. Look for these hybrids. Look for interesting milestones and leaps and don't ignore them. Don't pocket them away somewhere and ignore them because they don't happen to fit neatly into a category and categories are important but rather try to periodically violate those category boundaries and see what happens. See what it lets you think of that's new. And in order to go forward manage the seams, be architected, be seamful. Manage the applications so that you build things that empower people and don't control them and pay a lot of attention to technology transfer. And that means internally within the company as well as with your client basis. And at this point I'd be very happy to take questions and I'm sorry to have gone so fast but we didn't have a lot of time. Please. Can you hear me about wireless? About wireless? Okay, what, can you be a little more specific? What your interest is? How's that technology going to be? Okay. There's- You picked up slides that I could talk about this but I'm not good. Okay. I'd love to see the slides. Well, I can sort of tell you what's on it just as easily. The wireless world is pretty well split right now between voice and data and then between outdoors, mobile which is mostly like outdoors and then indoors. So let me sort of take a couple different pieces. Outdoors voice means cellular today and that is slowly moving both within the cellular market towards digital and they're trying to make up their minds what standards to use and also towards things called personal communication services which would be microcells, very low power hopefully very cheap handsets that you can carry around. The first versions- Are you strict into very small areas? The first versions of those would be dial out only and there's a system in London called TelePoint which has actually not been very successful and that would be effectively a phone booth in your pocket. You go up and you stand near a TelePoint which is like, almost like a phone booth but it's a big mark on the outside of a building in the downtown area and you pick up your phone and you dial out and it's like having a cordless phone and it connects up to that TelePoint. Yes. I'd like to follow on what he was saying too. Existing infrastructure strikes me it's a real impediment as much as friend-mind is open for the country of Namibia selling their telephones and because they're so backward they've never had a wire that would be much cheaper for them to hook up with leading edge technology. Exactly. Just like wire Germany and Japan has pearl harvests ahead of us it's like well they had all their infrastructure Yeah, exactly. The only major problem in those countries is foreign exchange. Talk to Argentina, talk to Soviet Union where there's no infrastructure there's also no money to pay for the state of the art systems and it's actually a personal interest of mine to try to figure some mechanism where I don't know habitat for humanity type things you trade sweat equity for the digital switch whatever it is. Also in New Zealand for instance there's no spectrum mess. New Zealand was able to pretty much auction off all of its spectrum and say you know guys do with it what you will well we'll talk about it later and in the U.S. there's this unbelievable and in Europe there's these unbelievable fights over who has rights over what frequencies and who is actually using theirs and who isn't. What's the width of the spectrum? So the TV pieces of the bandwidth are not really being used at all but they want to hold that because HDTV is going to consume a whole lot more of that spectrum. So you have the voice side you have the data side. Data outdoors means things like artists which is the joint venture between IBM and Motorola which is a relatively expensive data packet switch network. With a proprietary and they use Motorola modems of a particular type it's they have their own towers and so on and so forth. The other alternative there right now still emerging is RAM mobile data which uses a standard called MOBA-TEX which comes out of Europe. So there are already a couple MOBA-TEX networks up but five or six countries like in Scandinavia. So there you could use any terminal that adheres to MOBA-TEX standards and you could probably also use it in Europe and RAM has different architecture different sort of a slightly different offering but what's also interesting is with a cellular modem you can sort of do data over the wireless voice network. You can add some extra encoding and decoding like MNP10 and things like that and you can then ship data over the voice network. The only problem is that if you have only a second or two of information to send that way you might have a 15 second call because you've got to bring up the circuit get handshake, bring the circuit back down and that's why packet data is really interesting. There's one company called Cellular Data Inc that is pioneering they take the bandwidth between analog voice channels and cellular and they run a packet data network over that at about 2400 Baud right now. But that's pretty interesting because what you do is you take a cellular base station you add a little bit of equipment there and then you build a reasonably inexpensive terminal and a radio and you all of a sudden have this capability to walk around using this information. So there's some bleeding over between the two. Indoors on the data side you're talking about things like spread spectrum radio or licensed radio or infrared local area networks. And infrared is line of sight only it doesn't go through walls and the other stuff will go through walls to a certain distance which is both a headache and a good thing. I mean if somebody could conceivably put a receiver outside your office wall and decipher, you know, crack your network that's the bad side of radio frequencies. My own feeling is that people will be far more comfortable sitting next to something that has the same technology that's in their TV zapper than they will next to the same technology that's in a radar dome. So I see and also infrared has this there's no cap on the bandwidth. This is just, I mean infrared is going at trillions of cycles per second so what you need is faster sensors, faster whatever whereas a lot of the like cellular frequencies and stuff like that are maxed out and they need to shift technologies but there's only so far they can shift there's only that much movement in the way things like that. What about satellite? Satellite is interesting. It's particularly interesting in the mid range in the mid timeframe because you have things like I talked about personal communication services and telepoint. Well, if you have an outbound only phone how does somebody find you and trying to solve the problem of finding the user is really a thorny issue. Sorry, that's just a funny phrase inside of new science because we have one guy who we joke that he likes to end his articles with thorny issues still remain but trying to find the person's difficult. So how about in the interim before things solidify adding pager functionality to an outbound only phone so that I have an inbound call that lands at the CO at the central office, sorry a signal goes to the paging network which pages my system my system doesn't beep but it causes my phone to dial in and then at the central office the calls are spliced together. So you effectively have inbound calling in the near term without I mean the other answer that some people are looking at is using deep AI techniques to monitor your daily travel patterns and try to track where you might be to ring that particular office and broadcast out a, I mean it gets, by the way those printers are also scanners, faxes and copiers makes it a lot easier to cost justify that particular expense. So you can look for that kind of stuff. Those are just a couple ideas of how to go about it. Please. There's a couple of vendors that sort of do that right now in different ways some of them actually just do cross notification which means that if you call in for voicemail there will be a recorded voice that says by the way you have email waiting others like Centagram in San Francisco has a system called OneCall where it will, with a text to speech module it will read you your email so that is actually available today Oktail is trying to get that kind of functionality but they don't have it yet they just got cross notification and that's sort of a major vendor and then there's lots of probably little guys out there doing similar kinds of things but on a much smaller scale. Okay, we focus on emerging technologies so Gartner is all the traditional technologies and when Gartner cuts into advanced technologies like AI or case tools and things like that our services are dedicated to those particular subjects we just sort of overwhelm them and our scene is technically superior in our particular areas of expertise so we don't aim to replace Gartner Group what we do is we sort of add on to the ends of it Gartner is actually more worried about metagroup than about us because metagroup goes after their core business we're not as numbers oriented as IDC in fact I just my own the way I look at markets and things like that I try not to categorize too much because after a while if you've invested a whole lot of sweat and effort into a major broad market categorization and some new product like Notes comes along it's difficult to categorize you're bound to do something wrong you're bound to try to bludgeon it into the wrong categories or not be flexible enough about the categorization and so forth 80% of our client base is end users end user companies like American Express and Sears and Liberty Mutual and Texaco and Boeing and all industries yes this looks like a great potential for small companies except that the way it's pitched right now is really big accounts and it's hard to explain even what it is even if you show somebody once they start using it everybody starts to use it it goes wow exactly a couple things when I was in the running the Intelligent Document Management Service I wrote an article when we first heard about Notes and the title of the article was I was telling Barbara about this earlier Lotus is no longer a one product company despite the fact that I knew that you sold more CD-ROMs than anybody else and all this other kind of good stuff I saw Notes as being that kind of groundbreaking there's and so we've written articles about it we try to present our perspective and the people that read our stuff are usually directors of advanced technology groups in large corporations so there's one side another thing is I had an insurance company client call me up and say he was interested in negotiating a large site license for Notes and his question was really is there a price cap is there a max price that I know we won't have to pay over and the answer at the time was well you have to negotiate every single site license separately and sorry but what he was trying to do was wean people off of things like WordPerfect or whatever their people's favorite word processor and whatever else were and on to the email network on to the email backbone and I see that as a really really powerful paradigm shift because and I usually sort of tease this out sometimes when I present it but if you use WordPerfect in a wissy wig display your general impulse is to print it at the laser printer, walk over, pick it up and go eh, gotta move that paragraph over or you know this needs to be in bold very form oriented if your message backbone is where you create your messages you're very likely just to type in a message you might type it out to see if it's long what it is but your first instinct also is not to print it and put it in the mail but to just send it over the channel because that's the easiest most natural thing to do the channel is designed that way so the center of your mental universe shifts from being your word processor and this funny cycle with the printer to being your messaging engine and the ability to dump things into this very high speed channel and have things happen and that is what we try to work with a lot because it is a big paradigm shift for people yeah if you leave your name or a card or something like that I'll be happy to send you information please consumers and can I do that in like two minutes we're trying to do a lot of thinking about these things I recently wrote an article whose title was never mind the desktop who will own the household talking about the FNS at local monopoly and how all of a sudden cable vendors are starting to plug in and cellular vendors want to have a wireless last link to the household and so on and so forth just for starters we think the cellular vendors the cellular carriers are actually best positioned right now to control a lot of these links they have the infrastructure, they know how to do billing, they already send a bill to a lot of people a lot of, especially if it's the wireless carrier because they already do residential and landline billing and so forth a lot of them are aggressively exploring with things like data over the cellular network things of that nature fiber to the home is sort of, well my own perspective is that our multimedia extravaganza world is being overblown quite badly and IBM and Apple and companies like them that are saying multimedia is the next big thing I agree and disagree with them I agree because it's compelling and if you're trying to put together a report or if you want to see a movie on demand or whatever else you're going to want to see some video and maybe download it overnight or maybe see it live or whatever else I disagree because I think multimedia authoring is unbelievably difficult and whereas with pen computing you have this handy vocabulary like a squiggle and all this to do a good multimedia presentation you need to be an editor, a choreographer a sound mixer, a this, a that and you need to have this idea of being able to spend a lot of time on it because when you do a film you do this piece, it's like unit test, system test, integration test and you do every one of them over and over and over again until you see the whole thing and we don't yet have, and if somebody comes up with one this will be great, we don't yet have a great powerful vocabulary that's a shorthand for putting these things together so if I had a worker who was coming back to me and gave me a multimedia presentation just for me I wouldn't fire them but I'd get pretty hacked off because my interest again is in the content of the work the form. Now if what they had done was gone to a site and done a video camera tour of the site and they included that in the presentation that's a different thing and I take multimedia the ability to ignore what medium a message is coming in or the information is coming in as a given for your platform in five years so that kind of multimedia if that's all you mean by multimedia and you need a high bandwidth channel for it then that's sort of going to happen over time now N-Ren and Alternat and whatever you want higher speed networks are sort of anomalies and will help some populations but not your broad and users for a while Fiber to the home is going to be a big fight for a while and it depends on well imagine the difference between whether one of the R-Box gets the fiber into your house and whether your local cable TV operator gets the fiber into your house I let the disaster scenario unfold in your own mind so it depends who gets there and what they have to offer look also my real interest is look at the devices and the applications and how that's all going to change because that's going to create new uses that we don't think about today even if it's I scribble a recipe down and I send it to the neighbor and then I don't worry about it because there's a data network that just sends it over and it shows up on some big screen very simple kind of stuff yeah, if I can build a PC and put it on a board and charge you $10 for it I mean effectively the manufacturers cost of a 8086 quality PC and an HDTV is going to cost what, $3,000 to $5,000 or something why can't there be a dozen PCs in there some place so sure have you heard much about Echelon Echelon is a Mountain View startup started by Mike Markola who is the second president of Apple Computer who was tinkering around at home after he he and Jobs hired Scully and brought him in he was tinkering around at home trying to do home automation stuff and getting frustrated as usual anybody who's tried to do that has probably had a hard time and so he hired a team of engineers to go out and prove to him that you could not design a protocol and some circuitry and some hardware and some management systems that would let you do intelligent home control and maybe also distribute a device control in an office or on a manufacturing floor or on a wiring harness inside of a car so that instead of having eight different networks one for the safety one for this and that you'd have one and a bunch of little controllers on it to communicate it and what came out of it was not only is this not impossible but here's the set of protocols that might work and then they launched Echelon and they hired Ken Oshman who is the O in Rome to come in and be the president so they're doing pretty well I don't think they're going to set anybody's house of fire in the next year or two but they're on a very nice course toward establishing local operating networks like Echelons establishing some standards in that marketplace that will be used by lighting and like the Johnson controls of the world and Honeywells and all these people and once you start getting a common language embedded in the devices which is by the way, medium independent you could send the signals over the power line or over an infrared or radio link whoever OEMs this thing just builds the link but there's a media independent interface defined then and they're also shooting for like three to ten dollars per device manufactured cost and that's in the first sort of generation so that's really ambitious but it's good stuff because then you start getting these things embedded into switch plates and thermostats and stereos and what not like the TV answer people? that's one that I saw and I started looking into it there's one CD interactive there you have like a gun or a joystick sort of thing you point to the screen and you pick menu options I think it is a TV answer thing I think there are a couple of trade shows huge booths like 50 by 50 booths I don't know if anybody else has seen full page ads in the Wall Street Journal I talked with the TV answer people and it just seems like they're looking for they're trying to kick up interest so maybe a big booth was just so that they can corral more people walking through my own feeling is that some small companies like TV answer may come up with some interesting ideas but it's going to take a major cellular or cable operator to say okay we're going to go with that standard and start delivering service over it and make something of it but that's eventually going to happen somebody's eventually going to do it I just don't know which one in the next year or so Explorer the reason to start development right now with Windows for Pen is if you already have a legacy application of some sort that maybe runs on laptops and you've spent already a lot of money on it and it's maybe it's a Salesforce automation program or whatever an underwriting program that you know insurance salesman takes to a site and so forth and then you can migrate it onto the platform and you can have your application running otherwise you should at least explore Pinpoint and start getting a couple engineers who understand how to use it and what it lets you do differently and likely not to think of buying and build a strategic application which means first couple kids on the block just postpone you know I look at this as an opportunity to go back to your end users you being MIS let's say to go back to your end users and say you know those laptops you wanted to build, I mean buy you know how they're going to cost like 5,000 a piece and do this and do that well I'm actually going to tell you not to buy them and I'm going to tell you to postpone purchases for a year and a half and at that year and a half mark here's what we're going to deliver and then I would propose a Pen platform instead pardon momenta interesting machine very elegantly designed I mean their case is wonderful and the fact that you can unsnap one side and use the machine immediately that's wonderful I mean the fewer barriers the fewer things you have to do to do any task the more likely anybody is to use it and I think that curve is dramatically exponential or whatever hyper exponential and it's nice that they run Windows for Pen laptop compatible and they have a lot of stuff built in but I was pretty disappointed with their own software environment even though it's on top of digital talk small talk which is good but I don't think their interface design was that good I played around with one of the demos and I managed to lose a graphic and not know how to get it back and I'm supposed to I mean I'm supposed to know how to do those things never mind a retail customer three four years there'll be a major thing like the Macintosh what I tell my business clients is this is kind of like the PCN 81 and the Macintosh in 84 a year or two before nobody had one in their budget zero I mean except for you know with the Macintosh it was C1st Bank and Arthur Young and I don't know who else it's like three companies bought the first the 128k max that was it and then two years afterward when a couple other pieces fall in like Macintosh, Laser Writer Post Script and PageMaker all of a sudden markets open up a whole lot of sales and then there's a paradigm shift because everybody else looks at it and says oh yeah but I want to be doing it that way too so look at that kind of a pattern that kind of growth rate the better news is that whereas the Macintosh was only built by Apple here you have companies like Pinpoint Microsoft that are licensing the environment to everybody and you have Samsung building pen computers and getting out in the market testing the waters with prototypes and so there'll be a lot of different machines and I'm still waiting for somebody to announce a file fact sized one that just has a smaller screen in this portable one to tuck under your arm you mean like the Rayfollower but that doesn't run Pinpoint I mean I'm looking for one that that's a Pinpoint compatible machine sort of thing from whose perspective don't ignore the horizontal think about your whole office operations think about how people do their work and making their life easier even though it sounds trivial I gave a speech recently and I talked about all kinds of stuff and then the question came up which is an excellent question well what about business redesign and what about all the really strategic important stuff what you said is really nice and transparency is nice but it doesn't I don't hear the payoff or anything and my response is that the whole business redesign thing is sort of a different speech and it's another it's it's popping up a layer from what I'm talking about hopefully implementing some of the things I'm talking about in CIE gives you the infrastructure so that when you need to flex you have the capability to flex think you know object oriented network operating systems where you make a check you make a change and it's reflected in many places where directories are consolidated instead of having all them separated all that kind of stuff it would be really nice if the economy turned around just a little bit if you can drum up a few sales that'd be great part of what might happen is either user disenchantment because of poor interface design or poor application design general perceptions that these things aren't able to do or capable of doing what they're promising to do and part of that depends on how the system is first presented to somebody I mean the first time you see one is really important it tends to color your perceptions of the technology and what it's capable of doing and how it might help you in the future so there could be a lot of people that just get that wrong get it presented wrong understand it wrong and then and then decide to ignore it because honestly when I talk about the federal hybrid projects phase and people doing an OSI plan or an SAA plan or whatever that's all I mean they're running like crazy and that's all they have in their power to do is to finally decide that that's what they're going to do and then execute it so they don't have a lot of spare cycles to look around and say oh pen computing how does that fold in and that's where we try to bring some stuff in front of them that's short relevant maybe opens up a couple of new ideas and if it's not relevant throw it away or give it to somebody else or something but just try to kick that in for them in terms of right now the pen based stuff that's out there is in the traditional $3,000 but I'm seeing a lot of excitement around the 95LX you know at a price point of file effects you know not much more well maybe a leather bound golden scribe yeah but it does a lot more too but the point being that you know for $500 you can get into something you can carry around as opposed to $5,000 $5,000 is a loan and you can save up over all of it and just decide the bond and is that going to happen with pens any time soon and you might even be able to lease some of these things the way you lease a cellular phone which is again a different way of getting into the price point there's sort of two divergent ways of looking at how portable machines are going to evolve one of them is I think it's sort of represented by the momentum machine which is we're going to build this machine and it's going to do pretty much everything you want and you're going to carry it with you and in this haul you're going to have it with you and if you go on a trip you're going to carry it with you and if you go back to your desk you're going to set it back up and use it one machine and it'll have connections back to the network and there'll be a file server somewhere else that does more intense processing and so forth but this is your do everything machine another view is sort of espoused by Xerox the idea of ubiquitous computing where there'll be scalable machines of lots of different sizes and there might be one that's the size of my messages and receive messages so if I get an email message it'll show it to me I can maybe respond to it it probably doesn't even do handprint recognition because it might not have the horsepower early on but it has a wireless link or maybe it doesn't or maybe it has an infrared link and when I set it on my desk it goes ah I'm at the desk and it uploads everything it just it just heard and learned and then you have a series sort of a staggered architecture where there might be one that's real cheap and I buy that as my entry platform and then I have my old desktop keyboard machine and maybe I plug an infrared link into the back of that and they talk and maybe then I buy something that's in the middle when I move to that stage but you can start seeing sort of a family of products coming out with different kinds of capabilities I think that's a good way of looking at it what's difficult about that is communicating to the end user what this particular little platform or this particular one is and is not capable of doing because they'll all look like the omniscient computer I mean there'll be a flat screen that just sort of smiles at you I haven't heard about although this wouldn't suggest what kind of lobbying effort is there to get a space program so to speak a government throwing a lot of jobs and everyone else to add to this that for instance right here what is the right thing I think it's like the Glass Bead game there's an idea of linking all kinds of knowledge together here you've got the perfect environment if you could sell that concept as the new space program spin-off technology also the resulting system might look nothing like what you want conversely there's a real danger that if you centralize it, thanks for coming there's a real danger that if you centralize control and you do a DARPA style funding or you do a great society program that you actually stultify the kinds of mechanisms going on Japanese have tax write-offs if you want to build an intelligent building if you want to embed networks and controllers you get a write-off you're having a new agenda my own feeling is that there's no clear technology policy and that there's no clear vision behind from which one could even formulate a technology policy what you wind up having is spots like NREN where somebody really pushed hard and a whole bunch of other smart people got together and said this is going to go through, we don't care what happens we need this and it is happening it is starting to go through but there's so many intersections here NREN is one thing and the internet is one thing it's when you start to bridge some of the gaps and cut across different product categories that people haven't really done much thinking on those frontiers so there aren't a lot of groups where you have people collecting and discussing these issues and what kind of standards to push for and so forth luckily also the FCC is being very lenient these days and is giving a whole lot of leeway they just allowed the R-Box to do video dial tone they're opening up data PCS and hopefully they'll hold administrative video conference next year there'll be some more international agreements on that that's a really complicated area too but at least there's some leniency in that market but we're truly a laissez-faire system right now because of the Republican administration so there's it's not like France where you have the great projects there aren't any great projects here particularly not on the domestic agenda that was political one more question and then I'll wrap up please question was what's the future of video text and other kinds of value-added info let me focus just on the value-added info side and not use the word video text because it's sort of loaded and it's gone through different migrations and personally I'm not crazy about things like Prodigy and so forth themselves I think that the information content that's back there is unbelievably important I think also we tend to think of these devices, these data devices for instance as communicating us to other databases and we tend not to think enough about them as communicating us to other people and other groups and facilitating just communication between groups of people back to another human however there's a whole lot of opportunity on the data side for new uses of information in some cases in companies where they don't even realize they have information to give yet and it's not just the DMBs and it's not just the you know whoever runs the Reuters and the Newswire and all that kind of stuff there's actually a whole lot more stuff out there we just have to mine it right now as soon as an asset sort of thing there's going to be a small and growing economy of people that either decide they want to publish on CD-ROM or they want to publish online or they want to do a data broadcast service there was a recent announcement between McCaw and Oracle to do data broadcasting of different sorts but it's going to take frankly it's going to take 10 years for us to understand for us to see a big shift in how that information is used and what kind of information is provided because that's another big mental shift and a paradigm shift I think thank you very much for coming any other questions come on up appreciate your time