Gregory Farrington: Instead of having questions now, I thought
we'd just proceed to our next speaker, Steve Heibein, who is from
Silicon Graphics. Silicon Graphics made a major commitment
to being an industry which leads the re-invention of education,
and Steve Heibein has been responsible for their penetration into
markets involving the Internet, telemedicine, distanced learning,
and video servers. Now he notes that he has been with Silicon
Graphics while Silicon Graphics has grown from a 20 million dollar
company to a four billion dollar company. If that's his real
secret, he'll have lots of people looking for his services by
the end of this meeting. So it's a pleasure to have Steve Heibein.
Heibein: Thank you very much.
What I want to do is move right ahead and start talking about some
of the technology. And that's really what I am. I'm a technologist.
And as Dr. Lucky had mentioned, things are changing. In fact,
people are getting different careers. The things that you learned
in college, you don't really remember those items. You remember
the experience. I've gone through several different careers in
that sense. I was an architect. I did solar design for awhile.
I wrote some books on energy conservation. Now I'm a computer
scientist. But what's happened is technology is really what's
been pushing us along, and what's been pushing me.
What I want to do is talk about some of the technology trends
in a couple areas, specifically, when I start looking at the graphics,
what's happening with graphics. And a lot of people recognize
Silicon Graphics because of our graphics background, but what's
real important and the vision that was had by our founder, Jim
Clark, a long time ago has now happened. It's become prolific.
You now have 3-D. You can have it on the desktop. And in fact,
if you look at the things you have in a Nintendo today, the type
of graphic that's in there is what we had on a 150 thousand dollar
computer just three years ago, and I'll get more into that.
What's going on with CPU's is also impressive, and I want to
touch on those items and scalability, etc. [Aside: Can I move
forward here?] If you sort of look at the trends in graphics,
and I think what's kind of key for the educators to understand
is what you can now do with it beyond-What is that next generation?
What's the bleeding edge? What are the things that people are
doing to try and get better insight? And I think that the real
key that people always say, "Well, virtual reality."
Well virtual reality is a buzz word. A lot of us would almost
like it to go away, because what it conjures up is all these people
with headgear and etc. But what we're trying to do is now use
this so that people start having joint experiences.
If you start looking at what's going on at Stanford, they use
an electronic light table that actually has a stereo image. So
you're looking down at a model of a house or a village or something
when you're trying to look at something from an urban planning
sense. Or if you look at the type of graphics we can now do up
top here, the realism. So now you can experience something that
no longer exists: buildings that have since decayed or how it
would have looked like before environmental changes-the ozone
and different things-have happened to the walls. What we're doing
is we're using this in different ways to be able to say, "What
if?" Things that haven't been built yet. On the assembly
line, what would that assembly line look like? Should it be built?
What kind of room do we need? So now we're doing our designs,
playing out our designs on the computer. And having more and
more sophisticated graphics has allowed us to be able to see this
in real time. It's no longer quite as cartoonish as it once was.
You now have the ability to have things running at real time,
to make changes, also interacting with those programs that you're
saying, "O.K.., if this was to occur, this stimulus, what
now happens on this production floor?"
Other things that you're seeing are when people say, "Well,
I want to take large volumes of data-big data. I want to get
an insight from that. I want to understand what's going on."
And I mentioned this to somebody else, what we're trying to do
now is mean time to insight. O.K..? How quickly can you understand
something? How quickly can you mine a nugget of knowledge out
of this information you have? We all have data, and the trouble
with data is it's not enough. Information is not enough. The
insight is where we're going to. You need to take that to the
logical conclusion of being able to say, "O.K.., I want to
see what's happening with this N variables," and see it in
a way that's a little different.
Down the bottom you can see two examples of what a lot of companies
have started doing. It's a concept of a visionarium. So somebody
had mentioned earlier about video-conferencing. What we're seeing
is, for instance, Ford Motors. What they did is they built visionariums
on different parts of the world so that they could have designers
in Detroit and in LA and in Japan and in Milan and in London all
working together. And so what they're all looking at is the same
wall. And what they do is they use a way to rotate and take a
look at the car, so they can say, "O.K.., now that we're
all looking at the same car, notice what happens with the flare
on this hood." And they can say, "Ah, I see that,"
because they have a joint experience, and they're not wearing
goggles, they're not wearing glasses. What they're basically
doing is looking at that joint experience and that joint wall,
and that's what they're using video conferencing for, because
it's insight is what they're trying to get.
One thing that does get saved is not travel but generally it's
the time that's lost going to and from something. So what happens
is when you have an idea you can bring other people together quicker
than trying to schedule everything. And that's probably one of
the more powerful things.
Here's another example where people might have a vehicle and
they want to have that experience of trying this vehicle out.
And there's several different companies-in fact, that's probably
the biggest uses that we've seen, whether it's flight simulators,
auto simulators, or different types of equipment like that. Generally
it's too expensive to have everybody actually go and experience
something like that.
The same thing's happening now with the medical world. Right
now most of it is to help train you and get you prepared for an
operation, but not necessarily to do an operation. That's still
pretty far off there, but if you look at what is going on, right
now you'll see a physician, a surgeon sitting there and manipulating
handles that the handles are part of the robotics that are actually
doing the surgery and doing the cutting and doing the suturing.
What's the difference of doing this in the same room, holding
the handles, or being across the street or across the town or
across the country?
What's been interesting is we're finding that the military has
been very innovative in working with this because they are working
under different circumstances than everybody else, and so what
they'll do is they'll drive a HUM-V out into the field, do the
surgery there, where the surgeon may actually be at a more fixed
walled facility, so to speak.
Now we're seeing innovations coming in a lot of different ways.
I mean this is actually, this part here is part of an ad, but
it shows the Boeing 777. It says, "This is a 777."
And down there it shows, "And this is the box it came in."
You know, this little box down at the bottom. Basically the
triple seven, Boeing designed that whole system without building
a single prototype. In the past when they would build a prototype
they would be off as much as a foot and a half from end to end
on something like this. They didn't build a prototype. They
did the whole thing electronically on a computer. And what was
interesting is when they finally built the first one, with no
prototype, it was off by a half inch. O.K.., that was the smallest
tolerance they had ever run into, and what gave them the ability
to do that is now they could visualize every component of this.
And things that they hadn't thought of before were very powerful.
They could actually go down the cable runs, and they could pretty
much walk down like they were a mouse down there and see what
was happening, what interference they were running into. Before
they had to build a whole prototype, until they ran in and said
"Ah! I got this duct work that interferes with there,"
because all these flat drawings don't help you with that. So
you can see how this can help an aspect of it.
Another thing that was interesting is what's happening with Team
New Zealand when they won the Americas Cup. We had worked with
them on that, and we actually put a supercomputer on their dock
because-remember the ENIAC would fill a room larger than this.
Well, what we were able to do is now carry in, under our arm
basically, a super computer. And every time they ran a race we
were metering and monitoring what was going on, the flows around
the keel and everything else, and took that back, plugged it back
into the super computer, ran the test again, came out for the
next day. We had two keels: one we would change every day.
And we were able to get that mean time between insight, so-that
was before we bought Cray, mind you. So we had beaten Cray.
So Cray had this huge super computer that they couldn't bring
to the dock, and so they had to do all their tests and everything
three months earlier. We were able to do this three hours earlier,
and that made a big difference for us.
One thing that I think is real important is this 10x every three
years. If you basically look at what's going on with technology,
every three years computers are either ten times faster for the
same price or one tenth the cost for the same speed, either way
you want to look at it. Well that makes things happen is now
you go from something like this, a Cray super computer, to now
you have a desk top computer with the same type of capabilities.
Or let's put it this way: one tenth the price for the same capabilities.
You also have a scaling in size. Well what we're doing is all
vendors are basically pushing the technology down to the desk
top. So what you used to have to do in a closed room, we're now
making available for an individual student, for an individual
researcher, for an individual department, where you used to have
to have a whole university to be able to supply all their funds
to be able to have the type of super computing power that you
Sort of to take that even one step further in the logical conclusion,
we've basically taken what we sold as a literally about 150 thousand
dollar 3-D graphics work station, is now available in a Nintendo.
The Nintendo is under 150 dollars. So you go 150 thousand to
150 dollars. It's kind of interesting. And in fact, the two
chips that are inside the Nintendo cost 35 dollars-Quantity One
list for those chips. So if you start looking at what's happening
with this, it's making technology more accessible. You don't
see that quite yet, but you're going to be seeing this in the
coming year with things that are going to be going on with Network
Web TV, for instance, uses some of this same type of technology
that's out there right now. It takes your TV and puts it into
Web Browser. Well, what you're going to find now, taking the
same 35 dollar chip set, merging it with a monitor, a keyboard,
and a mouse, you now have something that you can start proliferating
on everybody's desk top, using the Web technology to be able to
get that information to them-get them access. And then some of
the things that are going on in terms of incompatibility of software,
what happens now is you don't really care what's on the other
side. You're using the Web to get to all these legacy systems,
all these systems that have the information on them, and you're
just putting new interface on it and making that available. I
think that's the real key of what's coming with 10x every three
If you don't recognize this, this is what we call a "rocket
dog". A lot of people have been strapping 600 megahertz
chips on PC Buses, and that's what I like to refer to as a "rocket
dog". The problem is, you can get faster and faster and
faster chips, which is really what's happening out there, but
the trouble over time is you're now running into new band bottlenecks.
Whether it's a Network bottleneck, whether it's the back plane,
whether it's memory access, other things are now occurring. So
new technology has to be evolving to overcome these problems,
because obviously this rocket dog isn't going to make it for the
Now, when you start looking at some of the things that are going
on with connections, currently most everything you're running
into is over here on this Bus Connects. You have one Bus, and
even if you have multiple seat views, you're plugging them all
into one bus, and you're sucking up bandwidth. So as I go, I'm
adding more CPU's, but I'm not necessarily getting much faster.
I'm not getting linear speed up. Well there's other interconnection
technologies. Whether you do a one dimensional ring or whether
you do some things with 2-D grids or 3-D grids, what basically
happens is you've got to get more interconnection. As you go to
the bottom right, what you see is this starts looking like something,
right? Like the telephone company. Or a Network infrastructure.
And that's really what you're going to see that computers are
What we've actually ended up doing is the same type of thing.
We've realized that we've got to go away from a Bus structure,
and we're going now to Router technology. So what we do is every
processor goes into some hub Router chips, and then as you grow
it, you start having the hub chips connect up to other Router
chips as you start growing and growing and growing. What happens
here that's different is I go from having a whole bunch of these
P-zero processors, and as I start adding processors, I start adding
bandwidth. And that's what's key. So now I start getting more
linear scaling, and I get to start dealing with those tougher
problems. Those fundamental problems that I want to solve that
before I didn't have enough CPU, I didn't have enough memory,
I didn't have enough bandwidth, I can now start solving. And
in fact in our case, we've done some things where we have three
tiers. So finally we take those cubes and we connect them up
and now we're into the hundreds and thousands of CPU's all interconnected.
And I think that's really key where you'll see this is going
Several vendors have started in this area. What this does for
you also in a software sense is, should memory systems are those-whether
the unit processors or SMP type of architectures, those were easy
to program. They're a simple programming paradigm. What happens,
though, when you start growing a lot of times, people say, "We'll
go with massively parallel architectures." And we've seen
just from what's gone on in the marketplace, people found those
are tough to program. They're challenging systems to program,
and what happens is they've actually died in the marketplace because
to get useful work out of them has been challenging. What we've
tried to do here is make it so that you have a shared memory programming
paradigm, so they're easy to program as you get these large systems,
but they're also easy to scale. That's been key. Key benefit
Now, one of the things that's happening out here is we are going
to be changing, and you're seeing that the Web and other technology
has been changing the way we work. Basically there's shorter
time to market, increased insight, better innovation. In fact,
what's nice is when you have better technology at your fingertips
I can now redesign three, four times in the design phase instead
of later on when it's more costly after I've built the prototype.
When I built the prototype, to go back and make a change is extremely
costly. It takes a long time to sort of back that into all the
other information. If I make multiple changes, test them in different
scenarios early on, it gives me a lot of flexibility there.
What I want to do is I want to kick off from the lecture and
really go into and show you a little bit of the technology itself
and show you some of the things that are going on. I'm going
to show you some Web-based training in some things that we've
been doing. Inside Silicon Graphics, we have a huge intranet,
and we've got just under a half a million URL's within our firewall.
So this is for Silicon Graphics only. This is for our people
to look at. We basically changed everything that we've done,
whether it's our human resources, whether it happens I want to
sign up for 401K, I want to change my dependencies, my medical,
I want to look somebody up-all these things have changed. We've
gone from where we finally decided, inside our company, about
three years ago that it was time to make that fundamental change
to be able to give everybody access, as was mentioned earlier.
What we've come up with is a couple things that we've done.
One is called "Webucator", which allows us to basically
do our Web training on the Net itself. So what I can do is go
through here, and let me go through just a couple pages of giving
you an example of what we've done. We've taken our education
and made computer-based training, all Web-based. In this example
here, for instance, if I want to talk about all the educational
challenges and if I go into the paradigm shift, whether I'm giving
a lecture or whether it's a student or somebody interested who
wants to see the information here, well, as I pass over my cursor
on top of it I've got a balloon there telling me what the paradigm
shift-what's the important things about it. How it's easy, immediate
access to multi-media content. Or I can go down here, talk about
the older student, how many are returning to life-long education.
Or same thing: new learning and teaching needs.
What's interesting now is if you look at this from a teaching
technique. I can give it in lecture format, I can have it in
the library, somebody can go through the lecture off-line, we
can use it for distance learning. It goes back to the paradigm
of saying, "Author once, deploy many." And so now
I have different ways of doing it. I mean even if I want to talk
about things like the Nintendo generation, right? Which if you
look at the students that are coming on line, all of them are
very, very comfortable with technology, and that's really what
we've been able to provide here.
The thing that's important is looking at a way that you can say,
"O.K., this isn't that different" when you start talking
with professors, and that's usually the point when I first meet
up with the professors. They all say, "Well, I don't know
how to do Web pages." And you say, "Great. I hope
you never write a Web page. You shouldn't ever have to write
a Web page. If you're writing a Web page, then you're not doing
what you're supposed to be doing. But didn't you write down your
notes, your lecture notes, someplace? Don't you have something
you're working off of?" "Oh, I did this in Power Point."
"Great. Take your Power Point, and just take the Power
Point-and right now there's so many different ways-there's tools
out there that are external that will take a Power Point presentation,
convert it to a Web page. Or now Microsoft will do that directly.
Get it so you have it now: a sharable format."
I mean here's the thing that-I mean, again, using this presentation
here. I mean we've seen this change coming, whether you started
out in 1970-the whole centric aspect of things-and how we're moving
on to new technology, the client-server that came out more in
the 80's/90's. And if you look at some of the things that have
gone on with visualization that I showed you earlier, combining
that with the desk top, desk-centric aspect of it, basically you
put all that together in a Network, and that's why the Worldwide
Web makes so much sense. We didn't throw away our main frame
computers. We just put another face on them. And I think that's
what's been real key.
I'm going to actually pop down here to a quiz, too. This is
another thing that we've been doing is working with education
to say, "O.K., let's put our quizzes on-line." Whether
that happens to be in an educational setting directly, in an institution
of higher learning, or whether it's places like Marriot. I was
working with Marriot, and they said, "We want to maintain
everybody having the same quality, that same attitude about hospitality,
and how can we do that?" And we said, "Well, with the
Web. We can combine so there's a movie showing you the right
way to welcome a guest, the right way to make a bed, and you've
got a video-the right way to do-" Some of these things are
very mundane, but what's nice is you can put it in a multi-lingual
format and have it on the Web so that everybody can go through
this, then do the test. And so they can, you know, "Down
here's my test, and I can do all my checklists," and you
know whatever. I click this and say submit query, and it gave
me, I got zero out of five, so zero percent because I only filled
in one and I did it wrong. To have that type of test, we're doing
it so that it goes back into a database so you can actually follow
the progress of a student or progress of an employee or the progress
of somebody on the way, and it's a nice way to do the metric.
It goes back to that self-enabling type of technology.
I want to show a different one. This is some training we did.
We trained 11,000 employees in a three day period, and we basically
took everybody in our entire company, brought them together, but
electronically. We brought them together separately. How's that?
A good way to put it. What we did was we had on-line these Web
pages and everybody had a 800 number so they could call in. So
they got their voice over the 800 number, they got the lecture
materials-all the presentation materials-over the Web. And so
they were able to say, "O.K. well, let's go and talk about
our new product." Well, what was interesting is, for those
people who missed that training, they could also get the lecture
and the audio notes. So if I wanted to get speaker notes, I click
on here, and it basically gives me the speaker notes, for each
page, on a page by page basis. Or I could actually go for the
audio notes themselves. And so what will happen is it will load
in the audio for me for the lecture, so as I go from-
Audio Notes Example: I'm the product manager for Irex
and ASD and the product manager for Cellular Irex. The announcement
that is occurring now marks the advent of a new software architecture
for scalability and large-
Heibein: O.K. And when I skip to another page, it brings
me up to the next audio and starts talking about that. What's
Audio Notes Example: Cellular Irex is aimed at realizing
the biggest dreams in-
Heibein: O.K.. I'll turn that off. Well, you get the
idea of what it basically goes into. Now, on-line, I've got the
lecture that I may have missed. O.K.. All we did was record
the voice. We did the very minimum amount of editing on that,
basically chopping it up so it went with each slide, and that
was about it. Down the bottom you'll see that query, where it
says "Send query." Basically that was so that during
that presentation we could type in and say, "Why did you
call it Cellular Irex?" You know? Boom. O.K. "Oh,
I see we get a couple notes." And what would happen is
we had a chat system, and that would add to the chat. And so
then the presenters would say, "Oh, I see, we have a couple
questions about why we called it cellular Irex. Let me explain."
It was very interactive, and we had, again, everything going-One
way out voice, but everything coming back electronically. And
it was like I said 11,000 employees in three days were done this
way. Part of the reason for the three days, it was actually two
days of material, and everybody was 8-hour shifted depending on
what part of the world they were in, and so that's why it took
three days to do that training. And it worked out quite well
Some other things that I think are important is that what we've
been doing is adding more and more and taking the technology,
taking the 3-D, taking the Web, and putting them all together,
merging all of these, and-we're going to load in cyber-anatomy.
And we've been working with a lot of universities and med. schools,
and there's been some information they always need to be able
to share with the students, and you can't always have a cadaver.
There we go. And sometimes it's nice to have something electronic
that you can say, you know, like there is a lot of junk on the
Net. What you want to do is you want to make sure this is information
that you've provided.
And here this is going to combine Java, VRML, several different
items here. What I can do is, if you'll look, I've got two balls
that are blinking there, and I can sort of drive up towards this,
get a little bit more into here. Let me turn off collision detection.
Boom. There. O.K.. So up here you can see there are two balls,
and if I click on this ball right here what it does is it evolves
this human form out of the center there. So obviously it looks
more like a crash dummy with all the checkerboard on it, but you
sort of see what's going on there. Let me move up here so I can
take a look. There you go.
Down on the bottom, what looks like a capsule, that sort of drives
me to pre-computed locations so I can sort of take a look at this
and say, "O.K. Well, what I want to do is I want to take
this form, and I'm going to have it evolve one more level,"
and so now it's going to remove the skin for me, and you're going
to see the bone, etc. What I can do is, now I've got-you sort
of hear the breathing there? A little bit of heartbeat? Actually
this is all 3-D audio, too. So what would happen is, if I pass
behind it, it would sound different than if I'm in front of it.
It actually takes advantage of the cranial mass to know how it
should alter the audio for front, back, and right and left, etc.
Let me move a little closer here. There's the form. O.K. What
I want to do-I'm going to move right over here, take this off.
I think when you move your arm back like this it makes the bones
go away. I think that's how that works. I didn't go to med.
school, so I'm not sure. Now if I click on something-I can click
on the lungs. Now what I'll get, I'll get something that's more
of a picture on the right hand side. You know, more of a schematic,
a drawing there, and so I can look for different information,
look down here again. All of it's linked. So I'm going from
the 3-D world to the 2-D world, all the different information.
If I pull this away I think I've got-I want to get a little bit
closer here. I can also take myself and say, "I want to
be halfway between here and this point," and it takes me
right to that item. There we go, that little beating heart.
And click on the beating heart and away it goes.
What you can do though, the key is, this may not be the most
sophisticated example in the world in terms of you could probably
have a little bit more realistic looking heart and all, but I
think the key is saying, "O.K., here's where they all work
in relationship to each other. These are the parts. This is
where they are in a 3-D world," and so I can move within
there and so I can move with there and, like I said, go grab that
Another one that we've done is something similar to that where
we did basically cyber-astronomy. With the cyber-astronomy, it's
a way to be able to go through the cosmos and visit different
planets and all, taking advantage of the same type of 3-D environment.
And so what I can do again is look at it. You'll see it first
brings up everything in a fairly crude form. Here's the paths
of the planets. But as it's downloading, it gets more sophisticated
for me, and so you'll see as that goes on there.
What's nice is I'm involved in it. I have the interaction.
So what you're going to see that I can move this, I can manipulate
it, go take the journey as I need to do the discovery. I think
that's what's real powerful. So if I take that and rotate it
from different points of view, whatever, or travel from one place
to another-there's Pluto. Let me go a little bit closer. And
so you can see Pluto a little bit on the dark side there. Neptune.
If I click on Neptune, you all of a sudden see the information
about Neptune in terms of different things: how many satellites
it has, all the information about the rings. If I click on the
satellites, it tells me about which satellites it has and so on
and so forth. So I can find more information. If there's a body
there going by, I can click on that body directly and say, "Oh,
that's Triton. That's one of it's moons." So again, I get
to do the self-discovery.
What I want to do is go right up to-there we go. Let's do the
approach here. As I get closer-let's get over there to takeoff,
get over to Saturn, because that's of course always one of the
most dynamic looking ones. But there you can see is a field of
stars and like the cloud of other satellites and things around
Saturn, what make up its ring. Of course, as I go into it I can
become more part of it. This is another way of looking at virtual
reality: being able to do things that you couldn't normally do,
which is real key.
One other thing that we've done, and this has been real powerful,
we've been working with the military where what they say is, "O.K.,
we need to send out somebody who might not be trained on a certain
piece of equipment. How do we take that person whose never been
trained on it and basically have them go out in the field-and
maybe it's a satellite down-link-how do I have them install this
thing, point it in the right direction, and do everything?"
What we've done is we're working with them on putting Web pages
on a CD ROM-same type of thing you can do for distanced learning,
right? Or for selling your educational materials to other universities
and distributing this-well, here's the case where this is the
computer I'm actually running off of, but if I go in here I can
click on it. It brings me up and says basically, "O.K..
click here." So I click on that. You can see on the monitor
there it's re-booting the system, an so we'll see it in a second.
The light goes from orange here to green in a second as things
start up. So what's nice is I say, "Ah, I understand that."
And let me move around to a different side of this, and you can
see the screen starts changing, etc.
I've also got a CD ROM here, and I can interact with that CD
ROM. Insert it in, whatever. Pull it out, or say, "Well,
let's take a look at that CD ROM. Is it the right one? Uht,
there it is. There's my CD ROM." So the quality of what
I'm doing with a Netscape Web Browser right now, it's there today.
This is today's technology. It's not really as far into the
future as many people think.
Let me cut down here and sort of show you, O.K., with that screen
again, but right here there's this piece. Well, I want to insert
everything and start making that go together. So I click this.
Oh, it's the AV module. Now it's doing the assembly. I can
do this either in a movie, which any of you who have done a M-peg
movie on the Web, you know it's mega, mega, megabytes. Well,
here we've got something that's hundreds of k-bites instead of
megabytes, and that's been what's real key in terms of what we
can do there. And that's why showing it 3-D, it's actually smaller
data, because instead I've got the four corners of some item and
then I just paste a picture to it. I don't have-I let the computer
calculate the rest of that for me. And again, I can also turn
this into transparent. So it's making it a little more transparent.
So when I insert this, I'll be able to see where the power supply
goes. And you basically get the idea. You know, this goes on
here. This piece goes on here.
Whether I want to interact with everything-I might want to say,
"Well, what's behind her?" Pull this module off. O.K..
And then I've got different parts I can look at. So let me get
a little bit closer in here. And so I can see that this module,
that's the memory. There's the display engine, and there's the
CPU itself. So I can click through, and I can start understanding
it a little bit more and say, "O.K.., well I want to insert
this. Now what all happens?" Boom. That top comes back
on, that goes there, and that all inserts. So, again, another
example of what can be done. And I can turn this thing even more
transparent, make the whole thing go away so just I see the modules.
But again, the interactivity I think is what's important of what's
Timing-wise, I'm just about up with my time, so what I want to
do is I want to finish my last two slides on here and go into
my conclusion for you. Basically, when you see what's happening
between the Worldwide Web, what you can do with visual computing
and how it's now become affordable, high performance computing,
the cost coming down, the availability coming down, so now whether
it's on a departmental level or whether it's on an individual
student level these things are now available. So this is the
one thing is technology is not out of reach like it once was.
The other thing is that things you saw are things that we're
doing today. It's not something that, again, is futuristic.
I'm working right now, and I've developed a distance learning
architecture based on the Web. And there's actually a commercial
company who sells education that's going to be using this architecture
so that they'll be able to distribute it and deliver it right
inside a corporation's firewall. And we're just using off the
The last thing I didn't mention large enough was just big data
and real world problems. Today because systems-although Microsoft
and Intelmate have just gone to 32 bit, a lot of the rest of the
world is already gone to 64 bit, and why that's important is,
when you start looking at very, very large problems, that's available
today so that you can go in there and do the computation on that
and be able to address more than four gigabits of data in a single
structure. And that's been real key is what you get with 64 bit.
We're doing some things right now with ten terabit single image
databases. Ten terabits, single image. And that's incredible.
A couple years ago-I mean well we couldn't have done it just
three years ago, and in fact this will be the first one over ten
terabits in a single image. So this is showing you what can be
So in conclusion, I think there's a lot out there to be used.
What I want to challenge you to do is take a look at this technology
and ways that you can implement it inside your operation. I'm
excited about it. I'm jazzed up. every time we've gone in and
implemented something like this, people have usually taken to
it. And those people who are fearful of the technology replacing
them or doing whatever, instead it actually enabled them to have
a more exciting job than they had before. And so they found out
after their fear was over they were able to now say, "I now
have got something new to be jazzed up about as an instructor,
as an educator, as a student." And I think that's been
real key. So thank you for your time.