[00:00.000 --> 00:01.780]  I'm going to try it again this year.
[00:02.840 --> 00:05.420]  My name is Adam.
[00:06.020 --> 00:08.520]  My moniker de jure is Javaman.
[00:09.680 --> 00:11.280]  You can call me anything you want.
[00:11.540 --> 00:12.620]  I don't really care.
[00:15.100 --> 00:17.140]  Let's see, we're not here to sell anything.
[00:20.290 --> 00:21.860]  So we'll end on time.
[00:23.740 --> 00:26.300]  Let's see, my areas.
[00:27.060 --> 00:37.500]  Generally, I spent five years studying hardware design and concentrated mostly in digital signal processing, communications, and modulation type issues.
[00:38.180 --> 00:44.000]  Now, most of my work has been researching computer network architecture, computer network protocols.
[00:44.240 --> 00:50.460]  And most of my hardware design development stuff has been in RF hardware communication systems.
[00:51.540 --> 00:56.260]  And now I'm just getting back to, you know, building stuff on my own, again, outside of industry.
[00:57.400 --> 01:00.360]  I notice that nothing really takes the fun out of something than doing it for a living.
[01:01.300 --> 01:03.300]  And I'm sure it's probably true for many people here.
[01:05.680 --> 01:08.820]  So if there's any questions along those lines, we can answer.
[01:09.000 --> 01:14.820]  And I'm also going to pass it over to Nick first, and he'll introduce himself to talk about some of the things he works on.
[01:14.940 --> 01:18.940]  And we'll pose a few questions just to get the ball rolling.
[01:18.940 --> 01:26.120]  And then we'd like people to start raising their hands with questions that they may have, you know, kicking around.
[01:26.620 --> 01:27.100]  Okay.
[01:28.920 --> 01:29.440]  Son?
[01:29.720 --> 01:30.060]  Okay.
[01:48.060 --> 01:48.100]  Son?
[02:00.200 --> 02:02.280]  I don't like here because this land is not balanced.
[02:02.500 --> 02:03.240]  Please go away.
[02:03.580 --> 02:03.680]  Sorry.
[02:04.380 --> 02:05.920]  And topics along those lines.
[02:06.440 --> 02:08.820]  So, okay.
[02:09.140 --> 02:09.440]  All right.
[02:09.570 --> 02:09.790]  All right.
[02:09.990 --> 02:11.160]  So my first question.
[02:11.390 --> 02:13.350]  How many people, this is a good one.
[02:13.450 --> 02:16.010]  How many people think that there's an infinite amount of bandwidth out there?
[02:16.270 --> 02:23.550]  And the only reason why people don't have very high data rate systems is because of the FCC.
[02:24.540 --> 02:25.850]  I think that's a good question.
[02:25.980 --> 02:27.230]  I want to get a show of hands.
[02:27.390 --> 02:27.890]  Don't be shy.
[02:28.350 --> 02:29.550]  Everyone who believes this, raise your hands.
[02:30.610 --> 02:31.050]  Come on.
[02:31.160 --> 02:32.200]  There's going to be more people than this.
[02:34.510 --> 02:35.420]  Infinite's a really big number.
[02:35.490 --> 02:35.770]  I know.
[02:36.100 --> 02:37.420]  But I want to get a show of hands.
[02:37.540 --> 02:40.230]  And it's very important for me that I get a show of hands on this one.
[02:40.830 --> 02:41.350]  Just curious.
[02:41.550 --> 02:43.600]  We have not too many people.
[02:43.670 --> 02:44.180]  We have a few.
[02:44.860 --> 02:45.260]  All right.
[02:46.600 --> 02:48.210]  And that's patently false.
[02:48.490 --> 02:49.860]  Because yes, infinite is a big number.
[02:50.160 --> 02:54.460]  And the only reason why I ask this question is because I read Slashdot every so often.
[02:54.580 --> 02:56.700]  And I truly believe that Slashdot rocks your brain.
[03:00.440 --> 03:06.050]  And the range of stupidity I see coming out of that area, it boggles the mind.
[03:06.360 --> 03:11.020]  And sometimes I just want to roll my eyes back in my head and go to sleep forever.
[03:14.500 --> 03:17.640]  And not too long ago... There it is.
[03:17.820 --> 03:18.540]  I'll do it right now.
[03:18.640 --> 03:18.880]  Watch.
[03:20.140 --> 03:26.440]  And not too long ago there was an article that said that all these technologies coming out which say that...
[03:26.440 --> 03:29.660]  which just increase bandwidth and that bandwidth is going to be infinite.
[03:29.710 --> 03:43.830]  And the only thing that's stopping this is the FM radio industry and the TV industry who won't prevent the average Joe from getting it so they'll defeat their monopoly on communications.
[03:44.770 --> 03:45.820]  That's bullshit.
[03:46.210 --> 03:46.540]  All right?
[03:47.740 --> 03:48.820]  That's the bottom line.
[03:51.050 --> 03:56.160]  There was very fundamental work done by a very, very brilliant man by the name of Claude Shannon.
[03:56.330 --> 03:57.710]  Done in the late 40s.
[03:57.880 --> 04:00.240]  And he is the founder of what's called information theory.
[04:01.020 --> 04:06.320]  He and Norbert Wiener were two people who really pushed this theory forward.
[04:06.660 --> 04:14.260]  And he actually laid down what is fundamental mathematical relation which defines a maximum capacity for any channel.
[04:14.640 --> 04:21.520]  And so the maximum capacity for a channel was related to bandwidth and a logarithm of the signal noise ratio.
[04:21.520 --> 04:28.350]  And what this said was that if there is absolutely no noise in a channel, zero noise.
[04:28.590 --> 04:30.760]  SNR is, you know, it's none.
[04:31.000 --> 04:32.850]  None whatsoever, yes, you have infinite bandwidth.
[04:33.470 --> 04:36.470]  The chance of having zero noise is zero.
[04:36.680 --> 04:43.210]  Noise comes from quantum fluctuations in devices, thermal noise, launch number, things like that.
[04:44.090 --> 04:48.680]  And what ends up happening is that this noise will kind of cloud your vision.
[04:48.680 --> 04:53.270]  Imagine taking, looking at, okay, good example.
[04:53.380 --> 04:56.410]  You go to the optometrist, okay, and there's a sheet of letters.
[04:56.700 --> 04:57.000]  You know what I mean?
[04:57.030 --> 04:57.760]  You have to read those letters.
[04:57.880 --> 04:59.080]  And you get down towards the bottom line.
[04:59.150 --> 05:02.150]  As they get smaller, those letters, you're able to fit more of them on the line.
[05:02.260 --> 05:03.090]  You should be able to read more of them.
[05:03.560 --> 05:07.000]  Let's say you take a sheet of wax paper and put it in front of it.
[05:07.800 --> 05:09.440]  That line gets even a little bit more blurry.
[05:09.520 --> 05:13.090]  And you're able to still pick out most of the symbols and read them, but it still gets harder and harder.
[05:13.090 --> 05:23.080]  As you increase the number of sheets of wax paper in front of that, you know, that optometrist sign, it becomes near impossible to differentiate those symbols.
[05:23.600 --> 05:26.080]  That's what noise really is in a communication channel.
[05:26.820 --> 05:38.340]  If there is zero noise, you can pack infinitely many symbols by different voltage levels, for example, if we're dealing with baseband signaling, into that area.
[05:38.340 --> 05:42.720]  But noise blurs that area between two signaling points.
[05:42.980 --> 05:49.060]  And so what ends up happening is that it's very difficult to actually make that distinction between different voltage levels.
[05:49.400 --> 05:57.720]  And just the same way it would be if you have, like, you know, some kind of haze in front of a large number of letters in a row that you try and differentiate.
[05:58.920 --> 06:03.560]  Now, technology such as spread spectrum, they say, oh, yeah, it increases all your bandwidth.
[06:03.560 --> 06:07.520]  Well, the real reason why people use spread spectrum is more for multiple axis technologies.
[06:08.550 --> 06:20.320]  Basically, the way you're able to, instead of having people change what frequency they're on, you'll take their signal, which may be 100 hertz, or 100, let's say it's one kilohertz wide.
[06:20.880 --> 06:27.500]  And you mix it in with a square wave, which runs at 100 times higher rate than your symbol rate of your channel.
[06:28.080 --> 06:30.000]  So your signal becomes 100 kilohertz wide.
[06:30.000 --> 06:35.440]  That doesn't mean they really have more bandwidth as in the idea of data rate transmission.
[06:36.120 --> 06:37.620]  Their signal takes up more bandwidth.
[06:37.800 --> 06:42.560]  But the reason why is because you have different square waves, different signaling codes.
[06:42.980 --> 06:47.700]  The two signals can occupy the same channel space at the same time.
[06:48.180 --> 06:50.800]  That doesn't mean that, you know, you're getting bandwidth for free.
[06:51.040 --> 06:56.700]  You just took one guy's one killer's and jacked up to, you know, 100 times larger without getting any additional data.
[06:57.490 --> 06:59.660]  And the other person's occupying the same space.
[06:59.920 --> 07:01.520]  And the reason why you do this is a couple of reasons.
[07:01.620 --> 07:03.720]  You get what's called a...
[07:07.490 --> 07:08.600]  I can't remember the exact...
[07:08.600 --> 07:09.900]  They can't remember the technical name for it.
[07:10.000 --> 07:17.620]  But basically what happens is that if you have a single noise point, when you apply the de-spreading code, it flushes out the noise.
[07:17.620 --> 07:19.220]  It spreads out the noise.
[07:19.940 --> 07:21.740]  I think it's called like time-space-gain.
[07:24.420 --> 07:30.400]  You also get that and also on the opposite end of your receiver, you have one receiver chain and you break it out into multiple...
[07:32.420 --> 07:37.040]  multiple paths of demodulation and you just mix in different spreading codes.
[07:37.240 --> 07:44.760]  So you have like one front-end amplifier to just do spread on different frequencies rather than having to have different receivers each at a separate frequency.
[07:44.760 --> 07:47.350]  It's actually a very convenient way of doing system design.
[07:59.860 --> 08:02.180]  For questions, if you can use them like...
[08:02.180 --> 08:02.200]  Yeah.
[08:02.360 --> 08:02.660]  Right there.
[08:03.780 --> 08:05.400]  Is this similar to the way...
[08:05.400 --> 08:10.580]  Is this similar to the way that cable modems work over the same line that your cable television runs over?
[08:10.800 --> 08:12.220]  That's something a little bit separate.
[08:12.460 --> 08:14.600]  That's just frequency division multiplexing.
[08:14.600 --> 08:24.360]  What I'm talking where basically all your cable TV occupies one frequency space and all your data rates occupy another data space.
[08:24.480 --> 08:25.860]  And that's exactly what happens with DSL.
[08:26.400 --> 08:32.780]  DSL lines is voices between zero and three kilohertz and DSL is between like 10 kilohertz and 20 megahertz, something along those lines.
[08:32.940 --> 08:35.540]  That's why you put the line filters on your phone when you get DSL.
[08:35.940 --> 08:40.340]  Is to keep, you know, all the extra DSL signals off your other, you know, off your other phones.
[08:41.640 --> 08:42.280]  And DSL...
[08:42.280 --> 08:43.220]  Actually, I'll get back to that.
[08:43.340 --> 08:44.240]  But DSL is...
[08:45.300 --> 08:48.000]  This is why the cable companies are going to beat out the telco companies.
[08:48.200 --> 08:50.220]  Because DSL is really a bad hack.
[08:51.140 --> 08:51.460]  And...
[08:51.460 --> 08:52.440]  Well, I shouldn't say a bad hack.
[08:52.520 --> 08:53.160]  It's a great hack.
[08:53.260 --> 08:53.860]  But it's a hack.
[08:54.140 --> 08:55.020]  And it only...
[08:55.020 --> 08:58.500]  The lines aren't designed to carry that wide signal path.
[08:58.640 --> 09:02.320]  And once you start getting a large number of customers in the area, all the lines are bleeding into each other.
[09:02.320 --> 09:04.180]  And your DSL data rates go down.
[09:05.460 --> 09:05.900]  So...
[09:05.900 --> 09:08.040]  Additionally, you can't be more than three miles from the telco.
[09:08.240 --> 09:11.520]  The reason why is because the cables physically can't carry that much data.
[09:11.740 --> 09:12.800]  It's like trying to run...
[09:13.280 --> 09:14.280]  A good example.
[09:15.100 --> 09:19.180]  It's like trying to take really bad twisted pair and run a gigabit Ethernet over.
[09:19.400 --> 09:21.160]  It's just the channel's not high enough quality.
[09:21.980 --> 09:22.860]  So that's...
[09:22.860 --> 09:26.320]  I mean, that's pretty much exactly what the DSL systems are.
[09:26.860 --> 09:28.520]  But let's get to your question in a second.
[09:28.660 --> 09:31.940]  And what I'm talking about spread spectrum is referred to as code division multiple access.
[09:31.940 --> 09:33.240]  CDMA phones.
[09:33.540 --> 09:34.080]  This is how they operate.
[09:34.860 --> 09:36.660]  So maybe we should start taking questions.
[09:37.200 --> 09:37.340]  You, sir.
[09:37.740 --> 09:37.940]  I don't know.
[09:42.240 --> 09:49.300]  In line with a most recent question, the cable companies are considering using phone access over their cable lines.
[09:49.460 --> 09:49.680]  Yes.
[09:49.920 --> 09:50.720]  They could...
[09:50.720 --> 09:51.480]  Put it this way.
[09:51.660 --> 09:55.140]  The phone companies have three kilohertz channels to everyone's house.
[09:55.140 --> 10:03.780]  The cable companies rolled out analog systems which provide, I think, up to one gigahertz wide channel to everyone's house.
[10:04.440 --> 10:09.800]  Cable companies have a sunk cost already in the air infrastructure of very wide channels to everyone's house.
[10:10.100 --> 10:11.600]  The phone companies have three kilohertz.
[10:11.720 --> 10:15.960]  The only way I can see the phone companies beating these people out, they start laying out fiber at everyone's house.
[10:16.060 --> 10:16.820]  And that's just expensive.
[10:18.420 --> 10:22.220]  I mean, so it's, it's just a matter of who put in the infrastructure first.
[10:23.720 --> 10:26.100]  We have fiber at our neighborhood in Miami.
[10:26.560 --> 10:27.760]  I mean, they're doing it...
[10:27.760 --> 10:28.660]  Is this new development?
[10:30.180 --> 10:33.020]  I guess they've been doing it now for the past five years.
[10:33.620 --> 10:35.960]  I mean, this is your housing complex a new development.
[10:36.040 --> 10:36.740]  Yeah, probably how it is.
[10:37.140 --> 10:38.540]  There's a fiber box.
[10:38.900 --> 10:40.140]  How long ago was your house built?
[10:40.140 --> 10:42.500]  Uh, my house was built in the American body.
[10:42.760 --> 10:42.960]  All right.
[10:43.040 --> 10:46.020]  And there was some houses built in the building now in Miami.
[10:46.200 --> 10:47.860]  They were laying out fiber boxes.
[10:48.240 --> 10:50.760]  Right, that fiber goes to something called a remote terminal.
[10:50.960 --> 10:54.080]  And then that gets changed over to your twisted pair.
[10:54.400 --> 10:55.400]  And it goes back to your house.
[10:55.580 --> 10:57.160]  That's how DSL is being deployed now.
[10:57.840 --> 11:05.220]  Where they're, they're finding that the, the distance limitations is causing them to basically put a DSLAM in every network, in every neighborhood.
[11:06.040 --> 11:09.380]  So that, that's just an effort to push it further, push the CO.
[11:09.960 --> 11:11.860]  So to speak, further closer to the customer.
[11:12.180 --> 11:17.220]  But still the cost of actually going through and drop, and, and it's actually trenching costs.
[11:17.620 --> 11:20.220]  Uh, of putting a fiber at one's house is just too expensive.
[11:20.540 --> 11:22.280]  And it's not the actual cost of the fiber really.
[11:22.360 --> 11:22.960]  It's the cost of labor.
[11:24.000 --> 11:28.240]  So, that's, that's actually the biggest cost by far of any networking technologies.
[11:28.360 --> 11:35.020]  The actual price they have to pay, uh, the people will actually go out and dig the trenches and, and actually lay the line.
[11:35.160 --> 11:36.940]  It's not the, the actual cable technology.
[11:37.380 --> 11:40.920]  If they give me access to the search engine, should I be able to run my own?
[11:41.300 --> 11:51.840]  Well, yeah, but I don't think, uh, the gas company would be glad when you, you know, drop a shovel right into one of their gas lines, they have to come out and, and fix it.
[11:52.020 --> 11:56.780]  I mean, you, you have to do these things in a, not, not as much in a centralized manner, but in an organized manner.
[11:57.440 --> 12:00.680]  Yeah, I mean, you can't have everyone going out and digging up the street and laying out their own fiber cables.
[12:03.800 --> 12:04.360]  Yeah.
[12:05.460 --> 12:06.020]  Yeah.
[12:07.400 --> 12:07.960]  Yeah.
[12:08.320 --> 12:08.720]  Yeah.
[12:09.980 --> 12:20.560]  Well, they're, they're running fiber to each neighborhood to deploy remote terminals or slits to service your phones, your TSL, and other services that they're planning to roll out.
[12:21.280 --> 12:24.620]  And, I mean, fiber to the, to your house really doesn't make sense.
[12:24.620 --> 12:28.260]  There's just not that much bandwidth out there, even if you did have nothing of a pipe.
[12:29.280 --> 12:30.620]  Uh, excuse me.
[12:30.960 --> 12:33.600]  We're, we're only getting half the conversation in the back.
[12:33.800 --> 12:36.840]  So, if you can either repeat the question or make them use the mic, we can all participate.
[12:37.680 --> 12:37.940]  Thank you.
[12:39.020 --> 12:40.220]  Yeah, we need to keep the mic.
[12:40.720 --> 12:41.080]  Okay.
[12:42.640 --> 12:44.100]  We asked people to use a mic.
[12:45.460 --> 12:47.400]  But, for now, I'll repeat the question.
[12:47.900 --> 12:49.040]  I feel chastised.
[12:49.360 --> 12:52.120]  Why don't, why don't, why don't people walk up to the mic?
[12:52.240 --> 12:52.380]  Yeah.
[12:53.060 --> 12:55.660]  So, uh, everyone walk up to the mic from now on.
[12:56.220 --> 12:57.060]  Or else.
[12:57.060 --> 12:57.840]  Or else.
[12:58.900 --> 13:00.280]  We send you to France.
[13:01.060 --> 13:05.540]  We were talking about laying fiber to housing complexes, apartments, whatever.
[13:05.740 --> 13:06.440]  I live in Houston.
[13:06.740 --> 13:10.060]  Warner Cable runs fiber up to some point.
[13:10.920 --> 13:17.580]  You know, everybody's provided with cable modem access within the range of Time Warner communications through Roadrunner.
[13:17.900 --> 13:19.620]  This is not what you're talking about?
[13:20.440 --> 13:26.160]  Your cable modem actually is delivered via fiber up to a certain point and then it's brought back to the head end.
[13:26.740 --> 13:29.980]  So, the remote terminal can be many different things.
[13:30.180 --> 13:33.780]  It can be your transceiver for your cable access.
[13:34.060 --> 13:37.860]  It could be a slick for serving phone lines or digital lines.
[13:37.860 --> 13:42.040]  Um, it could be a DSLAM for DSL services.
[13:42.480 --> 13:44.900]  Um, Sprint, I don't know if you've ever heard of Sprint Ion.
[13:45.280 --> 13:46.780]  It, uh, was a project that failed.
[13:46.920 --> 13:51.200]  They were trying to deploy, um, voice over DSL to everybody's house.
[13:51.280 --> 13:53.980]  Everybody would get two phone lines and a high speed DSL line.
[13:53.980 --> 14:03.500]  And they actually had to run fiber to all these neighborhoods, install a DSLAM, go out, have a truck roll, go and install CP equipment in everybody's house.
[14:03.940 --> 14:07.120]  And it was just taking too long and not that many people were signing up.
[14:07.220 --> 14:20.940]  But still, in order to serve everybody, to get everybody DSL or everybody cable modems, they have to get your high speed pipes out past the CL infrastructure and into the neighborhood.
[14:20.940 --> 14:22.560]  Uh, and then from there it splits off.
[14:24.060 --> 14:26.480]  Actually, uh, that reminds me of something else.
[14:26.700 --> 14:29.500]  Uh, there's a lot of people talking about dark fiber.
[14:29.860 --> 14:33.640]  Uh, that, that there's a huge amount of fiber infrastructure that's laid out there that's just not even lit.
[14:34.320 --> 14:37.380]  Uh, the reason for this is because, again, the trencher costs are so high.
[14:37.620 --> 14:42.980]  They say, you know what, we're going to do this once and we're not going to, we're not going to be able to do this again for another 50 years or something like that.
[14:43.340 --> 14:44.620]  Or another 30 years.
[14:44.840 --> 14:46.980]  So what they're doing is just laying out huge amounts of fiber.
[14:46.980 --> 14:51.020]  In case they say, oh, well, in case we fill up the capacity of our fiber channels as this.
[14:51.800 --> 14:58.820]  Um, people are going out and developing very, very dense, uh, modulation techniques for fiber.
[14:58.940 --> 15:02.800]  I know that, uh, Sprint, I think...
[15:02.800 --> 15:03.260]  DWDM?
[15:03.420 --> 15:05.480]  Yeah, Sprint runs, like, I think, 100 colors.
[15:05.800 --> 15:08.920]  It's moving to 40 colors, I believe, or maybe 100 colors.
[15:09.200 --> 15:14.160]  By me 100 colors, I mean 100 separate, or 40 or 100 separate laser channels than one fiber.
[15:14.160 --> 15:14.480]  Yeah, different wavelengths.
[15:15.220 --> 15:16.440]  Different, yeah, different...
[15:16.440 --> 15:21.140]  You can think of those colors because that's what different wavelengths are at the optical domain.
[15:21.740 --> 15:23.420]  They're running, you know, between...
[15:23.420 --> 15:27.240]  I think they're right now 40 and they're going to 100, or they're just going to 40 now.
[15:27.560 --> 15:29.600]  Uh, different wavelengths down single fiber.
[15:29.720 --> 15:31.440]  And they only run at 25% capacity.
[15:31.940 --> 15:34.180]  Actually, I think most of the time they even run far, far less.
[15:34.400 --> 15:36.540]  Because there really is a glunt of bandwidth in the core.
[15:36.860 --> 15:37.380]  People just...
[15:37.380 --> 15:39.960]  There's so much bandwidth in the network core, it's not even funny.
[15:41.280 --> 15:41.680]  Question?
[15:42.720 --> 15:44.640]  I'm not going to walk because I'm sitting right next to him.
[15:47.060 --> 15:57.340]  Since fiber tends to be very expensive to take all the way down to the end point where the customer is located, do you think that wireless is a simple and inexpensive solution to make that last mile stretch?
[15:57.560 --> 15:59.780]  It's not the fiber that's expensive, it's the labor cost.
[16:00.160 --> 16:01.180]  Okay, but I think...
[16:01.180 --> 16:03.520]  Yes, I think wireless is, but there's only so much sky.
[16:04.200 --> 16:05.500]  That's really what it comes down to.
[16:05.500 --> 16:07.280]  Uh, if...
[16:07.280 --> 16:10.200]  I think that wireless is an effective way of doing it.
[16:10.300 --> 16:15.260]  And I think that you're actually going to first see it crop up in India or Africa.
[16:15.820 --> 16:18.080]  Uh, they're actually starting to do wireless local loops over there.
[16:18.380 --> 16:20.500]  And the reason why is because, um...
[16:21.600 --> 16:23.740]  In a way, these countries are...
[16:23.740 --> 16:25.580]  I don't want to say lucky, but there's a...
[16:26.360 --> 16:28.600]  The term was used when this...
[16:28.600 --> 16:30.540]  After the Soviet Bolshevik Revolution.
[16:30.540 --> 16:31.920]  And it was called the...
[16:31.920 --> 16:32.460]  The, uh...
[16:32.460 --> 16:33.520]  The reward of backwardness.
[16:34.040 --> 16:35.840]  And what happened was that, uh...
[16:35.840 --> 16:38.140]  The United States in the 1930s were just getting over...
[16:38.140 --> 16:39.440]  We were right into the depression.
[16:39.900 --> 16:41.800]  And we were saddled with all this industrial equipment.
[16:42.780 --> 16:43.220]  Uh...
[16:43.220 --> 16:44.960]  The Russian, when they went to...
[16:44.960 --> 16:50.660]  After the Bolshevik Revolution, moved from an agrarian society to an industrialized society and bought all new machine equipment.
[16:50.940 --> 16:52.700]  And because of that, they were...
[16:52.700 --> 16:54.400]  I mean, they were pretty much...
[16:54.400 --> 17:00.320]  They were very much capable and getting to be ahead of the United States because they were able to buy a whole mess of brand new stuff.
[17:00.720 --> 17:02.760]  And it really is this price of backwardness.
[17:02.880 --> 17:04.320]  And a good example is that when, um...
[17:05.760 --> 17:05.980]  Uh...
[17:05.980 --> 17:07.520]  I don't want to say when, but, uh...
[17:07.520 --> 17:09.220]  North Korea is probably going to...
[17:09.220 --> 17:10.140]  I don't know.
[17:10.760 --> 17:12.960]  Eventually North Korea and South Korea will probably become unified again.
[17:13.200 --> 17:20.640]  And what's going to happen is that because North Korea is so poor for so long, and they're going to start rolling out, you know, phone lines to everyone, they're just going to roll out fiber in everyone's house.
[17:21.280 --> 17:23.560]  So, you're going to see, you know...
[17:23.560 --> 17:26.080]  And being that South Korea is so industrial...
[17:26.080 --> 17:30.280]  You know, so technological and so industrialized, when reunification probably will happen.
[17:30.980 --> 17:35.780]  You're going to see North Korea become one of the tech centers of the world because it's just to be saddled with brand new hardware.
[17:38.600 --> 17:39.000]  Uh...
[17:39.860 --> 17:40.260]  Mike...
[17:40.260 --> 17:41.660]  Does anyone pass the mic back to him?
[17:41.660 --> 17:42.720]  Can I just speak up?
[17:43.540 --> 17:44.160]  I don't know.
[17:44.300 --> 17:45.240]  This guy doesn't like that.
[17:50.940 --> 17:51.940]  He's actually serious.
[17:58.170 --> 18:01.470]  I just want to take issue with your economic interpretation.
[18:01.750 --> 18:02.150]  I mean...
[18:02.150 --> 18:02.850]  Dude, I can't...
[18:02.850 --> 18:03.970]  I can't hear you.
[18:04.870 --> 18:05.330]  Hold on.
[18:08.710 --> 18:13.070]  I just want to take a second to get away from this mic.
[18:14.330 --> 18:17.250]  I just want to take issue with your economic analysis.
[18:17.250 --> 18:24.030]  I mean, when it comes down to it, the Soviet Union didn't have any advantage by starting as a peasant country.
[18:24.490 --> 18:32.390]  North Korea is not going to have any advantage starting from scratch when it has zero experience picking market participants.
[18:32.830 --> 18:37.790]  I mean, their chance of getting it right the first time out is like slim and numb.
[18:37.870 --> 18:39.790]  I'm just using the terms that historians use.
[18:39.930 --> 18:40.250]  I understand.
[18:41.090 --> 18:43.390]  And I'm not saying North Korea is going to get it right the first time.
[18:43.450 --> 18:45.610]  I think South Korea's government is going to run things right.
[18:45.610 --> 18:47.570]  I think that's happening in eastern Germany.
[18:47.830 --> 18:52.150]  As they're starting to build new infrastructure, they're getting more advanced infrastructure than what's already in place in Germany.
[18:52.510 --> 18:53.410]  That's the bottom line.
[18:53.530 --> 18:54.010]  I mean, it's like...
[18:54.010 --> 18:54.530]  Put it this way.
[18:54.610 --> 18:56.110]  They roll out technology over there.
[18:56.590 --> 18:58.290]  And it's, you know, at some date N.
[18:58.470 --> 19:02.430]  And then you have to build out a whole new country of technology at some date N plus 10.
[19:02.570 --> 19:05.710]  Well, you know, N plus 10 is greater than N.
[19:05.710 --> 19:08.250]  And it's not going to give them the old technology of the new stuff.
[19:08.510 --> 19:10.610]  Well, you can use Japan as an example.
[19:10.750 --> 19:12.650]  I mean, Japan has 3G wireless phones.
[19:13.270 --> 19:16.010]  Almost no common channel signaling in their infrastructure.
[19:16.790 --> 19:18.210]  And they were just...
[19:18.210 --> 19:21.270]  Because they've had to buy new equipment and roll it out.
[19:21.650 --> 19:27.250]  I don't know if it's so much experience in picking the people that are going to build that network out.
[19:27.370 --> 19:29.110]  I mean, there's only a handful of those.
[19:29.270 --> 19:30.250]  And they're all well experienced.
[19:31.070 --> 19:32.510]  Actually, Japan's a real good example.
[19:32.650 --> 19:34.990]  Japan's got a big problem with the fiber infrastructure.
[19:35.750 --> 19:36.910]  And what it is is because...
[19:37.950 --> 19:39.130]  Alright, fiber technologies.
[19:40.790 --> 19:44.650]  Fiber optics are a lot like wireless channels.
[19:44.830 --> 19:50.950]  They actually have a lot of the same characteristics in that you see first order, second order, distortions, and things like that.
[19:51.130 --> 19:52.750]  They call it dispersion in fiber.
[19:52.910 --> 19:56.730]  And it's from being a limited bandwidth and impurities in fiber.
[19:56.870 --> 19:58.490]  And someone came up with zero dispersion fiber.
[19:58.490 --> 20:00.390]  And they thought this was going to be the greatest thing in the world.
[20:00.570 --> 20:02.250]  And Japan just rolled it out all over the place.
[20:02.690 --> 20:09.130]  And what turned out happening was that when you have zero dispersion fiber, you still have the second and third order characteristics which limit your capacity to rechannel.
[20:09.790 --> 20:13.530]  So the United States, what they do is they do one run of...
[20:13.530 --> 20:19.470]  A partial run of one kind of dispersion and another partial run of another kind of dispersion fiber.
[20:19.610 --> 20:20.570]  And they cancel each other out.
[20:20.790 --> 20:23.510]  So what happens is we're getting better fiber infrastructure than Japanese.
[20:25.710 --> 20:33.350]  But that's just because they rush to roll out fiber before people really understood what was capable of engineering.
[20:33.630 --> 20:34.550]  I've got another question.
[20:35.390 --> 20:43.010]  Companies that market laser transmission lines claim that they're the ultimate solution to the last mile problem in an urban environment.
[20:43.130 --> 20:43.890]  Can you count in on that?
[20:44.730 --> 20:45.870]  We have one.
[20:46.050 --> 20:47.990]  I'm a graduate student at Drexel University.
[20:48.390 --> 20:51.210]  And we just acquired Hahnemann University.
[20:51.350 --> 20:52.230]  It's now our school of medicine.
[20:52.470 --> 20:56.650]  And we have a gigabit line of sight laser connection.
[20:56.930 --> 20:59.230]  And when it gets foggy out, the bandwidth drops.
[21:01.950 --> 21:02.370]  So...
[21:02.890 --> 21:03.410]  I know.
[21:03.570 --> 21:06.730]  I think that any technology along these lines, I think it's great.
[21:06.890 --> 21:08.650]  Anyone who's developing stuff like that, I think it's great.
[21:08.650 --> 21:10.650]  Because it flushes competition.
[21:10.910 --> 21:12.390]  It challenges people.
[21:13.110 --> 21:15.010]  I think that, you know...
[21:15.010 --> 21:20.270]  I mean, really, when it comes down to the reason why everyone's got high speed connections in everyone's house is pornography.
[21:20.970 --> 21:22.030]  And the web.
[21:22.670 --> 21:22.990]  No.
[21:23.150 --> 21:27.850]  On the internet, pornography drives technology.
[21:28.030 --> 21:28.930]  The reason why VHS...
[21:31.310 --> 21:34.070]  And it's not... I wish I was joking.
[21:34.070 --> 21:35.090]  You know why?
[21:35.190 --> 21:37.570]  Beta was a superior technology to VHS.
[21:37.850 --> 21:39.190]  The reason why VHS...
[21:39.190 --> 21:40.270]  Everyone's got VHS...
[21:40.270 --> 21:43.050]  Well, you know, had VHS VCRs in their house and now they have DVD.
[21:43.710 --> 21:48.350]  And the reason why is because VHS tapes are cheaper to shoot on.
[21:48.630 --> 21:53.450]  So, all these people who are, you know, making pornography decide to shoot on VHS tapes because they're cheaper.
[21:53.630 --> 21:56.390]  And people who want to watch porn were buying VHS VCRs.
[21:56.990 --> 21:57.350]  Alright?
[21:57.510 --> 22:01.810]  Now, I think, for all I know, the reason why DVD is so popular is because of multi-angle porn.
[22:02.590 --> 22:03.570]  You know what I mean?
[22:06.370 --> 22:11.470]  But, I'm telling you, once they figure out how to get boobies in space, we're going to have cheap launch technologies.
[22:11.990 --> 22:17.730]  Once they figure out a good way of looking at Cornell on a low-LC panel, we'll have high data right to our handsets.
[22:18.970 --> 22:19.570]  Who knows?
[22:23.210 --> 22:24.270]  So, any questions?
[22:26.090 --> 22:27.470]  I'm sorry, I made a woman leave.
[22:34.570 --> 22:39.210]  Can you comment or do you have any knowledge on the 802.16?
[22:39.610 --> 22:40.050]  802?
[22:40.050 --> 22:40.890]  Which technology is that?
[22:41.030 --> 22:41.470]  Is that 802?
[22:44.110 --> 22:49.070]  802.11bA, but it's just 16, which is supposed to be the next.
[22:50.570 --> 22:53.010]  It's 802.11b, then A, then G.
[22:53.310 --> 22:57.030]  And G is supposed to be backwards compatible with B and operating the same frequency in the range.
[22:57.030 --> 22:59.550]  I thought 16 was Bluetooth or 17 was Bluetooth.
[22:59.810 --> 23:00.750]  It just did something.
[23:01.030 --> 23:01.570]  What was that?
[23:01.690 --> 23:02.310]  Last mile.
[23:02.430 --> 23:03.530]  Last mile, 16?
[23:03.890 --> 23:04.930]  I really don't know.
[23:05.030 --> 23:06.870]  I'm not familiar with the actual protocol spec.
[23:07.430 --> 23:26.410]  And can you, I don't know if it's any of your experience or knowledge, but the difference between CDMA and GPRS, why is it still that wireless technology on a cell phone is so low with the data capacity?
[23:26.410 --> 23:27.250]  Okay.
[23:28.170 --> 23:32.870]  The data capacity, the technology is there to go very high data rates on cell phones.
[23:33.170 --> 23:37.850]  The reason why the technology is not there just yet is because of economic reasons.
[23:37.990 --> 23:39.210]  It takes time to roll these things out.
[23:39.410 --> 23:43.210]  And there's no demand for people to have high data rates on their phones.
[23:43.490 --> 23:49.250]  Again, if there's some way to get porno on your phone, then they would want it.
[23:49.250 --> 23:51.610]  I mean, it's one hand and the other hand is somewhere else.
[23:54.350 --> 23:58.010]  And for the love of God, I mean, I don't want to see someone on the subway doing that.
[24:00.650 --> 24:02.510]  Again, it's existing infrastructure.
[24:02.750 --> 24:12.470]  Whereas we have an existing CDMA infrastructure where other countries or smaller countries that are just deploying that technology will just go with the latest thing.
[24:12.470 --> 24:15.230]  Why spend all this money when we have something that already works?
[24:15.250 --> 24:19.770]  And the reason why the Japanese have rolled out this technology so fast is that their population density is so high.
[24:19.850 --> 24:23.990]  If you roll out a single 3G cell, they will be able to service far more people.
[24:24.130 --> 24:24.430]  That's one.
[24:24.550 --> 24:34.590]  And number two, the reason why you actually went from AMPS to TDMA to CDMA is you're able to squeeze three customers in for every one AMPS customer.
[24:34.590 --> 24:41.410]  And for every CDMA, you're able to squeeze three CDMA customers in practically for every one TDMA customer.
[24:41.610 --> 24:48.430]  So what happens is that for one cell site, the cell company purchased a cell site for a lot of dollars.
[24:48.750 --> 24:52.570]  They drop it into a place until the point where they'll run that cell.
[24:52.730 --> 24:59.510]  And they'll run it for as many years until the cost of making a phone call is less than the cost of electricity and the telco services that sell.
[25:00.230 --> 25:01.830]  And that's all they care about.
[25:01.830 --> 25:07.570]  And if they're able to get six customers in for a spot of every one customer before, that's all the more power to them.
[25:07.690 --> 25:09.810]  Because they're able to make, you know, it's more of a cash cow.
[25:09.930 --> 25:14.690]  And that's the reason why I really, really went from TDMA to CDMA and out of AMPS.
[25:14.750 --> 25:18.650]  And it wasn't because, you know, it's better quality or anything like that.
[25:18.810 --> 25:23.170]  And I'll be honest, I kind of think that AMPS usually, or TDMA wouldn't be, when it first was coming out, it sounded worse.
[25:23.170 --> 25:31.890]  Because, you know, you're right at the fringe of coverage and digital noise is much harder for the human ear to fix than analog of noise.
[25:46.080 --> 25:49.760]  What it is, it's, I think a CDMA is not even your own voice, it's just a vocoder.
[25:50.100 --> 25:52.580]  And basically, it's a model of your own, it's a model of your vocal track.
[25:54.100 --> 25:54.400]  And...
[25:54.400 --> 25:55.700]  Yeah, I mean, it sounds like...
[25:55.700 --> 25:56.480]  Can you repeat the question?
[25:56.480 --> 25:57.880]  The point is, you're able to...
[25:57.880 --> 25:59.120]  Oh, the question was...
[26:00.600 --> 26:04.920]  Well, it wasn't really a question, it was a comment, it said that AMPS always sounds far better than CDMA or TDMA.
[26:05.020 --> 26:11.060]  And the reason why is because, I mean, it's just, it's a vocoder and it's digitized and it's doing a reconstruction.
[26:11.820 --> 26:16.960]  That's, I mean, it's not as much for the noise issues, but it's the fact that they're able to squeeze six people in for every one channel.
[26:17.080 --> 26:23.740]  Actually, in China, the CDMA systems, they jack in far more people into a CDMA signal than they do over here.
[26:23.740 --> 26:25.340]  So it really sounds like crap over there.
[26:25.600 --> 26:27.920]  But they're able to get more, you know, more customers on it.
[26:28.100 --> 26:28.420]  I have a question.
[26:29.120 --> 26:30.580]  It's just a technical question.
[26:30.700 --> 26:30.820]  Yes.
[26:31.060 --> 26:33.960]  I have a CDMA phone from Verizon and...
[26:33.960 --> 26:34.780]  I don't know how to program it.
[26:35.280 --> 26:36.100]  No, it's not...
[26:38.900 --> 26:44.980]  I was told by Verizon that I could get 140 kilobits per second if I buy their expensive internet kit.
[26:45.620 --> 26:50.060]  And a lot of my friends disagree with me and they say that I probably couldn't get anywhere near that.
[26:50.220 --> 26:51.380]  What's your opinion on that?
[26:52.360 --> 26:53.560]  And it's a CDMA phone?
[26:53.760 --> 26:56.040]  Yeah, they said 140 kilobits per second.
[26:58.480 --> 27:00.120]  I mean, assuming I'm getting good reception.
[27:00.520 --> 27:02.760]  I didn't think that CDMA would actually go up there.
[27:02.960 --> 27:07.480]  Because, like, CDMA vocoders oscillate between 2400 baud and 28.
[27:07.920 --> 27:10.380]  I thought they might use some kind of data compression or something.
[27:11.760 --> 27:13.540]  I mean, I'm sure they could probably do it.
[27:13.640 --> 27:15.600]  Because there's enough signal space out there to do it.
[27:15.600 --> 27:17.520]  But I don't think they're going to be able to roll out to a lot of customers.
[27:18.480 --> 27:20.380]  Because there just isn't that much channel capacity.
[27:21.100 --> 27:26.140]  If you wanted to do your own over that 150 kilobits per second data connection...
[27:26.140 --> 27:26.880]  It would sound really good.
[27:26.900 --> 27:27.860]  You would have really great audio.
[27:28.160 --> 27:28.400]  Yeah.
[27:29.220 --> 27:30.140]  There's a question over here.
[27:30.360 --> 27:30.660]  Wait.
[27:31.040 --> 27:31.480]  Pass the mic.
[27:32.960 --> 27:34.620]  Then we'll send it over to this guy over here.
[27:35.760 --> 27:36.800]  I think this gentleman.
[27:36.980 --> 27:37.500]  Did you have a question?
[27:37.500 --> 27:39.120]  Yeah, I just have an answer to the previous guy.
[27:39.120 --> 27:42.980]  In Korea, they have about 10 million customers with CDMA 1X.
[27:43.180 --> 27:53.200]  And the throughput on a 144K connection is about 60 or 70 real data range, which is great.
[27:53.320 --> 27:54.800]  I mean, better than dial-up.
[27:56.160 --> 27:57.060]  So it works.
[27:57.220 --> 27:57.700]  It exists.
[27:58.320 --> 28:00.060]  Who knows if Sprint's going to make it work.
[28:00.320 --> 28:00.640]  But that's...
[28:00.640 --> 28:00.840]  Yeah.
[28:01.020 --> 28:02.780]  And I think part of it's also...
[28:03.640 --> 28:07.900]  It would work great in 99 when everyone had laptops and companies expensing their cell phones.
[28:07.900 --> 28:08.680]  But now...
[28:08.680 --> 28:11.820]  I mean, this is why everyone's like, oh, well, you know, people broke the ricochet network.
[28:12.040 --> 28:16.460]  Well, maybe part of the reason why a ricochet network doesn't exist isn't going to be successful right now.
[28:16.740 --> 28:19.960]  Because there's not a lot of people whose companies will expense a ricochet modem.
[28:21.260 --> 28:23.500]  I mean, and...
[28:24.260 --> 28:27.400]  What about the security of CDMA phone?
[28:27.600 --> 28:31.580]  Somebody says landline phone is more secure than cell phones.
[28:31.700 --> 28:34.320]  Some other people say cell phone is more secure than landline phone.
[28:34.320 --> 28:38.120]  I mean, for someone else to overhear your conversation on the phone.
[28:39.100 --> 28:43.760]  I know all these CDMA phones you can throw into sometimes the bug mode and sit there and listen to other people's conversations.
[28:44.100 --> 28:45.900]  If a cell is able to negotiate with a...
[28:46.300 --> 28:51.980]  If a cell is able to negotiate with a phone, then someone else is able to make that negotiation.
[28:52.860 --> 28:53.940]  That's really what it comes down to.
[28:54.020 --> 28:57.180]  You have to know if it's a random code, then, you know, nothing's impossible.
[28:58.280 --> 29:01.800]  Whether or not one's more secure than another, let's look at the reality of it.
[29:01.920 --> 29:02.860]  I don't think you have to worry about it.
[29:03.440 --> 29:14.540]  I mean, the probability of someone being within, you know, hearing range of your phone, which is actually pretty small, because those things are in, you know, think about 180 milliwatts max on most handsets.
[29:15.160 --> 29:24.260]  I mean, which is, it's only, like, maybe a, I don't know, half a mile radius, depending on, like, if you're in an urban environment.
[29:24.640 --> 29:30.620]  The chance of someone being in that range and listening to you while you actually read off your credit card number is, you know, pretty slim.
[29:31.160 --> 29:32.220]  I won't worry about it too much.
[29:34.720 --> 29:35.040]  No,
[29:38.500 --> 29:41.860]  I mean, but still, how many people are sitting there, you know, listening to your cell phone conversation?
[29:41.860 --> 29:45.860]  There are people who have more things to do in their lives than this.
[29:45.980 --> 29:46.760]  You'd be surprised.
[29:46.960 --> 29:49.940]  Oh, I'm surprised every day at these things.
[29:50.520 --> 29:55.810]  By the way, we're very nice of you guys.
[29:55.910 --> 29:57.370]  Probably much more than a half mile.
[29:58.710 --> 30:04.250]  Yeah, but also, then you have to sit there and, and you sit there and build all the custom hardware to sit there and demodulate it.
[30:04.450 --> 30:07.970]  And also, listen, pick your one conversation out of everyone else's.
[30:08.190 --> 30:10.070]  I'm just talking about practicalities.
[30:10.470 --> 30:10.910]  Okay?
[30:12.630 --> 30:17.430]  If someone wants to listen to your phone call, or if someone wants to follow you around, they're going to follow you around and listen to your phone calls.
[30:17.910 --> 30:18.430]  All right?
[30:18.590 --> 30:20.790]  There's not shit you're going to be able to do about it.
[30:22.790 --> 30:23.410]  Right, right.
[30:23.410 --> 30:25.450]  It's all the same transmission after the tower.
[30:26.030 --> 30:26.530]  Good question.
[30:27.410 --> 30:32.430]  Yeah, I was wondering, you were talking about CDMA phones, but what about the GSM?
[30:33.110 --> 30:34.750]  Is that any better or any worse?
[30:34.750 --> 30:35.050]  Yeah.
[30:35.050 --> 30:38.550]  And what do you think the chances are of GSM spreading in the U.S.
[30:38.690 --> 30:40.550]  as opposed to CDMA or something like that?
[30:42.690 --> 30:44.690]  It's all about 3G now.
[30:45.010 --> 30:50.550]  And because of a patent fight involving Qualcomm, the United States and...
[30:52.050 --> 30:54.050]  I want to say Japan, but I don't know.
[30:54.390 --> 30:56.090]  It's the United States and our country, I think.
[30:56.210 --> 30:57.210]  It might just be the United States.
[30:57.490 --> 30:58.750]  They're rolling out one technology.
[30:59.090 --> 30:59.830]  It's CDMA 2000.
[30:59.990 --> 31:01.650]  The rest of the world is rolling out in our technology, WCDMA.
[31:04.650 --> 31:06.970]  And until... I don't know why they're doing it.
[31:07.090 --> 31:14.490]  I think it really comes down to a patent fight, and also it's a part of American culture, that we don't have to interact with everyone else because we're in the United States.
[31:14.930 --> 31:20.370]  Until the United States really admits that it's more part of a world economy, we're not going to have a global cellular system.
[31:20.550 --> 31:21.590]  This is economics.
[31:21.790 --> 31:23.950]  It's not technology-motivated.
[31:24.170 --> 31:25.370]  It's really just a hassle.
[31:26.370 --> 31:28.850]  Sorry, it's just a hassle if you've got to go anywhere else.
[31:28.990 --> 31:31.030]  Do you have to have a GSM phone or rent one?
[31:32.730 --> 31:40.870]  I mean, yeah, but the thing is the number of Americans, or the fraction of Americans who have to travel overseas and get in our phone and do business overseas compared to...
[31:40.870 --> 31:46.390]  I mean, if that fraction starts increasing dramatically, then yes, the cell phone companies are definitely going to roll it out because there will be an economic demand.
[31:46.830 --> 31:48.570]  So how much of technology is driven by economics?
[31:48.910 --> 31:50.090]  Well, it's all driven by economics.
[31:50.210 --> 31:50.890]  That's the bottom line.
[31:52.250 --> 31:52.750]  Thank you.
[31:55.550 --> 31:55.870]  Questions?
[31:59.620 --> 32:00.400]  Is there any hardware?
[32:00.600 --> 32:01.920]  I mean, there's a lot of wireless questions.
[32:01.920 --> 32:02.220]  Good.
[32:02.460 --> 32:03.500]  I mean, I've got no problem with this.
[32:03.640 --> 32:04.820]  Are there any other questions on other topics?
[32:04.820 --> 32:06.620]  You're more than welcome to ask them.
[32:07.580 --> 32:11.160]  I have a question on 802.11b.
[32:11.500 --> 32:17.120]  I heard that you can put an amplifier in line there to increase the power output.
[32:18.200 --> 32:19.900]  How does the switching of that work?
[32:20.160 --> 32:23.920]  Does the amplifier itself do the switching between receive and transmit?
[32:23.920 --> 32:25.020]  Yeah, you would have to have...
[32:25.020 --> 32:27.940]  I'm still thinking of the old linear amplifier model from ham radio.
[32:28.120 --> 32:29.120]  That's why I'm wondering how that is.
[32:29.400 --> 32:30.740]  You don't have...
[32:30.740 --> 32:32.520]  I mean, it's not that much power.
[32:32.800 --> 32:33.980]  It's like 30, 50 watts.
[32:34.060 --> 32:34.960]  And you can do it with a pin diode.
[32:35.500 --> 32:36.320]  Oh, exactly.
[32:36.320 --> 32:44.660]  So you come into your chain, you have a pin diode and you switch over to your amplifier and you flip over to the other end and you have an LNA on a reverse pin.
[32:45.280 --> 32:46.940]  There's no need for...
[32:48.380 --> 32:50.500]  There's no need for a mechanical TR switch.
[32:50.780 --> 32:53.220]  We have really high power off electronics nowadays.
[32:53.740 --> 32:54.840]  I mean, we really have to...
[32:54.840 --> 32:56.600]  Put it this way, I'm a ham radio operator too.
[32:56.820 --> 32:59.680]  And honestly, I've worked in the cell industry.
[33:00.000 --> 33:03.780]  And the shit that the cell industry has compared to the amateur radio operators is a disgrace.
[33:04.060 --> 33:06.000]  I mean, ham radio operators are so far behind, it's not even funny.
[33:07.600 --> 33:08.100]  One sec.
[33:09.800 --> 33:13.820]  I mean, yeah, but for power levels, nowhere near.
[33:15.460 --> 33:19.180]  Even if you get the cell systems compatible, you're still going to need a different cell phone.
[33:19.280 --> 33:22.400]  The frequency allocations are different in different parts of the world.
[33:22.480 --> 33:24.520]  So your cell phone isn't ever going to be transportable.
[33:24.960 --> 33:25.940]  I'm sorry, what was that?
[33:26.620 --> 33:29.240]  You have different frequency allocations in different parts of the world.
[33:29.300 --> 33:30.460]  The frequencies are set up differently.
[33:30.820 --> 33:33.860]  Even if your system is compatible, it's going to be on a different frequency.
[33:34.140 --> 33:38.380]  There is international agreements upon frequency allocations done by the ITU.
[33:38.440 --> 33:44.460]  And it wouldn't be that much of a challenge for everyone to agree on a frequency set Because there's already the infrastructure in place for people to do it.
[33:44.600 --> 33:48.440]  So if they really wanted to have global phones, they could have them.
[33:48.480 --> 33:50.420]  You have to rewrite everybody's system.
[33:51.440 --> 33:51.840]  Question.
[33:52.840 --> 33:58.200]  I've helped a couple of friends, a few friends over the last couple of years, get like four or five DSL connections set up.
[33:58.860 --> 34:09.800]  And we've been in a constant battle with Verizon, getting the right lingo down, talking about things like bridge tabs and digital carriers to get our line clean enough to get a DSL connection on there.
[34:09.800 --> 34:18.960]  Can you provide some insight as to what a line needs to have technically in order to qualify for DSL so that we can speak the right language when we can actually get a hold of tech?
[34:19.180 --> 34:22.160]  I don't know what language do tech people speak.
[34:22.260 --> 34:23.380]  And sometimes I don't think it's English.
[34:23.820 --> 34:27.520]  What type of DSL service were you trying to get is the first question.
[34:27.700 --> 34:32.160]  Because there's many different types and they all have different line characteristics that are required.
[34:34.180 --> 34:37.260]  ADSL, SDSL, DMPDSL, ADSL?
[34:37.260 --> 34:38.080]  Yeah, that's what they often are.
[34:38.080 --> 34:41.680]  Yeah, I believe ADSL is what, 28,000 feet?
[34:42.040 --> 34:42.920]  Yeah, I think so.
[34:43.160 --> 34:51.860]  Okay, on the traditional copper phone network, you had things, load coils, bridge taps,
[34:55.060 --> 34:57.700]  really archaic, rotting cable basically.
[34:57.700 --> 35:07.980]  And all of these problems with the line can be shorted, it can be wet, it can be brought to ground.
[35:08.640 --> 35:14.400]  There's acceptance tests that the phone company will do and it can tell you if you have a bridge tap or a load coil.
[35:14.700 --> 35:22.100]  Load coils were used to try to balance the impedance for incredibly long lengths of transmission cable.
[35:41.950 --> 35:43.430]  Do you know where your CO is?
[35:43.610 --> 35:44.050]  Like how far?
[35:44.170 --> 35:44.790]  Yeah, we're real close.
[35:44.790 --> 35:46.350]  We're like about three quarters of a mile there.
[35:46.470 --> 35:47.650]  You shouldn't have any problem.
[35:47.810 --> 35:48.750]  They're giving you a hard time?
[35:48.910 --> 35:48.970]  Yeah.
[35:50.050 --> 35:50.810]  Check your house.
[35:51.450 --> 35:51.810]  Okay.
[35:52.410 --> 35:54.650]  Do you live in our old house in an apartment builder?
[35:55.390 --> 35:56.870]  No, it's four years old.
[35:57.490 --> 35:58.850]  Four years old and they're giving you a hard time?
[35:58.850 --> 36:00.650]  Yeah, this is, I've been through this four or five times.
[36:00.810 --> 36:03.410]  One time we had a digital carrier that I moved again and then it's the thing.
[36:04.410 --> 36:05.650]  It's like it's something new every time.
[36:06.030 --> 36:06.590]  Yeah, hi.
[36:06.850 --> 36:07.630]  There's no reason for that.
[36:07.810 --> 36:11.270]  Yeah, there's just no reason for three quarters of a mile that they shouldn't be able to...
[36:11.270 --> 36:11.910]  No, wait a second.
[36:12.130 --> 36:14.390]  And you should, the bottom line is this.
[36:14.750 --> 36:16.150]  You should tell them, do you want my money?
[36:16.670 --> 36:17.610]  Do you want my money?
[36:17.810 --> 36:18.610]  I'm being serious.
[36:19.090 --> 36:24.390]  They, their goal is to acquire money from you and your goal is to acquire a service.
[36:24.390 --> 36:33.330]  And you have to go up and say, look, I am willing to give you $50 a month, $80 a month, you know, whatever their fee is, in order to get high data rate.
[36:33.830 --> 36:34.490]  The end.
[36:34.730 --> 36:35.190]  All right?
[36:35.330 --> 36:36.450]  Do you want to give me that data rate?
[36:36.490 --> 36:38.430]  If you don't want to give me that data rate, I'll find some other way.
[36:38.590 --> 36:39.810]  You lose that $8 a month.
[36:40.150 --> 36:41.990]  I mean, that's really what it comes down to.
[36:42.190 --> 36:43.430]  One possible thought.
[36:43.430 --> 36:51.970]  If you're trying to get your DSL from a company other than your local Bell operating company, a lot of times they will try to make it difficult for you and claim they can't do it.
[36:52.090 --> 36:57.350]  But if you then contact them and tell them you want it from them, they magically find a way to make the line work.
[36:58.070 --> 37:00.070]  That happens a lot with Verizon, at least.
[37:00.230 --> 37:01.150]  I think other ones too.
[37:02.170 --> 37:02.850]  They're notorious.
[37:03.170 --> 37:03.370]  Keep that in.
[37:03.370 --> 37:07.930]  Even if it does, again, you know, they call it the Better Business Bureau.
[37:07.930 --> 37:10.650]  If you really want to make these people miserable, you can do that.
[37:11.550 --> 37:12.090]  Freak out?
[37:14.270 --> 37:17.210]  We spent a lot of time on DSL and wireless talk.
[37:17.350 --> 37:18.690]  Is there any hardware questions?
[37:19.150 --> 37:22.350]  Actually, we all have these nifty little smart card branches.
[37:22.510 --> 37:22.910]  There's nothing in them.
[37:24.570 --> 37:26.090]  But tell us about them.
[37:26.270 --> 37:27.770]  What can you, you know, do with them?
[37:28.130 --> 37:33.030]  What household electronics can I use to peer inside this and put stuff on it?
[37:33.210 --> 37:34.190]  Do something with it.
[37:34.470 --> 37:35.490]  What household electronics?
[37:36.150 --> 37:40.650]  There's been some work to do where they try to read the memory cells by shaving them down.
[37:41.670 --> 37:44.950]  The people who design these things aren't dumb shits.
[37:45.590 --> 37:49.210]  I mean, no, they actually put a lot of effort into trying to make these things secure.
[37:49.350 --> 37:53.730]  I remember some of the initial ones, they tried photocopying them to try to read the memory cell.
[37:53.810 --> 37:55.010]  They'd self-destruct.
[37:55.490 --> 37:56.750]  There's a lot of different things that have happened.
[37:56.870 --> 38:02.750]  There's a report that came out about this guy in Britain who was able to shave down slightly and use a photo flash to read the memory cell structure.
[38:02.750 --> 38:05.870]  Yeah, that was not, like, really repeatable.
[38:06.050 --> 38:07.370]  He said you could do it within a batch.
[38:07.690 --> 38:12.710]  Like, you could destroy several hundred in the process of one batch trying to read the memory cell structure.
[38:12.890 --> 38:14.950]  So they're moderately secure.
[38:15.230 --> 38:22.550]  As in, I mean, common household structures to access them, there's a really good paper on how to attack these things.
[38:22.550 --> 38:27.730]  And basically looked at external characteristics, such as you examine supply voltage used.
[38:27.870 --> 38:32.610]  So let's say before it authenticates, most of the circuit's off and may draw five milliamps.
[38:33.130 --> 38:40.050]  And as you try to authenticate against the card, until it authenticates, you know, it'll still draw five milliamps.
[38:40.190 --> 38:42.950]  And once it authenticates, it draws 20 milliamps, for example.
[38:43.370 --> 38:45.130]  So what you do is just watch the power structure.
[38:45.290 --> 38:48.270]  And I think that's one of the more creative ways of going about accessing these things.
[38:49.850 --> 38:50.690]  There's one in the back.
[38:50.830 --> 38:50.950]  All right.
[38:51.050 --> 38:52.770]  And then we'll come up here, then over here.
[38:53.330 --> 38:53.770]  Oh, yes.
[38:53.870 --> 38:55.070]  Just a small hardware question.
[38:55.210 --> 38:55.310]  Yeah.
[38:55.770 --> 39:00.110]  Does anybody here know where I can find a 27C512 Amtel chip?
[39:01.130 --> 39:01.490]  27C?
[39:01.490 --> 39:02.230]  Have you tried Mouser?
[39:02.630 --> 39:03.070]  Or DigiKey?
[39:03.070 --> 39:03.430]  Not yet.
[39:03.770 --> 39:04.590]  1-800-DigiKey.
[39:07.390 --> 39:07.750]  All right.
[39:07.750 --> 39:09.070]  DigiKey will deliver the next day.
[39:09.510 --> 39:09.790]  Oh, yeah?
[39:10.110 --> 39:10.250]  Yeah.
[39:10.590 --> 39:10.710]  Good.
[39:15.170 --> 39:18.710]  I had good luck ordering chips from Radio Shack.
[39:18.810 --> 39:19.030]  Yeah.
[39:19.230 --> 39:19.870]  No, just go with DigiKey.
[39:19.870 --> 39:20.470]  Really obscure ones.
[39:20.710 --> 39:22.990]  Mouser and DigiKey are an engineer's best friend.
[39:24.690 --> 39:25.130]  There.
[39:25.690 --> 39:27.250]  Also, try calling for samples.
[39:27.890 --> 39:28.710]  Say that you're developing...
[39:29.450 --> 39:30.770]  Right over here, this gentleman.
[39:31.490 --> 39:32.770]  Bend back that way.
[39:33.210 --> 39:38.450]  If you're saying that you're doing a large spin of systems, saying, I'm such-and-such engineer for such-and-such company.
[39:38.550 --> 39:39.490]  We're developing this product.
[39:40.110 --> 39:41.750]  I want to get some samples of this device.
[39:41.830 --> 39:42.970]  We're comparing it to this and this.
[39:44.230 --> 39:44.610]  You know.
[39:45.590 --> 39:47.030]  And usually what happens is they'll send it to you.
[39:47.090 --> 39:49.970]  Then you'll get some phone calls from the sales staff and say, oh, we chose not to use it.
[39:50.630 --> 39:51.570]  But no, it makes...
[39:52.010 --> 39:53.610]  No, I mean, this is...
[39:53.610 --> 39:54.970]  You had to do this in the industry.
[39:55.050 --> 39:55.890]  This is how people operate.
[39:55.990 --> 39:57.110]  Because you have to get the components fast.
[39:57.210 --> 39:58.190]  You can't wait for the...
[39:58.190 --> 39:59.350]  You know, it's not to turn them over.
[39:59.490 --> 40:00.830]  But DigiKey is really fast with stuff.
[40:01.930 --> 40:06.790]  What device or the hardware I need for recording my own cell phone conversation?
[40:07.150 --> 40:08.690]  Recording your own cell phone conversation?
[40:08.850 --> 40:08.990]  Yes.
[40:09.010 --> 40:10.470]  It depends on what you're working with.
[40:10.690 --> 40:13.210]  I mean, if you're working with amps, just a scanner.
[40:13.390 --> 40:16.350]  If you're working with TDMA, there's an Ericsson phone.
[40:16.590 --> 40:18.570]  I think the Ericsson web phone, there's a way of...
[40:20.410 --> 40:21.850]  There's a way you can run into bug mode.
[40:21.890 --> 40:22.770]  You can listen to each channel.
[40:23.290 --> 40:26.470]  If you're working with TDMA, the specialized equipment.
[40:27.290 --> 40:28.290]  They test that, yeah.
[40:28.550 --> 40:28.730]  All right.
[40:28.730 --> 40:29.510]  There's a question back here.
[40:41.090 --> 40:46.730]  I've heard that x86 has some CPU instructions that just didn't get used over the years.
[40:46.830 --> 40:47.730]  What exactly is that?
[40:48.950 --> 40:53.750]  That could be, if they had instructions that haven't been used over the years, that could be because of the compiler designers.
[40:55.310 --> 40:57.470]  I mean, that's really what it comes down to.
[40:57.470 --> 41:03.590]  The compiler, if you think about it, in a way, the architecture people define what instructors are capable, like MMX, for example.
[41:03.730 --> 41:04.870]  I remember when MMX came out.
[41:05.030 --> 41:07.950]  It was like, oh, it was going to be a big screen increase.
[41:08.290 --> 41:12.370]  But until the compiler designers develop it, it goes underutilized.
[41:13.290 --> 41:17.390]  Another good example is the G4 has the Altevec processor, which I think is so badass.
[41:17.830 --> 41:23.750]  And it's basically a vector processor unit, which is something you didn't find unless you got the really high-end processors, usually.
[41:24.270 --> 41:28.350]  And until people write good compilers to utilize them, they'll go underutilized.
[41:29.670 --> 41:31.150]  This gentleman did not ask a question yet.
[41:31.350 --> 41:32.850]  On your left.
[41:38.410 --> 41:39.130]  Is this on?
[41:39.330 --> 41:39.590]  Hello?
[41:39.830 --> 41:39.970]  Okay.
[41:39.970 --> 41:45.730]  In terms of processing, based on level 2 cache...
[41:45.730 --> 41:47.650]  In terms of processing based on...
[41:47.650 --> 41:56.550]  In terms of processing, with NEX 86 processor, what's the highest level amount of level 2 cache have you seen?
[41:56.810 --> 42:01.330]  Because I've seen stuff like, for example, on the G4s, I think they use one or four megs.
[42:01.970 --> 42:06.290]  And everybody on the PC process, you know, uses anywhere 256, 384, 512.
[42:06.630 --> 42:09.610]  I've seen meg cache, and I think I've seen two meg cache.
[42:09.750 --> 42:09.850]  I don't remember.
[42:09.970 --> 42:11.650]  I know that this... I'm more of a Spark guy.
[42:12.270 --> 42:14.010]  And I wear Spark boxes of eight meg cache.
[42:14.750 --> 42:15.130]  What is it?
[42:15.130 --> 42:16.130]  What is it?
[42:19.450 --> 42:23.730]  This guy is saying that it's three meg Intel cache, and I mean, it's...
[42:24.850 --> 42:29.970]  I mean, for most people, until the customer demands it, you know, it's not going to appear.
[42:30.610 --> 42:35.410]  But for most people, most people don't use X86 hardware for scientific computing, so...
[42:35.410 --> 42:36.970]  Or for having number crunching.
[42:37.430 --> 42:40.770]  And most of their number crunching nowadays is done with video card because they play video games.
[42:41.150 --> 42:43.990]  So, you know, it's not as much of a concern for most people.
[42:43.990 --> 42:47.550]  And the people that demand it are, like, people do scientific processing or stuff like that.
[42:47.690 --> 42:48.610]  And that's where you see it.
[42:48.950 --> 42:52.370]  Like, on some V880s, you usually have a lot of cache per processor.
[42:52.830 --> 42:53.110]  Back?
[42:53.650 --> 42:53.790]  Yeah.
[42:54.030 --> 42:56.370]  I was wondering what your thoughts were on CG.
[42:58.550 --> 43:04.150]  NVIDIA's new compiler for, you know, using, like, you know, vertex and pixel shaders.
[43:04.310 --> 43:13.450]  And it's like a C-like language that's, you know, since it doesn't have to be programmed for specifically, you know, like, ATI cards or, you know.
[43:13.450 --> 43:15.010]  It's an industry thing.
[43:15.150 --> 43:20.470]  I mean, what they're doing is they're basically saying, we're going to write a language so more people will use our hardware.
[43:21.470 --> 43:23.790]  I mean, make people more comfortable using our hardware.
[43:23.890 --> 43:24.950]  That's pretty much what it comes down to.
[43:26.070 --> 43:29.570]  I mean, whether or not it's a video card or things, I'll be honest with you.
[43:30.670 --> 43:32.270]  I don't really play video games.
[43:33.510 --> 43:35.530]  I play some Yahoo games about it.
[43:36.210 --> 43:40.050]  But I get motion sick from, like, the 3D things.
[43:40.570 --> 43:47.590]  But, I mean, really why they're coming out of those compiler things is because they're trying to make it easier for programmers to write stuff for their video card.
[43:47.810 --> 43:51.690]  And if people write more software for their video card, people are going to buy more of their video card.
[43:52.850 --> 43:53.250]  So...
[43:53.250 --> 43:55.690]  I've just been told you have time for one more question.
[43:56.670 --> 43:58.090]  Someone who hasn't asked anything yet.
[43:59.150 --> 44:07.490]  Can you comment on the Tempest technology of remotely monitoring a monitor's output, the feasibility of building a custom system?
[44:08.910 --> 44:11.010]  Of building your own custom Tempest system?
[44:11.390 --> 44:12.370]  I'm sure it's possible.
[44:12.810 --> 44:15.010]  I mean, I don't know how much time you want to spend on it.
[44:15.390 --> 44:17.310]  But it's a lot of optimization stuff.
[44:17.590 --> 44:26.990]  I mean, there was an old Heathkit computer that you were able to put a TV next to it and you would see the exact same thing on the TV as you would on the computer.
[44:27.290 --> 44:33.410]  Because the V-Sync and the H-Sync bled so much, it would bleed into the TV next to it.
[44:35.710 --> 44:38.210]  Doing it on your own, I don't know, it's a good challenge.
[44:42.430 --> 44:44.250]  Let's just get this one last question here.
[44:45.590 --> 44:46.390]  Do you have the microphone?
[44:49.030 --> 44:52.930]  It sounded like you talked about a differential power analysis with the smart cards.
[44:52.930 --> 44:53.330]  Yeah.
[44:54.330 --> 44:56.430]  Where you examine and extract the key.
[44:57.030 --> 44:57.810]  Are you aware of...
[44:57.810 --> 45:00.910]  It's not only extracting a key, it's just a brute force attack against the key.
[45:01.350 --> 45:01.530]  Okay.
[45:01.850 --> 45:01.970]  Yeah.
[45:02.270 --> 45:07.990]  But developers are encountering that with running like a random number generator while they're doing it?
[45:08.070 --> 45:08.670]  Are you aware of...
[45:08.670 --> 45:12.910]  Yeah, they could do that or they could put like a load resistor across it.
[45:13.230 --> 45:14.410]  I mean, it's really a cat and mouse game.
[45:14.910 --> 45:15.030]  Yeah.
[45:15.030 --> 45:15.690]  And it's kind of cool.
[45:16.030 --> 45:18.810]  Are you aware of any other emerging attacks?
[45:19.550 --> 45:19.910]  No.
[45:20.050 --> 45:24.390]  The last thing I heard was someone doing an attack with shaving off the surface and using a flash.
[45:24.690 --> 45:27.150]  That's not something you attack a specific card with.
[45:27.230 --> 45:30.870]  You attack a make of cards to try to reverse engineering structure.
[45:31.950 --> 45:33.630]  Off the top of my head, I don't know anything new.
[45:33.810 --> 45:47.570]  If you can cut out the back of the card and dissolve it with metric acid, if you can gain access to like a research lab and use a scanning electron microscope to read the membrane.
[45:48.750 --> 45:49.310]  You can.
[45:49.470 --> 45:49.810]  I mean, no.
[45:49.930 --> 45:50.730]  I think it...
[45:50.730 --> 45:54.910]  I have a feeling by the time you read the first couple bits, the last couple bits are going to start and destroy it from the radiation.
[45:56.630 --> 46:00.930]  I mean, because the thing about it, you're bombarding it with huge amounts of electrons.
[46:01.310 --> 46:04.250]  So you might destroy more data.
[46:04.430 --> 46:06.690]  You might destroy some of the data before you actually get any of it out.
[46:08.610 --> 46:09.910]  I think that's all we have time for.
[46:10.590 --> 46:10.990]  Oh.
[46:12.590 --> 46:14.590]  Your DSL question to get...
[46:14.590 --> 46:18.370]  If you ask Verizon for a pair change, complain about your modem data rate.
[46:18.570 --> 46:22.510]  They should give you at least one pair change and try to get your DSL provision for that.
[46:23.150 --> 46:24.150]  We should open that.
[46:24.150 --> 46:24.290]  We should open that.
[46:24.550 --> 46:25.170]  Thank you.
[46:25.330 --> 46:25.810]  Thank you.