While it's true that the article hardly touches on the interesting aspects of these teams' design choices, I did find the closing somewhat interesting- the contrast that one team concentrated all their money/effort on the sensors and control of a normal car, while the other team concentrated their money/effort on a vehicle with bad-ass off road capability and didn't worry as much about the sensors and driving systems.
Who would I put my money on? Well, from all the pictures I've seen of the course, I'd bet that I could significantly kick the ass of all these autonomous vehicles (even if they perform two orders of magnitude better than they did last year) driving my '97 Subaru station wagon. On the other hand, whether you're driving a dune buggy, a Humvee, a Sherman tank, or NASA's space shuttle launch platform, boulders, mountains, canyons, underpasses, and rivers will put a real crimp in your style if you don't navigate well. Thus, I think that control systems are where it's most worth concentrating one's effort.
You are confusing latency and bandwidth, which are completely different things. Latency is the time between stimulus and response; bandwidth is the data transfer rate.
Saying "your fingers can handle a lot more latency than 250 Mbps" is nonsensical- 250Mbps is a measure of bandwidth, not latency. I couldn't find any statistics on a latency difference between USB and Firewire interfaces, and any latency either has is going to be something in milliseconds that will never be apparent to humans no matter what the application. External drives that have dual Firewire/USB interfaces don't even bother to quote different latency specs for the different interfaces, it's so close to identical.
As I said, Firewire had greater bandwidth, so if you needed to move a lot of data in real time, then that was an advantage. Firewire became the standard for video instead of USB because USB 1.1 didn't have enough bandwidth to handle a DV stream, and it probably helped stop USB2 from taking over later that Firewire was designed specifically to handle a DV video stream and has great protocols for doing so. USB2 could probably do as good of a job- the latency's effectively the same as Firewire, and the bandwidth is competitive. You can find many pages online testing, measuring, and debating the merits of Firewire and USB2 for various real-time uses, like MIDI. Note, this article on MIDI latency doesn't even mention the latency of USB and Firewire, only the read/write speeds- the bandwidth, because the latency of the interfaces is irrelevant. USB2 actually wins the realtime data transfer test in their comparison because it achieves faster write speed. If you look around, there are a lot of other real-world tests online showing USB2 and Firewire to have similar bandwidth, and the latency of the interfaces isn't even an issue.
Again, your division of tasks with non-realtime using USB and realtime using Firewire is a coincidence of the two things you pointed out. Plenty of realtime applications are done through USB, and plenty of tasks that aren't time sensitive are done through Firewire. I could as easily switch your sentence around to say "Which is why non-real time tasks like tape backup drives use Firewire, while real-time webcam video uses USB." You can get webcams, printers, hard drives, and all sorts of things with either interface or both. Firewire rules video transfer for the reasons I've mentioned, and USB rules keyboards and mice because USB chips were much, much cheaper than Firewire chips a few years ago. Neither ever had anything to do with latency, and neither has anything to do with current bandwidth differences between Firewire and USB2.
"it isn't used for keyboard and other areas where real-time performance isn't critical"
Funny, I've always had a strong preference for keyboards that work in real-time. Rather than, say, typing a paper now, and then coming back later to see if the computer's finished importing it so I can look at it. In fact, some people even like to use keyboards as part of a real-time interface for controlling their computers.
Firewire's advantage isn't it's ability to work in real time; plenty of real-time devices like keyboards, mice, webcams, and networking run on USB. It's initial advantage for video was bandwidth, which stopped USB1 from doing real-time DV video, and a good protocol for handling realtime video, including remote controlling a video source.
Never mention to anyone that they sell fake watches on Canal street in New York.
Never tell anyone that there are drug dealers in the park down the street, even for their own safety.
You had also better never report a crime to authorities. That is also providing information on how to locate illegal activity.
Someone should print out the web address of a stolen copyrighted work that's freely available online, go into a court house in Australia, and stick it to a bulletin board. Then they should sue the government for hosting that information, citing this case as precedent.
I basically agree with everything you said, but thought I'd clarify a couple of points:
"a $99 Epson inkjet that uses Ultrachrome inks"
I think the cheapest Ultrachome printer is the R800 at $400 MSRP. Not a bad deal, and they do have Durabrite printers with inferior but still good longevity for $100.
"For home and business photo printing there really isn't an alternative to inkjet besides dye sublimation"
I agree that there's no viable alternative to inkjet for home photo prints except dye-sub, but for businesses with a lot of money, there are alternatives. There are digital photoprints like the Durst Lambda, Epsilon, and Zeta, the Oce LightJet, or the ZBE Chromira, the Fuji Pictography, and many more. There are also electrostatic printers like the IRIS/IXIA printers. Then there's Xerox's solid-ink printers, which are still technically piezoelectric inkjets, but are pretty different functionally from what most people mean when they talk about inkjets.
Everyone says "video iPod," but I think they have the wrong thing in mind. Jobs has said before, and I agree with him, that mobile video playback just isn't a killer app. People want to listen to music in their car, working out, walking around, working- everywhere they don't want to watch TV and movies. People want to watch their movies on their giant home theater setups. Maybe a few people who take public transportation want to watch downloaded video on their iPods, but the potential market for portable video just isn't worth designing a product line to go after. Sure, if you can plop it into your existing product as a software addition, like the PSP (and probably future iPods), you might as well, it may be handy. But it's not a killer app.
What might be a killer app to design a video iPod around is the DV (or HD) camcorder. Clip your iSight onto your iPod. Now you have a camcorder that's smaller than any other on the market and records approximately forever, strait to hard disk, no messing with tapes. Maybe in H264. I think that's what a "video ipod" is going to be.
Have and iPod Video and want an HD camcorder? It'll cost a heck of a lot less than buying a DV camcorder, all you need is the iSight, which, by the way, you can still use as a webcam. Want to upgrade to an HD camcorder? Instead of giving Sony another $1000 to replace your DV camcorder with HD, pay Apple a quarter as much for their new HD iSight and plug it into your existing iPod Video.
You have a couple of typos. The last sentence should read "It's a bad thing city engineers don't have to make a profit and aren't thrown out on their asses."
Although, perhaps if government actually got out of their way and let them do their jobs, while also holding them to the standards non-government employers insist upon, they'd manage to get their jobs done on time and within budget instead of getting thrown out on their asses?
What's with the mice they're talking about? I feel totally lost, the poll says "Top Mice compared," but I've never heard of any of them. If I were doing this, I'd have compared:
Email addresses aren't a good representative for all text. I may only have four or five email addresses memorized (although that's due in good measure to the fact that my email program knows them all and associates them with names and nicknames in its own DNS-like system), but I have hundreds of website addresses memorized, compared to only a couple of dozen phone numbers. Look at how many people's names a person knows, or at their vocabularies, compared to how many numbers they have memorized. I stand by the idea that words and names are easier to memorize than long strings of numbers.
Email addresses are bad, and I think there are a number of reasons for that. Primarily, they are almost always at some arbitrary seeming domain- uchicago.edu, columbus.rr.com, insighttelco.net. etc. These are hopeless to remember. Also, a lot of email addresses are assigned, or chosen poorly. For example, at Chicago, it was first initial, middle initial, first 5 letters of last name. How stupid is that to remember, among each of a bijillion other arbitrary systems?
But if you choose your own, and don't have any domain (or the domain for all of them is "phone" or something), then you don't have that problem.
This is neat, but they're going the wrong direction. Memorizing a 10-digit number instead of the word "Google" isn't making things much easier for people
I want the opposite: I want my phone number to be arbitrary text of my choosing. I want the whole phone system to use good voice recognition, so that any time I want to call anyone, if I can remember their phone-moniker, all I have to do is say it into the phone. A DNS-like system will go reference this to their actual phone number, IP address, or whatever address and protocol they choose to use for voice communications, and connect me.
They used radically different image sets for different areas, and it's available at different resolutions for different areas.
Check out some of the radical boundaries around Michigan.
Mapquest used to have satellite and USGS topographical map data for the whole US, and the top of every map had tabs letting you switch between these or their street maps, maintaining the same view and scale. Mapquest also used to let you set your preferences to view much larger pixel count maps. Why they removed all these features is beyond me.
Yeah, that's the first thing I noticed. If there's a company that does 100 projects per year, and delivers 99 of those on time, then one project is late, so that company goes in the "delivers projects late" pile.
Why not have a headline "95% of IT employees will lose their jobs this year," based on an article showing that 95% of IT firms will terminate at least one employee this year.
While highly plausible, this 95% statistic is worthless, and the headline is not at all supported by the article. Actually knowing what percent of IT projects are late might be interesting.
There used to be a time when people wrote software at home on legal pads and tried to debug it in their heads before driving somewhere where they could punch it out onto tape for a teletype to send over a modem to a remote time-sharing computer. Now everyone expects to be able to sit in front of a computer and interact with it while writing and debugging code. Wimps.
Oh, but wait. Less than a hundredth of a percent of the population was interested in using a computer that way.
As long as Linux is happy to be a niche toy for geeks and computer professionals, then by all means, defend keeping it the kind of system that you have to go to school for computer science in order to use. That's what Marcus Ranum was saying will happen if it stays so diverse; it will maintain the kind of market share numbers you'll see if you stick to CS majors as your potential user base.
Most PC and Mac users don't require a formal education in computers to do their email, word processing, and web browsing. If Linux has the ambition of gaining serious market share, it needs to adopt a lot of the silly user friendliness of the competition, rather than complaining that too few people are willing to take computer classes so that they can use it.
OK, I learned some math. Here's a little math lesson.
Inelastic goods tend to have a lump sum, or flat tax (as opposed to a flat rate tax). Cigarettes, in every state in the US, are taxed by the pack. And guess what, gasoline? Taxed by the gallon And holy crap, look at this, I'm looking at my SBC local phone bill right now, and the tax is charged by the line. A portion of my long distance tax charges is actually done as a percentage, but guess what, they're changing that to a flat rate, too. And hang on, I've got my Adelphia broadband bill here, and it's got a flat rate monthly tax too!
So in your example, where they sell twice as much at 1/3 the price, tax revenue would be exactly... twice as much. So I'm looking forward to the second insallment of your math lesson where you contue to explain how I'm such an idiot for thinking that if people buy more of these things, the tax revenue will increase, and enlighten me as to how it will "definitely be less." Of course, without knowing the supply and demand curves, you can't know if it would be more or less even if it were a percentage tax, and in almost all cases, you're wrong, because total sales almost always increase when prices drop. But that's another story.
So, instead of spending $50 on a private ISP, if they spend $25 of their taxes on "free" government service
Try supposing it's $100 for the government's service, and that it ends up going out all the time, being a fifth the speed of the $50 private service, and if you think tech support is lousy these days, imagine having to drive somewhere and stand in line for 5 hours just to have them tell you they don't care and won't help.
According to IDC, 5 million americans have a wifi card now. That's about 1.6% of the population. Of course, not all of those are 802.11 b to work with these networks, but let's round up in the government's favor. Philadelpia, the first US city to try this, has a population of about 1.5 million. Philadelpia's spending $10 million on setup and expects operating costs of $1.5 million. Thus, if Philadelphians own wireless cards at about an average rate, then about 24,000 of them have cards. So for the first year, they're spending $458 of tax money per resident who could even possibly try to make use of the service. Of course, this is just to put one wireless hub on each block; what pecent of people do you think could actually get service without leaving thier home? Our 802.11 hub only reaches some rooms in our house, and it's base station is right here. The people installing the networks admit that the base stations only reach about 100 feet, and that's when they aren't going through brick walls. So I wouldn't be surprised if their first year cost is closer to $4,000 per regular user. But other people will be paying that money, instead of them paying $50 for their own access, so it's good, right? Of course, more people will buy 802.11 cards in the future if there's "free" broadband available, so the numbers should improve, if the government can keep program costs under control.
Just out of curiosity, do you think those 24,000 out of 1,500,000 who have computers with 802.11 are among the poor? How many do you think have incomes at least two standard deviations above the mean? You do realize this is a tax on everyone, including the poor, to provide services overwhelmingly consumed by the rich, so they don't have to pay the fee themselves?
"So, instead of spending $50 on a private ISP, if they spend $25 of their taxes on "free" government service, they would have mo
I don't quite follow your point as to "why" there is no free market in telecom. OK, so it's a service with inelastic demand, like cigarettes and gasoline (in the short run). So as goods with low demand elasticity, cigarettes and gasoline get really high taxes; but that's no reason for the government to give them a monopoly. In fact, in economic terms, if the government is trying to maximize tax revenue, then they want a competitive market, especially for inelastic goods with giant taxes on them. Competition brings down prices and increases total sales, and every time there's a sale the government gets their huge tax. The government doesn't care if you go to BP or Texaco, if you smoke Marlboros or Virginia Slims. They get their taxes either way. So why the monopolies in telecom?
And I don't see how any of this applies to your last point, "if you believe for a second it's about being nice to you and giving you free Internet, you're the biggest sucker out there. Government and big business is a serious sport. Wake up and look at who's putting the money behind the efforts you're idealistically supporting." Well, government is putting money behind municipalities providing broadband as a government utility. If governments collect taxes to pay for the service, then give the service away for "free" on the margins, that will prevent them from being able to collect a tax on it. If they're giving it away for free, they not only can't tax it, they have to use other tax revenue to subsidize it. Also, by offering it at no marginal cost, people will discontinue their paid DSL and Cable broadband service (which they pay taxes on), and switch to the "free" government service (which they're already paying for). Thus providing free broadband will erode their existing tax base!
So I disagree with your insinuations regarding the motives of those promoting municipal broadband. They probably are acting for their own self interest, as they understand it. But the locals are probably for it because they don't understand that when the government provides something for "free," it's likely to cost them more in total, and be lower quality. The government backers just want more programs to administrate; it makes more jobs to hand out, more "good" they can say they're doing for the community, more accomplishments for their resume, more media coverage, larger budgets, more people to supervise, all leading to more pwoer, prestige, and higher salaries for themselves. Basically, all the normal incentives for government to expand.
Lawrence Lessig argued this in the latest Wired- that free markets were failing to provide decent service, and that the government can provide services better and cheaper than markets can.
This is my reply:
RE: Why Your Broadband Sucks
Perhaps Lawrence Lessig should stick to law, rather than demonstrating his ignorance in economics. Lessig starts with "city leaders stepped in where the free market had failed." What free market? The government given monopoly in cable and telephone? Where in the US is there anything approaching a free market in broadband? Does Lessig believe that Taiwan is so far ahead of the US because US businessmen can't compete with their Taiwanese counterparts in broadband? He calls government-supplied telecommunications "free." Most government services cost more than free-market equivalents; the bill just comes in taxes.
There is a problem with all of Lessig's comparisons to other government functions: he never claims that government does them better. Simply doing something is no argument for doing other, similar things. Congress passed the DMCA and the Mickey Mouse... I mean Sony Bono Copyright Extension Act. Does Lessig believe they should therefore pass more similar legislation? These comparisons with other government services either compare apples to oranges, or don't support his case. Street lights and roads are textbook examples of "free rider" problems. The difficulty of charging for street light usage probably outweighs the efficiencies of market provision; but billing for internet access is easy. As for busses and water, studies indicate that, ceteris paribus, markets provide them better, making a counterexample to his point.
Closer comparisons to government broadband are Soviet industry, or the nationalized Jaguar. Real costs are higher than those of competitive services, quality is worse, and innovation stops. Sensible deregulation brings lower prices, higher quality, and more options.
The worst fallacy is the conclusion: "let the markets, both private and public, compete" What public markets? What does "competition" mean when your "competitor" forces every customer to pay them any price, makes rules for you, and taxes your income to fund themselves? This is like the "competition" between a mugger and his victim. What do you think would happen to FedEx if the USPS was "free?" Perhaps not all innovation would stop. Like public schools, public internet access might become so bad that high quality services will serve the rich elites who can afford them, leading to the opposite of his egalitarian goals.
If we want to see the US plunge from 13th on the list down to wherever Cuba is, then by all means, let the government "compete" in the telecommunications industry. If you want to stand up to the "self-serving lobbyists," stand against their government-granted monopolies, not against saving themselves and us from socialized industry.
Umm, wouldn't this make a better Poll Topic than an Ask Slashdot? Is someone going to tabulate all the responses?
You still get the discussion with the poll, but you also get the raw data, which might add something of interest to the collection of posted anecdotes, each explaining where one individual feels more productive.
Actually, it's all part of their evil plan. They will copyright silence, then refuse to sell that track anymore, and then sue anyone who's NOT listening to music, as they are listening to silence, which is copyrighted, and hasn't been legally purchased.
Therefore, you must always listen to music to avoid fines. In this way, they will sell more music.
Slashdot Mirror
Who would I put my money on? Well, from all the pictures I've seen of the course, I'd bet that I could significantly kick the ass of all these autonomous vehicles (even if they perform two orders of magnitude better than they did last year) driving my '97 Subaru station wagon. On the other hand, whether you're driving a dune buggy, a Humvee, a Sherman tank, or NASA's space shuttle launch platform, boulders, mountains, canyons, underpasses, and rivers will put a real crimp in your style if you don't navigate well. Thus, I think that control systems are where it's most worth concentrating one's effort.
Saying "your fingers can handle a lot more latency than 250 Mbps" is nonsensical- 250Mbps is a measure of bandwidth, not latency. I couldn't find any statistics on a latency difference between USB and Firewire interfaces, and any latency either has is going to be something in milliseconds that will never be apparent to humans no matter what the application. External drives that have dual Firewire/USB interfaces don't even bother to quote different latency specs for the different interfaces, it's so close to identical.
As I said, Firewire had greater bandwidth, so if you needed to move a lot of data in real time, then that was an advantage. Firewire became the standard for video instead of USB because USB 1.1 didn't have enough bandwidth to handle a DV stream, and it probably helped stop USB2 from taking over later that Firewire was designed specifically to handle a DV video stream and has great protocols for doing so. USB2 could probably do as good of a job- the latency's effectively the same as Firewire, and the bandwidth is competitive. You can find many pages online testing, measuring, and debating the merits of Firewire and USB2 for various real-time uses, like MIDI. Note, this article on MIDI latency doesn't even mention the latency of USB and Firewire, only the read/write speeds- the bandwidth, because the latency of the interfaces is irrelevant. USB2 actually wins the realtime data transfer test in their comparison because it achieves faster write speed. If you look around, there are a lot of other real-world tests online showing USB2 and Firewire to have similar bandwidth, and the latency of the interfaces isn't even an issue.
Again, your division of tasks with non-realtime using USB and realtime using Firewire is a coincidence of the two things you pointed out. Plenty of realtime applications are done through USB, and plenty of tasks that aren't time sensitive are done through Firewire. I could as easily switch your sentence around to say "Which is why non-real time tasks like tape backup drives use Firewire, while real-time webcam video uses USB." You can get webcams, printers, hard drives, and all sorts of things with either interface or both. Firewire rules video transfer for the reasons I've mentioned, and USB rules keyboards and mice because USB chips were much, much cheaper than Firewire chips a few years ago. Neither ever had anything to do with latency, and neither has anything to do with current bandwidth differences between Firewire and USB2.
Funny, I've always had a strong preference for keyboards that work in real-time. Rather than, say, typing a paper now, and then coming back later to see if the computer's finished importing it so I can look at it. In fact, some people even like to use keyboards as part of a real-time interface for controlling their computers.
Firewire's advantage isn't it's ability to work in real time; plenty of real-time devices like keyboards, mice, webcams, and networking run on USB. It's initial advantage for video was bandwidth, which stopped USB1 from doing real-time DV video, and a good protocol for handling realtime video, including remote controlling a video source.
Never tell anyone that there are drug dealers in the park down the street, even for their own safety.
You had also better never report a crime to authorities. That is also providing information on how to locate illegal activity.
Someone should print out the web address of a stolen copyrighted work that's freely available online, go into a court house in Australia, and stick it to a bulletin board. Then they should sue the government for hosting that information, citing this case as precedent.
"a $99 Epson inkjet that uses Ultrachrome inks"
I think the cheapest Ultrachome printer is the R800 at $400 MSRP. Not a bad deal, and they do have Durabrite printers with inferior but still good longevity for $100.
"For home and business photo printing there really isn't an alternative to inkjet besides dye sublimation"
I agree that there's no viable alternative to inkjet for home photo prints except dye-sub, but for businesses with a lot of money, there are alternatives. There are digital photoprints like the Durst Lambda, Epsilon, and Zeta, the Oce LightJet, or the ZBE Chromira, the Fuji Pictography, and many more. There are also electrostatic printers like the IRIS/IXIA printers. Then there's Xerox's solid-ink printers, which are still technically piezoelectric inkjets, but are pretty different functionally from what most people mean when they talk about inkjets.
What might be a killer app to design a video iPod around is the DV (or HD) camcorder. Clip your iSight onto your iPod. Now you have a camcorder that's smaller than any other on the market and records approximately forever, strait to hard disk, no messing with tapes. Maybe in H264. I think that's what a "video ipod" is going to be.
Have and iPod Video and want an HD camcorder? It'll cost a heck of a lot less than buying a DV camcorder, all you need is the iSight, which, by the way, you can still use as a webcam. Want to upgrade to an HD camcorder? Instead of giving Sony another $1000 to replace your DV camcorder with HD, pay Apple a quarter as much for their new HD iSight and plug it into your existing iPod Video.
Although, perhaps if government actually got out of their way and let them do their jobs, while also holding them to the standards non-government employers insist upon, they'd manage to get their jobs done on time and within budget instead of getting thrown out on their asses?
- Algernon
- Mrs. Brisby
- Mighty
- Mickey
- Danger
- The Brain
Email addresses are bad, and I think there are a number of reasons for that. Primarily, they are almost always at some arbitrary seeming domain- uchicago.edu, columbus.rr.com, insighttelco.net. etc. These are hopeless to remember. Also, a lot of email addresses are assigned, or chosen poorly. For example, at Chicago, it was first initial, middle initial, first 5 letters of last name. How stupid is that to remember, among each of a bijillion other arbitrary systems?
But if you choose your own, and don't have any domain (or the domain for all of them is "phone" or something), then you don't have that problem.
I want the opposite: I want my phone number to be arbitrary text of my choosing. I want the whole phone system to use good voice recognition, so that any time I want to call anyone, if I can remember their phone-moniker, all I have to do is say it into the phone. A DNS-like system will go reference this to their actual phone number, IP address, or whatever address and protocol they choose to use for voice communications, and connect me.
Yeah, I played E.T. on Atari too.
FYI, that doesn't help.
Forget the sand in central park. Can anyone tell me what's going on here?
They used radically different image sets for different areas, and it's available at different resolutions for different areas. Check out some of the radical boundaries around Michigan.
Mapquest used to have satellite and USGS topographical map data for the whole US, and the top of every map had tabs letting you switch between these or their street maps, maintaining the same view and scale. Mapquest also used to let you set your preferences to view much larger pixel count maps. Why they removed all these features is beyond me.
Why not have a headline "95% of IT employees will lose their jobs this year," based on an article showing that 95% of IT firms will terminate at least one employee this year.
While highly plausible, this 95% statistic is worthless, and the headline is not at all supported by the article. Actually knowing what percent of IT projects are late might be interesting.
The more monotonous and boring the delivery is, the funnier the material seems.
Oh, but wait. Less than a hundredth of a percent of the population was interested in using a computer that way.
As long as Linux is happy to be a niche toy for geeks and computer professionals, then by all means, defend keeping it the kind of system that you have to go to school for computer science in order to use. That's what Marcus Ranum was saying will happen if it stays so diverse; it will maintain the kind of market share numbers you'll see if you stick to CS majors as your potential user base.
Most PC and Mac users don't require a formal education in computers to do their email, word processing, and web browsing. If Linux has the ambition of gaining serious market share, it needs to adopt a lot of the silly user friendliness of the competition, rather than complaining that too few people are willing to take computer classes so that they can use it.
OK, I learned some math. Here's a little math lesson.
Inelastic goods tend to have a lump sum, or flat tax (as opposed to a flat rate tax). Cigarettes, in every state in the US, are taxed by the pack. And guess what, gasoline? Taxed by the gallon And holy crap, look at this, I'm looking at my SBC local phone bill right now, and the tax is charged by the line. A portion of my long distance tax charges is actually done as a percentage, but guess what, they're changing that to a flat rate, too. And hang on, I've got my Adelphia broadband bill here, and it's got a flat rate monthly tax too!
So in your example, where they sell twice as much at 1/3 the price, tax revenue would be exactly... twice as much. So I'm looking forward to the second insallment of your math lesson where you contue to explain how I'm such an idiot for thinking that if people buy more of these things, the tax revenue will increase, and enlighten me as to how it will "definitely be less." Of course, without knowing the supply and demand curves, you can't know if it would be more or less even if it were a percentage tax, and in almost all cases, you're wrong, because total sales almost always increase when prices drop. But that's another story.
So, instead of spending $50 on a private ISP, if they spend $25 of their taxes on "free" government service
Try supposing it's $100 for the government's service, and that it ends up going out all the time, being a fifth the speed of the $50 private service, and if you think tech support is lousy these days, imagine having to drive somewhere and stand in line for 5 hours just to have them tell you they don't care and won't help.
According to IDC, 5 million americans have a wifi card now. That's about 1.6% of the population. Of course, not all of those are 802.11 b to work with these networks, but let's round up in the government's favor. Philadelpia, the first US city to try this, has a population of about 1.5 million. Philadelpia's spending $10 million on setup and expects operating costs of $1.5 million. Thus, if Philadelphians own wireless cards at about an average rate, then about 24,000 of them have cards. So for the first year, they're spending $458 of tax money per resident who could even possibly try to make use of the service. Of course, this is just to put one wireless hub on each block; what pecent of people do you think could actually get service without leaving thier home? Our 802.11 hub only reaches some rooms in our house, and it's base station is right here. The people installing the networks admit that the base stations only reach about 100 feet, and that's when they aren't going through brick walls. So I wouldn't be surprised if their first year cost is closer to $4,000 per regular user. But other people will be paying that money, instead of them paying $50 for their own access, so it's good, right? Of course, more people will buy 802.11 cards in the future if there's "free" broadband available, so the numbers should improve, if the government can keep program costs under control.
Just out of curiosity, do you think those 24,000 out of 1,500,000 who have computers with 802.11 are among the poor? How many do you think have incomes at least two standard deviations above the mean? You do realize this is a tax on everyone, including the poor, to provide services overwhelmingly consumed by the rich, so they don't have to pay the fee themselves?
"So, instead of spending $50 on a private ISP, if they spend $25 of their taxes on "free" government service, they would have mo
I don't quite follow your point as to "why" there is no free market in telecom. OK, so it's a service with inelastic demand, like cigarettes and gasoline (in the short run). So as goods with low demand elasticity, cigarettes and gasoline get really high taxes; but that's no reason for the government to give them a monopoly. In fact, in economic terms, if the government is trying to maximize tax revenue, then they want a competitive market, especially for inelastic goods with giant taxes on them. Competition brings down prices and increases total sales, and every time there's a sale the government gets their huge tax. The government doesn't care if you go to BP or Texaco, if you smoke Marlboros or Virginia Slims. They get their taxes either way. So why the monopolies in telecom?
And I don't see how any of this applies to your last point, "if you believe for a second it's about being nice to you and giving you free Internet, you're the biggest sucker out there. Government and big business is a serious sport. Wake up and look at who's putting the money behind the efforts you're idealistically supporting."
Well, government is putting money behind municipalities providing broadband as a government utility. If governments collect taxes to pay for the service, then give the service away for "free" on the margins, that will prevent them from being able to collect a tax on it. If they're giving it away for free, they not only can't tax it, they have to use other tax revenue to subsidize it. Also, by offering it at no marginal cost, people will discontinue their paid DSL and Cable broadband service (which they pay taxes on), and switch to the "free" government service (which they're already paying for). Thus providing free broadband will erode their existing tax base!
So I disagree with your insinuations regarding the motives of those promoting municipal broadband. They probably are acting for their own self interest, as they understand it. But the locals are probably for it because they don't understand that when the government provides something for "free," it's likely to cost them more in total, and be lower quality. The government backers just want more programs to administrate; it makes more jobs to hand out, more "good" they can say they're doing for the community, more accomplishments for their resume, more media coverage, larger budgets, more people to supervise, all leading to more pwoer, prestige, and higher salaries for themselves. Basically, all the normal incentives for government to expand.
This is my reply:
RE: Why Your Broadband Sucks
Perhaps Lawrence Lessig should stick to law, rather than demonstrating his ignorance in economics. Lessig starts with "city leaders stepped in where the free market had failed." What free market? The government given monopoly in cable and telephone? Where in the US is there anything approaching a free market in broadband? Does Lessig believe that Taiwan is so far ahead of the US because US businessmen can't compete with their Taiwanese counterparts in broadband? He calls government-supplied telecommunications "free." Most government services cost more than free-market equivalents; the bill just comes in taxes.
There is a problem with all of Lessig's comparisons to other government functions: he never claims that government does them better. Simply doing something is no argument for doing other, similar things. Congress passed the DMCA and the Mickey Mouse... I mean Sony Bono Copyright Extension Act. Does Lessig believe they should therefore pass more similar legislation? These comparisons with other government services either compare apples to oranges, or don't support his case. Street lights and roads are textbook examples of "free rider" problems. The difficulty of charging for street light usage probably outweighs the efficiencies of market provision; but billing for internet access is easy. As for busses and water, studies indicate that, ceteris paribus, markets provide them better, making a counterexample to his point.
Closer comparisons to government broadband are Soviet industry, or the nationalized Jaguar. Real costs are higher than those of competitive services, quality is worse, and innovation stops. Sensible deregulation brings lower prices, higher quality, and more options.
The worst fallacy is the conclusion: "let the markets, both private and public, compete" What public markets? What does "competition" mean when your "competitor" forces every customer to pay them any price, makes rules for you, and taxes your income to fund themselves? This is like the "competition" between a mugger and his victim. What do you think would happen to FedEx if the USPS was "free?" Perhaps not all innovation would stop. Like public schools, public internet access might become so bad that high quality services will serve the rich elites who can afford them, leading to the opposite of his egalitarian goals.
If we want to see the US plunge from 13th on the list down to wherever Cuba is, then by all means, let the government "compete" in the telecommunications industry. If you want to stand up to the "self-serving lobbyists," stand against their government-granted monopolies, not against saving themselves and us from socialized industry.
Yeah, they don't even mention the most popular internet body.
You still get the discussion with the poll, but you also get the raw data, which might add something of interest to the collection of posted anecdotes, each explaining where one individual feels more productive.
Therefore, you must always listen to music to avoid fines. In this way, they will sell more music.
Let's hope it at least goes better than it went last time.
Yellow Dog Linux, based on Fedora Core, also supports Mini Mac already, although they don't support Airport Extreme (yet) either.