Remarks by Bill Gates, Chairman, Microsoft Corporation
May 14, 2008
KEVIN TURNER: Next we've got certainly our -- it's a pleasure to be able to introduce our chairman and founder, Bill Gates. He's going to come out and talk to you about continual innovation, as well as business productivity and how innovation complements that. So, please give a warm welcome to Bill Gates. (Applause.)
BILL GATES: Well, good morning. I get to do this every year, and share my optimism about how innovation will change the rules. And, of course, if you take any long period of time, the innovation in information technology really has been the biggest change agent out there. It is the thing that makes globalization work. It's the thing that allows people to design products, to make decisions in a very different way.
Microsoft Chairman Bill Gates demonstrates the Microsoft Touch Wall – a new multi-touch user interface technology from Microsoft that creates touch-sensitive interfaces on vertical surfaces – during the Microsoft CEO Summit 2008. Redmond, Wash. May 14, 2008.
This is all built up around the Internet, the PC, the kind of databases that businesses are building, but what I want to put forward today is that we are still only at the beginning of really empowering the workers.
If you think of key decisions that these workers make, whether it's promoting somebody or how you price a product, certain design decisions, we still fall very far short of the ideal that when that employee wants to make a decision, their ability to see all the data, to connect up to the other employees, have it be completely up to date, easy to dive in, get the details, that they have everything they need to make the best decision.
But I also think we're on a path where that's going to get solved. I think the leading companies in some areas now are providing exactly that. So, over the next five years, as this spreads out, it's going to make a huge difference.
Economists are still trying to understand why in the 1990s productivity grew so much, and there's really only one answer, and that is the improved use of IT. And yet that was only a partial use compared to what we're going to have as we get this broad adoption.
Now, we get to be more ambitious in our use of technology year by year. That's partly because this miracle of chip improvement has not slowed down. Intel, of course, is great at driving that forward, and we expect at least another decade of that exponential improvement.
The bandwidth of the networks likewise has gone up very dramatically, and so that as a cost has not held things back at all.
The cost of storage now makes it so that talking about even the world's largest database, taking every one of your customers and every click they ever made on your Web site, having that stored on a server system that costs less than $100,000, and being able to data mine it and see some fairly subtle patterns of what happens to different types of customers, that's all within reach. The hardware cost is not holding us back.
The idea that you take all your training sessions in your company and put those videos online and make them available to your employees to access wherever they are in the world, anytime day or night, that's a trivial cost, and yet gives them immense flexibility, including even having interactive portions that test their knowledge there.
And so one thing we need to do is open our minds up and realize that things that would have been impractical now are incredibly practical; really say to the IT department, okay, I want all that information, and it shouldn't be a budget busting thing to make sure it's available.
The software itself is advancing a lot, and one dimension of that that I'm going to emphasize today is that taking this information and putting it in the hands of end users in a way that they can pick how they view it, so they don't have to call up IT and say, no, I'd like to do the currency this way or the time periods this way or handle some special product bundle this way; they should be able in a few clicks to not only do it but to find that easy.
So, direct empowerment of those end users, making it so that they can pick and set up a Web site, they can pick their colleagues they want to collaborate with, and it doesn't require the time delay or the cost of IT involvement to get those things done.
This is what spreadsheets started a long, long time ago, but the spreadsheet had limitations. It wasn't really connected to the corporate database, it really couldn't hold the rich amount of information, it really didn't have the ways to navigate and visualize. So, it's only now that that type of software is becoming available.
The actual computers themselves are, of course, getting smaller, lighter, faster. Even the gap between what's a portable PC and what's a high-end phone, there's almost no gap there at all. The readability, the size of the screen, the speed of that performance means that we can have some great applications.
Things like 3D rendering would have been something only for a few workstations, specially high-priced computers. Now that's available on virtually all of these devices, and the software people are just catching up to realize, yes, let's put this kind of navigation into the hands of our end users.
Now, when we think about information that empowers people, the most basic thing is just to have electronic mail. That is a wonderful tool. Having it be super reliable, having the directory be up to date, having good group names, having it so people on the road can get it, people on their phones can get it, that I think most corporations have focused on. You should have very tough metrics to say that that's a mission critical system and should basically never be down.
But that form of communication is ineffective when you get to decisions that involve groups. Simply using e-mail and having attachments and long copy lists is a very poor way to get things done.
And, of course, the answer to that is to have Web sites, have URLs that people navigate out to, and make it easy for people to build those things.
Now, what happened when this started exploding now about eight years ago was that there was unbelievable fragmentation. People had all sorts of terms like document management, portal, search site, workflow site, business intelligence site. And so if you were inside a corporation trying to go to these various sites, you'd have a completely different user interface, a completely different way of searching. If you wanted to take data from one of those sites and build your own site, say a departmental or project-focused site, there was no way to just point to it and have that come across. In fact, the cacophony and the amount of money spent and the duplication on all these different sites was pretty phenomenal.
It echoed in some ways what had happened on the desktop itself in terms of document creation where different employees had had different tools and modules and different formats, and that all got resolved when a few key products, particularly Microsoft Office, came along and became this high volume, low priced thing that literally was on every desktop. So, if you were exchanging a budget, it was in an Excel format. If you exchanged a presentation, it was in a PowerPoint format. That same thing needed to happen in terms of these Web sites, but what was it that would let people do what they wanted?
The range of these sites is quite broad, and that's what I'm trying to capture in this slide is that you have a few sites -- I'll call them Web portals here -- that are very few in number, and they are the definitive site say for your financial reporting data or the portal that your customers come in to do transactions on. Those have to be very high performance, they're very high volume.
Then you move down and you have something that's more numerous, but less volume oriented, things like your human resource site or your purchasing site. Those are still company enterprise level, probably driven by central decision-making, but far more numerous.
Then you move down to the department, ad hoc teams, and all the way down to a site for an individual. This is the business equivalent of the Facebook site. This is one where the information is kept private inside the company, it connects up to the org chart, you can have business documents on there that get indexed.
And the point here is that you need a standard way of having every one of these sites, but they're a continuum. Whenever you click to one of them, you know how to search. If you see a piece of data and you want to drill down on that data and have more detail, you know how to do that. If you want to take information and get it down onto your computer to work using the Office tools, you know how to do that.
And so today I would say about a quarter of Microsoft customers actually have this in place, and I'd say about three-quarters have a plan, with varying degrees of length in terms of how they get it together, to take and have this pervasive approach that everyone just knows there's a SharePoint template when you want to create a site, and it's very low overhead in terms of doing this.
Sometimes when people look at Microsoft from a pure business point of view and they say, wow, over the last three years we've had a lot of success, our profits have grown over 70 percent, they think, well, is that about these highly visible things like advertising markets or various consumer things, and the answer is, no, all those opportunities in terms of profit growth are in front of us, not part of the incredible success story of these last three years. The center of the story these last three years really has been adoption of software in business, having these tools higher volume so that people can get more productivity.
What it's meant is that the actual number of software packages inside a company -- workflow, document management, things like that -- is actually a lot less. So, the idea that you could get used to it, even bring in somebody from another company or somebody coming out of college, and they'd already have this familiarity, that's moving towards the same level that exists with Office, so whatever the scenario is, you can go and use this tool.
I thought I'd take a couple examples of how in my work activity these things are used and they help me get things done in a way that if you forced me to go back and do it the way it was done five years ago, I would be incredibly upset, there's no way I'd go back.
The first one I'd pick is the personnel process. That includes the way that you survey groups to understand what their sentiment is about their schedule or their management or any of their attitudes, and everything you go through in the regular review process.
Now, the review process historically at Microsoft was incredibly paper intensive. We would get people to write things up, you know, we want to have your peers comment on your work, we want to have your direct reports comment on your work. We had a lot of constraints in terms of the average raise, the number of promotions, various things like that. So, people were printing out Excel spreadsheets and mailing them around, and it was an incredibly complex process, and you would spend a lot of time on the mechanics.
Well, today, we're finally at the point where it is a paper-less process. The idea of the employee saying, hey, I want you to hear feedback from these people, they select the names of who they think should give input on their work, the manager picks other people, sometimes the same set, sometimes different, who he's asking to get input on. The system tracks who's already done that. If you're somebody giving feedback, you can easily come in at any day, just click and see, okay, who have you finished your feedback on, who have you not. When you're assigning people's rankings and new salary data, the system shows you right there, are you staying within the guidelines that have been provided for the various finite resources that are there.
It's actually become a process where the focus on the mechanisms goes away, and the focus moves to what it should be, which is the actual question of did this employee achieve what they should.
The fact that the commitments go online and then six months later you're looking exactly at those commitments and have people giving feedback on them, it's a really phenomenal thing.
The ability to do these digital surveys and get quick feedback from groups on how they feel about things, it's actually our best way of getting the pulse of a business. So, this is phenomenal, and it's just taking that kind of SharePoint software and doing a site for human resource activity.
Another one that just becomes taken for granted is the idea of scorecarding. This is very straightforward also. It's the idea that when you come into work you should be able to bring up a screen that shows in sort of green light, yellow light, read light form the objectives that you've signed up to. For any one of those if you want more information you click on it and drill down to any level of detail, getting down to a specific customer, if that's what's involved, to a specific set of tasks that are in a project management format, and that are behind schedule.
I actually played around with the numbers here, so don't think those are the actual numbers. But this actually is exactly a typical scorecard other than changing some of the data. So, the idea of having a chart, drilling in -- and we have these by division, by group.
It's been particularly good for our global nature, where we want to be able to compare results. Whenever I'm visiting a country, I just say, hey, let me see their scorecard. I call it right up and I can drill in on the red items and then I can send mail off to Kevin saying, hey, should I give them a hard time about this, what is the story here, is it something headquarters is doing that I should understand a little bit better? It creates a common dialogue, we're all on the same page about what's going on.
And actually building these things and getting them together wasn't that hard. Now it's something that you don't have to go off to IT if you want to create one of these metrics or change one. If you want to do a special page that takes a subset of them, just copying that over is a very straightforward thing.
So, it's just a form of empowerment that is very basic, and again not driven by paperwork, it's the data that's up to date, and because it's not printed out, this navigation capability and putting annotations on it and passing it around for discussion, including on other sites, is something that comes very, very easily.
So, just examples of where the work process has changed. The equivalent of this used to be a complete mess, and we really never agreed on what the metrics were and how to measure them. We'd come to a meeting at the end of six months, and somebody would say, you know, I met this goal, somebody else would say, no, the way I looked at it you didn't. That is mechanism, and that's a distraction from the real content that you want to get into.
So, there are many of these things. The idea that the Web sites are common and easy to use, that is a building block. And as I've said, we've made real progress there, and that is almost becoming just commonsense everywhere like Office did.
There are many other of these things that are perhaps in an earlier stage. The social networking within enterprise where everybody has one of those Web sites that they talk about, their interests and their background; getting that to be widespread is very important because it's one of those things, if only 30 percent of the people do it, then it's worth almost nothing -- in fact, it's basically a waste of time to have done it. You really need to get full participation so that if anybody is trying to say who's got expertise in this, who's got background in it, they go in and use this powerful employee search capability, and then create sites, create discussion boards about those things.
The communications people, the desktop phone, that's being changed to be a software-driven device. The idea that when you call somebody up, if they're not there, if you have the right relationship with them, the software sees that and they've kind of marked on their schedule time that you can just schedule for when the follow-up call is going to take place, that should be just commonsense.
Getting rid of phone numbers, the boundary between the mobile space and that desktop space, a lot of the cost that exists now in that telephony world, and it's lack of integration with the Internet world where setting up a screen sharing thing or the quality of teleconferencing is not nearly as good as it will be.
So, communication is at an earlier stage in terms of the integration, social computing is at a fairly early stage in terms of mapping it into the business world, taking the things that are good about it as a consumer project but still bringing the special requirements of connecting the data and the confidentiality of data into the business world; and then another level of visualization. We have good visualization today; we think we can do more there. So, as far as we've come on this, we think that there's still a long way to go.
There is even today a huge range in terms of how well companies do these things. Empowerment is a fairly basic thing. The big cost actually is one that every company pays. No matter how well they use it, they are buying an Internet network, they're buying state of the art hardware, and so this is just the thing that determines the value you get out of it, which is the software processes that you put on top of it. Amazingly, doing a lot of this stuff well actually gets the IT department out of the loop so they can go and focus in on other things.
I thought I'd ask Jeff Teper, who is the person who runs our SharePoint business, to come up and interactively show you some examples of how this is working, particularly how you navigate between the different pieces so that the whole is more than the sum of the parts. Welcome, Jeff. (Applause.)
Demo: Empowering Employees: Microsoft SharePoint
JEFF TEPER: Thanks, Bill. Thanks.
Well, I thought I'd show you some of these tools and how they can be used to drive better understanding your business and improved business performance.
A lot of what I'll show you is previously places where people would have hit silos, silos within the organization or silos within the technologies. And as Bill said, we're trying to break those silos down by taking the advances we're doing in business intelligence and unified communications and social networking and enterprise search, and deliver them to the end user in an intuitive, integrated experience, just as we did with the Office desktop. If you look back sort of 20 years ago to the advent of the graphical user interface, that was really the key innovation that led the vast majority of people in business to use these kinds of tools. We're trying to sort of replicate that experience with the next generation.
So, I'll start my day as an executive in this fictitious company called Littware. You can see my personal Web site that Bill mentioned is up and running. One of the questions that was in the introduction, somebody brought up people wanting to spend more time on Facebook than the golf course. Well, this is the kind of thing where we're taking the tools that have been popular in the Internet and applying them to an enterprise experience, and it's actually a great way to track and excite new employees to your organization.
This SharePoint technology Bill mentioned is really I think of it as the smart fabric that is the way to have Web sites for the employees, the customers, the projects, the scorecards, and tie that all together so you're not constantly hunting around for information in the organization.
So, you can see my personal site has gone out, and it knows -- the software is smart -- it knows I'm part of the marketing team, it knows I'm part of the finance team, it knows what metrics are relevant to me, it knows who I collaborate with. The software is actually integrated with our e-mail system, and understands, if you opt into it, who you communicate with on a regular basis, and you can actually track what they're doing. So, I can see that Holly Holt has added competitive research as an area of focus for her. The software is smart and does all that based on sort of the organization, usage, and things that are targeted to me based on my role.
Since I'm in finance, one of the things that's targeted to me is a sales dashboard that Bill gave an example of the things we do internally. Let me go click on that for this company, and you'll see up comes up a view of their business.
The thing I thought I'd highlight that's different than traditional performance management systems is we're not just bringing in the numbers, but the context about what's behind those numbers. So, you not only see sort of dashboards and graphs that you can open up in Excel and interact with, but there are things like report plans and case studies, and even a sales team blog.
Bill mentioned social networking technologies. Blogging is a way that people have gotten familiar with sort of keeping a running tally of what's going on.
But you can see various people in finance, in marketing, and in sales are all collaborating on what's going on behind the numbers in this organization.
Sort of looking into the specifics I can see my sales are declining in the Western region. I've got a big red triangle for that, a big red diamond for that. So, I'm going to go and look at what's going on in a little more depth behind those numbers.
I'm bringing up here what we term a next generation search experience. One of the things you'll sort of notice as I go through this is I'm never actually going to explicitly search. I'm not going to type in a query and hit enter and see what comes back.
The software should be smarter than that. It should sift through all the information in the organization, figure out interesting ways to structure and navigate, and help me browse through it. This is based on an acquisition Microsoft just completed a couple of weeks ago. It's a company called Fast who's the leader in enterprise search. What they've done in this example is we've indexed everything, not just documents and Web sites, but actually our Customer Relationship Management system, our performance metrics, e-mails, and sort of synthesized it all in an integrated experience.
So, you can see it's pre-filtered here into the Western region, but I could go back and pick a different filter for the East, and it automatically sifts through in real time all that content. Let me to back and take a look at the West.
The big innovation I think we'd like to highlight around search is going to the next level of understanding meaning behind all this content that exists in all the systems. It's not a great accomplishment if we can use technology to overload the user with sort of millions of hits. We have to help people sift through it.
So, I'm actually going to sort through our customers -- it's probably a little hard to read in the screen there -- by what we term "sentiment." This is a capability in the Fast technology to actually sift through the content and look at things and figure out are things -- is it a happy day, is it a sad day, are things going well, are things going poorly. It's just one of the many dimensions of which we could sort of organize information.
Sure enough, we've sorted by sentiment, we see in the Western region Contoso Aerospace is unhappy. And I'll go click on them and bring up on the right side a little profile.
And this graph is actually composed by the search engine going through all this content and determining that the customer sentiment has sort of gone through the floor, that the customer is unhappy. And I can go in and look at the specifics, and sure enough, here there are words like "rejected" and "problems." Rather than have a human being sift through all that stuff, the software knows that's sort of bad news.
So, I'm going to see what is going on, a little bit more depth behind Contoso in this particular issue, and I'll bring up another scorecard. The first one was for the business overall. This one is for this particular customer. And again sort of I'm bringing and synthesizing the collaborative data, the social data around the account team and so forth with this business intelligence data.
It's a lot of metrics to look at. I'm going to go and filter those to the ones that are off target. I can see we've got a return rate that's declined of 9 percent. So, I'm going to look at the comments. Sure enough, I can see we've got a return rate due to bad shipments to this customer we've got to resolve.
So, I'm combining the human element with the data behind it, and I'm going to go in and take a look at that.
I thought before I drill into the account managers, and I've got that right here, Mark Hassle, I could go give him a call, I could instant message him, there's a whole bunch of ways I can communicate with him.
Sort of one of the things we've done in this smart fabric of SharePoint is any time there's sort of a person or a relationship, the software automatically puts those links in. There's not sort of an army of Web publishers going in and linking manually what's going on in the company. The software should be smart enough to do that.
So, all the ways I can communicate with Mark are up here. Before I go bug him personally, I'm going to go to his personal site. So, just as I had a personal site, Mark, has gone one, and I can go up and down the org chart. I can see his responsibilities and interests, I can look at his documents. All this is filtered, so only the things that I have the rights as an external user to see about Mark I can.
I can see our colleagues. So, maybe I don't know Mark. We've got a big sales force. But the software is smart enough to figure out people that Mark knows, that I know, and can suggest to me, hey, if I want to find out if Mark is going to do a good job closing this account. Pat Coleman, somebody we both know, I trust Pat, and I can go call her up and check on Mark's track record.
But I know Mark, he's doing a good job, and I'm going to take a look at what he's done to communicate with the customer.
These are workspaces that not only are inside your organization but we've put out on the Internet in a secure way to collaborate more effectively with our customers. So, this is the Contoso external workspace.
Microsoft actually has about 15,000 of these sites out on the Internet that we use with our customers and partners to collaborate more efficiently with them.
If you can imagine instead of everybody traveling to every meeting or being on conference calls or even e-mail distributions, imagine if Littware was a giant company and Contoso was a giant company, for everybody to be on every e-mail is just overloaded. So, here is a single place we can go to and find out what's going on.
Sure enough, I can see that we've set up a wiki -- so again sort of an area where we've taken Internet technology and applied it to the enterprise. Folks are familiar with things like Wikipedia. I've applied a wiki to our relationship with Contoso for shared meeting notes, so people between the customer and our organization can collaborate and share information on what's going on.
Sure enough, in this wiki we've got a rolling status of the meeting notes, and I can go in and scroll through this and say, yep, we're actually on top of this one. Littware is going to replace the parts next week, and that we're not only going to do that, we're going to provide engineers onsite to address the issue so that we can turn that sort of low sentiment customer into happy sentiment customer and drive sales.
So, now I think I know what's going on for our business overall and this individual customer relationship, and I'm going to reach out to Mark to make sure that -- so we're going to close this down.
Again, I don't have to go switch to a separate tool, look in the address book, search, find Mark. The software is smart enough to know, again, Mark is the account manager for Littware, and I can bring up the unified communications gumdrop that we sort of refer to it sometimes that gives me all the ways I can talk to Mark, and I'll go just reply with an instant message and say, "Are we okay with Contoso?" And I should hear back from the magic mark behind the stage, and he types a lot faster than I do, but sure enough the CEO of Contoso is going to be at the CEO Summit, which is great, convenient. I can go talk to him and reassure him that we're on his problem, and we're going to resolve it very quickly.
So, that's just a quick tour of how again the software can be smart and break down the barriers of these sort of silos of information in the organization for a unified experience. Thanks very much. (Applause.)
BILL GATES: So, information empowerment is getting stronger and stronger.
Now I thought I'd focus on something that is more of a discontinuity, and something that I think really will surprise people, although it's very complementary; and that's the way we interact with all these devices. It is overwhelmingly today the keyboard and the mouse. Even on your cell phone, that little keyboard, you're essentially typing like you would have many, many years ago.
Natural User Interface: Touch-Based Technologies
Well, we're beginning to see a change to that into what I call natural user interface. Last year here I showed you a prototype of Microsoft Surface, and that was the flat table that had a Windows machine with a little camera looking up, and so whenever I would touch and change things, the software would understand that and respond. So, that is a very different thing.
Natural user interface covers a wide range. It's speech, talking into your phone and just saying who you want to call or the stock price you want, the information. That's coming along, and it will be very mainstream.
This idea of touch on the horizontal surface is becoming a real thing as well. In fact, a few months ago, we sent out our first units of Microsoft Surface. One of the first customers is AT&T for use in their phone stores. And what happens is somebody who's got a phone comes in, puts it down on the table; the software recognizes it, suggests what calling plans they might use, suggests another phone model they might use. When they select that, it shows them comparison, what features it has that that existing phone does not.
And the reception has been phenomenal. This idea that you just sit there and interact, touch, you don't have to learn anything, that naturalness really draws people in. So, it's been a strong success so far, and that form factor is going to get cheaper and smaller.
We're also going to have that in a vertical form. So, think about the whiteboard in your office becoming intelligent.
Our view is that all the surfaces, horizontal surfaces, vertical surfaces, will eventually have an inexpensive screen display capability, and software that sees what you're doing there, so it's completely interactive.
When I say everywhere, I mean the individual's office, I mean the home, the living room, all of those things. The cost of the hardware is not that great, and the quality of the software is improving substantially.
The early uses of this, besides Surface, include things like the touch on some of the phones, including Apple's, it includes the pen on tablet computers that are very popular in verticals like medicine, and we expect to catch on with students who want to take notes or people who sit in meetings.
I've also shown here the RoundTable, which is the videoconferencing thing that takes the entire view, the 360-degree view of everybody in the room by using multiple cameras, and creates that teleconferencing interaction that's far, far better. That's a new type of interaction.
Little thin, tablet-like computers that have both the pen capability and that finger touch are also going to be coming along, and you'll be able to switch back and forth between those.
Well, I always like to show something that's new, because that's kind of risky and exciting, and so what I thought I'd show is this future whiteboard, the intelligent whiteboard. So far, this display here has just had the nice meeting logo, but, in fact, it's running a new piece of software. It's got some scanning cameras down here at the bottom, so whenever I go up to it and say just touch it, the software will notice that, theoretically. (Laughter.) There we go. Makes me a little worried but --
So, in fact, this piece of software has all sorts of information behind it. And typically when we think about information now it's very linear. A PowerPoint presentation is slide by slide, and it's something happens during the presentation, you want to skip around, that's hard to do.
Here we've got things laid out in this nice two-dimensional form. So, I can zoom in on anything, I can move around. I've got different types of information that I'm using here. And it's all just easy to navigate to, because at any time I can look at different things.
Here I can point, I've got an actual document. This is a Word document. So, I can go and step through this page by page, see different things. If something is interesting, I can blow that up or I can just go back to where I was.
I can have a PowerPoint presentation. There, too, it's easy to step through different things.
But, of course, I don't have to do this in this one-dimensional form, I can have things laid out on this surface.
I could have my sales data. I talked about being able to look by region and product. Well, that kind of pivot navigation is also kind of a natural thing to be able to do. Any type of diagram or flowchart, you know, you can go in and look at these things in a very detailed way.
Of course, this document, I could take it and edit it on my PC, as well as looking at it here.
One nice thing about this form is that if multiple of us are working together, let's say we've got this flowchart here, and we're talking about the different steps, we can all walk up and interact with the information that's here.
It's fully multimedia. So, if I go across I think somewhere here I've got this is actually a video. So, as soon as I play that, we're looking at, say, our new ad, our promotion thing talking about that. It happens to be for the touch capability. So, all the different types of information.
I can even take and say if I want to ink on here, I just touch that, and say I want to circle this, say that's something important, and then when I go back to the presentation mode, as I zoom in and out, that's there just exactly like you'd expect.
I can also let's say we have a bunch of slides and we're working together, figuring out how we want to order those in order to make the presentation to a customer, I can go in and anything I point to, it will let me just move that and organize these things. So, whether that's a piece of work information that I want to do something with or a photo album, something like that, as soon as I've made those changes, those are just there.
So, you can see how you could train somebody to use this pretty quickly. They could work with a group, work with the information, and it's kind of a natural extension of Office. We can take spreadsheet, word processing, presentation data, and get it into here.
And this kind of whiteboard with a little bit of hardware advance over the next couple of years will not be an expensive thing, and that's why we're saying that it will be absolutely pervasive, and people's ability to get at information and not want to see it on paper, want to just have it here where it's so much more interactive, will be dramatically changed.
When I'm done with this, I can find that slide, I just point and it goes back, and it's just like it was when I got started.
So, that's in our R&D labs right now, but already the Office group is thinking about, okay, what do they do in the next version that has built-in capabilities for this. In fact, the Windows group is also building in down at the operating system level, so any piece of Windows software will be able to have this type of interaction.
When you put it on a horizontal surface it's really the same, except there you can also put objects on the surface and the camera software can actually do recognition not just of your hand but of, say, a business card or a sheet of paper like an optical scan, so you add even a few additional things to that as well.
And we're using the same software we used for pen input, and so when the finger is not good enough, when it's not high enough resolution, if you wanted to write a little note here, if it's the right hardware basis, then you get that pen interaction as well; so an opportunity to go back and rethink a lot of how people work with information.
Well, now let me zoom out from that and take it in a broader context, and talk about how a company drives innovation. I think this is a tough challenge for all companies. Because the rate of change is higher than ever, because the mix of skills of people who see the opportunities for change are often different than the people who understand the customer needs, how do you bring that together? Particularly as a company gets large, you're in a dynamic industry, which certainly IT is, thinking consciously about these processes that allow for continuous innovation are important.
I just wanted to share some of the techniques that have helped us and how we kind of stitch them together to get the best of each one of these things.
Shared Vision: Five Practices
I'm going to touch on five different things that are practices, starting with what I call shared vision. This is one that over the last three years we've put a lot of energy into, is separating out from the actual product development cycles the idea of, okay, where do we want to be in a five or 10-year period, because after all many of the advances it's simply not possible to put into a single release, but if you just have that mentality of, okay, what can I do in the next two years, you don't set that big, exciting direction and make sure you're making progress towards it. Then again, you don't want to have people expecting that you can do that five years of work in one release, and so you need different types of personalities involved.
So, for the shared vision we've created a thing we called quests where we divided our types of customers down, and we got the best thinkers on these things, both the very practical people who are with the customers, the engineers who write the code, and the researchers who may be more unbound in terms of their timeframe and imagination, and put them together. And, in fact, we allowed a lot of fairly young junior people to participate in those brainstorming sessions.
We get together yearly and break out into different groups. You don't have to go to just groups that relate to the business you work in. In fact, if you go into a group and you're very passionate, they may try and recruit you into their group to try and help them achieve their quest.
So, these are actually written down, shared pretty broadly in the company, but you can see the five different groups here that we talk about, and all of them sit on an overall platform, the idea of what computing will be like, what software development will be like, that feeds into each one of these things.
The level of broad participation we've gotten now really in our third year in driving this forward has been quite fantastic. We had some early learning pains in terms of did people see this as central, did they see it as something we were really going to stick with. Year by year, it's surprising, not that much changes in these visions. Some things like social networking you could see is a little bit stronger now than it would have been a few years ago. Some of the progress in the hardware that enables natural interface is a little stronger. But there's a lot of continuity, because after all many of these are problems that are going to take about a decade to get done. And we can always checkpoint to see is there something missing in these, and also checkpoint what progress we've made towards that vision.
Using Customer Feedback
Another key thing is the feedback we get from customers. As you get large, this can become very diluted. I mean, the idea that somebody is out in the field hearing something, how does that get back not only to headquarters but actually into that product planning process? Here we can be helped a lot by digital processes, by taking the feedback, seeing where it's recurring, let people navigate through it.
And, in fact, some of this comes in, in structured form, customers indicating what they like and don't like, but by far the best, in our experience, is what we call the verbatims, where you have actual text of a customer saying, I wish it would do this, I'm upset it doesn't do that. Making sure those verbatims get translated into English, get up and to the very product person, that's a process that you've constantly got to reinforce how important that is. Even once you do the digital enablement, that won't necessarily make sure that it happens the right way.
There are some additional feedback things in terms of being able to watch the software as it works. We see when somebody boots up Windows and it's looking for a driver they don't find. We see which of the error messages are coming up on a regular basis. We see if there's any type of hang situation; and that broad data is fascinating. We can even look and go to a customer and say, look, the way you're managing your Exchange network is causing you to have more delays or less reliability than is typical; let's dig into that, is it the hardware, the way the software is configured, what is that, and have an understanding, and then see if it's a change at your level, or if there is something in the product itself that needs to change.
So, I think these are the two key systems, the broad vision out in front, and then the feedback we get.
We also think it's very important to set our sites high though that we have pure research activity, unscheduled, unknown what will come out of it, very risky things that are often even more than 10 years out into the future.
This has become a big asset for us, not just because we've brought bright people into it, and been able to build centers in many place around the world. We're literally on the Cambridge University campus; that's our European center. We're across the street from Tsinghua University in China. We're closely associated with India Institute of Technology in India. We just opened a new research site back in Boston to draw on the great universities that are back there. And, of course, we have one in Silicon Valley and the largest group of these researchers here at headquarters.
The cultural divide you typically get between researchers and product people we've had to do a lot of things to make sure we bridge that. I've talked publicly about a thing we have called TechFest where the researchers take booths and show off their stuff. It's like a county fair sort of, and each one is trying to say, hey, my research is the coolest, come by, and don't just get excited about it, actually put it into your product and ship it, and that's a positive dynamic.
The connection out to the universities is a big part of that as well, because if you think about creative ideas, the manpower out there is greater than any company. So, seeing that at an early stage, having very positive relationships with them so we can see, okay, how could we license that, how could we put that into our products, how could we take it further, that's made a very big difference. So, this is a very strong leg of bringing in new ideas and understanding what's going on.
Another element is that we get new blood and new ideas through acquisitions. There will often be an area that a company is dedicated to. Take speech recognition; this company TellMe was a startup down in the Valley, it had had ups and downs. The speech market, although in the long run is super important, hadn't developed that quickly. So, we were able to go into them and say, we can really up the R&D on this, we can integrate it into our products, and it's kind of a textbook case of their energy and understanding where they have been taking all these information calls, the 411 calls that people like AT&T and Verizon get, and teaching the computer to get better and better at recognizing that speech. We brought that in and put that onto the mobile phone, put that into a broad search context.
There are a number of them up here that have been good for us.
Executing on acquisitions is a special skill. You can look three years later and say, were the really good people who came with that acquisition, are they still there or not? Has the product group kind of embraced them, and yet not taken away what was special about them? We've got a report card that we've gone through for ourselves, taking our predictions when we made the acquisition, and we don't have anywhere near a perfect record, but the successes far outweigh the cost of the ones that don't work well. This is an important thing, and Steve will actually quiz a group that's not getting involved in any acquisitions and say, are you being open-minded enough, are you looking out there, do you know the different things going on, to make sure that it's at least a tactic that they've got clearly in mind.
A final approach is one that has really become important over the years is what we call Think Week. Traditionally this was more focused on my going off for a week and people sending me a lot of ideas, generally 300 or so, and then I would take and provide feedback on those.
Now what we've done is taken the broad technical community, the top 50 engineering thinkers in the company, and they all participate. That's possible because it's gone from being a paper-based process where I would literally take boxes out to a beach place and sit there for a week reading them day and night and scribbling on them, to putting it entirely online. So, it's all just a SharePoint site. We start out the period by taking 20 or 30 papers that are relevant to a particular expert and saying, hey, we want you to comment on these, and then at any time we can look, have people comment on the ones they refined, that they commented on more. We can look and see which papers are getting good ratings, recommend to a broader set of people they ought to look at those, recommend to the product group involved that they really ought to think that through. So, we've institutionalized it as kind of a grassroots process, and this is a way that somebody who's even just a year or two into the company and has ideas that may or may not relate to the group they're in can write something up and there's great examples which we highlight so that it encourages more and more of it where those things come in and make a big difference.
So, it's now an institution that has a regular twice a year rhythm, and is just part of the innovation cycle. In no way does it substitute for those other things, but in terms of some of the more far out things that you need to grasp and bet on five to 10 years before they become mainstream, it is a valuable element.
Changes Over the Next 10 Years
So, what kinds of things will change in the next 10 years? I've said that I think the Office environment in terms of how that whiteboard works, and the amount of information there, that's one that will happen a lot faster than people think.
Some scenarios really have been digitized: photos and music, everybody knows that, it's taken for granted. Bill paying, we're more than halfway there. Communications, getting rid of phone numbers, unifying the different ways you interact, definitely over the next five years that PBX will go away, it will be a rich integrated software experience.
The TV moving onto the Internet, the United States is actually one of the leaders in this, because the telephone companies, in order to try and get ahead of what the cable providers have offered, are using the Internet to deliver the video. So, the content providers are now looking at the interactivity, targeted ads, things like news where you can skip over things that aren't of interest or get more data about what you want, and taking those videos that aren't popular enough for broadcast but because you might be interested in them, understanding that and putting it into your TV Guide.
So, TV is changing, reading, note-taking, those will change. The datacenter in terms of being able to delegate tasks off to a mega datacenter that Microsoft and others will build, those things will change as well.
So, year by year it doesn't look like these things change much, but when you take that accumulation, you get to a point where the scenario is utterly different, and I think for every one of your businesses there are some interesting opportunities to get out in front and take advantage of these things.
Thank you. (Applause.)