Recently in Technology Category

Unix weenies everywhere will be partying like it's 1234567890 this Friday.

That's because, at precisely 3:31:30 p.m. Pacific time on February 13, 2009, the 10-digit "epoch time" clock used by most Unix computers will display all ten decimal digits in sequence. (That's 6:31:30 Eastern, or 23:31:30 UTC.)

Airlines should allow voice calls, at least for business travelers, one woman posting on Computerworld.com said, adding that she thought businesspeople could be counted on to use the service in a "respectful, quiet manner."

But that optimism isn't shared by others who assert that, as a blogger elsewhere put it, "these Type-A business people are the worst in bellowing on their cellphones." Another comment on Computerworld.com supported the in-flight blocking of Skype and similar programs "until phone users learn to speak in a normal conversational tone instead of shouting."

Yes, I upgraded to Vista when I bought the pieces for a new computer. I did my research and came to the conclusion that for most of what I do I would not run into the many problems that have been described. A couple of days into the experience and I am still confident of that. However, there is one thing that really annoys me.

Vista doesn't trust you.

Oh, I've turned off the User Account Control. I don't need it asking me if I really meant to do everything I do. No, I'm talking about the fact that there are certain things that you need administrator privileges to do in Vista, just like XP. No problem, I figure. I created my account as an administrator. No dice. Apparently being an administrator in Vista doesn't mean that you can do administrative tasks. Ok, so I'll run the shell (cmd) as an administrator. That ought to do it. No dice, again.

Well, ultimately I did what I wanted to do. I was trying to install Flash Player on my new machine. It didn't want to do it in Firefox. Just for fun I thought I'd try it in IE (not that I'd use it there, this was more for curiosity's sake). Well, the IE installation didn't complete, so I tried to remove it. (And that's where all the administrator fun came in.) Never did get it removed. No big deal. It will just sit there until either MS or Adobe can get their products to play nicely with each other. I was able to install it in Firefox thanks to the hint from this nice little site.

And now I'm back on track to putting my system the way I want it. But I now understand the frustration many people have had with Vista. In the end it was not a critical problem, but more of the sort of thing I wanted to follow through on just to see how (or indeed, if) it could be done. Now I know that Vista has some quirks when it comes to file permissions. Basically, it doesn't trust you--which can be sort of an inconvenience during installations. Sadly, Vista does not appear to have an equivalent to chmod 755.

How to make a local backup of your Gmail

I mentioned these once before. Now, I'm starting to contribute to the library of demonstrations. My first contribution is a simple Keynesian IS-LM model. I chose this for my first attempt as it was easy to code and could be widely used in classroom presentations. Granted, I don't make the IS-LM the centerpiece of my class, but it does have some value and students should know what it is.

If you teach macroeconomics, you might want to try it out. It requires the Mathematica version 6 player (free download). Here's a link to the demonstration. Here's a link to all the economics demonstrations. And this is what the interface of my IS-LM demonstration looks like...

Please let me know if you find it useful. I hope to do some more of these in the near future.

Today is the 50th birthday of the integrated circuit (at least if you're going by the date that the patent was filed).

Press release from the Nobel Foundation

Relevant excerpt:

This year's physics prize is awarded for the technology that is used to read data on hard disks. It is thanks to this technology that it has been possible to miniaturize hard disks so radically in recent years. Sensitive read-out heads are needed to be able to read data from the compact hard disks used in laptops and some music players, for instance.

In 1988 the Frenchman Albert Fert and the German Peter Grünberg each independently discovered a totally new physical effect – Giant Magnetoresistance or GMR. Very weak magnetic changes give rise to major differences in electrical resistance in a GMR system. A system of this kind is the perfect tool for reading data from hard disks when information registered magnetically has to be converted to electric current. Soon researchers and engineers began work to enable use of the effect in read-out heads. In 1997 the first read-out head based on the GMR effect was launched and this soon became the standard technology. Even the most recent read-out techniques of today are further developments of GMR.

Tyler Cowen has been pushed too far. Why? Listening to people complain about Apple cutting the price of the iPhone.

One customer, Kevin Tofel, was quoted in the NY Times as saying:

“I just felt so used as a consumer,” he said. “They hyped up the iPhone for six months and built up our expectations, and then they grabbed our extra $200 and ran.”

This was too much for Tyler.

It is you people, you who resent Coase (1972), you people who induce wage and price stickiness and widen the Okun gap. You people, who don't know what it means to sit back and enjoy your consumer surplus. You beasts! (emphasis in original)

Ok, so who among us did not expect that this would happen? In fact, when the iPhone came out, one blogger wrote:

There is one certainty however. In a few months, there will be a better model that will be released at about the same price and this one will be sold at a discount. For tech products like this, there is most certainly a dynamic form of price discrimination partly due to the nature of quality improvement and innovation over time and partly due to calculated profit maximizing behavior. The effect is to segment the market into the patient and the impatient.

Man, that sounds familiar.

The only thing I didn't get right about it was how fast it would happen. Two months is sooner than I would have expected, but not terribly so.

So let this be a lesson to you. Big hype around a high tech innovation just cries out for this sort of dynamic price discrimination. The market will be segmented into the patient and the impatient. If you are impatient, you will pay more than those who are patient. There is a price for being the first on the block with a new toy. You gave away some of your consumer surplus, but you've probably got some left. So enjoy being first as long as it lasts. The patient masses will soon join you in enjoying what Tyler calls those "icons of modernity". Should I feel sorry for someone who buys a $2000 computer and complains that a few months later a better one sells for $1800?

"You people...who widen the Okun gap." That's just beautiful.

UPDATE: Wired has a story that should soothe Tyler

"If they told me at the outset the iPhone would be $200 cheaper the next day, I would have thought about it for a second - and still bought it," said Andrew Brin, a 47-year-old addiction therapist in Los Angeles. "It was $600 and that was the price I was willing to pay for it."

In his Economic Scene column, Hal Varian links to this study by researchers at UC-Irvine that does a breakdown of the value added throughout the supply chain for the iPod.

I think I'll assign the article in my principles course this fall.

If you liked this one, you'll enjoy the sequel.

Hat tip to Felix Salmon.

So you want to be first on your block with an iPhone? It'll cost you.

NEW YORK (Reuters) - Apple Inc. said on Tuesday its hotly anticipated iPhone could cost as much as $3,000 with a required two-year service contract, and a handful of eager buyers started lining up to spend their money.

They don't go on sale until Friday. And to be clear, that is $3000 over the life of the two year contract, as the article later explains...

Apple and AT&T on Tuesday outlined three iPhone rate plans that will be available, from $60 per month for the most basic to just under $100 per month for more talk time. AT&T said iPhone customers could also choose other AT&T service plans listed on its Web site.

While that is consistent with other AT&T wireless plans, it adds $1,400 to $2,400 to the cost of what many say is already a steeply priced $500-to-$600 gadget.

Of course, many people already pay $60 for their wireless plan, so using the economist's favorite concept of opportunity cost, it's the $500-$600 for the iPhone itself that is of interest. Still, that is several times what I paid for our family's two phones combined. (Our monthly rate is comparable to theirs, but again, that's for two phones.)

The story wouldn't be complete without this...

Not only Apple and AT&T are hoping to profit on the iPhone phenomenon. Advertisements on the New York and San Francisco online message boards at Craigslist.org solicited payment for waiting on line to buy an iPhone.

One listing from a self-professed "professional waiter" offered to stand on line for a fee of $100 per eight hours wait. The person will throw in delivery for an extra $50.

One wonders just how high Apple could have pushed up the price for this initial release. There is one certainty however. In a few months, there will be a better model that will be released at about the same price and this one will be sold at a discount. For tech products like this, there is most certainly a dynamic form of price discrimination partly due to the nature of quality improvement and innovation over time and partly due to calculated profit maximizing behavior. The effect is to segment the market into the patient and the impatient.

UPDATE: Felix Salmon thinks they are cheap. Yes, I think they could have gone higher. But on the other hand, as they come out with better models, they may be intending to stick with the $500-$600 price for the top shelf model for some time which will capture a moderate amount of consumer's surplus at each stage of product evolution rather than try to squeeze out too much in the first round. Just a thought.

Oh, and via Wired, we read that the iPhone will be sold in a box just like the iPod and you'll set it all up through iTunes.

Customers will choose their AT&T voice and data plans through iTunes, while iTunes is transferring songs and video files to the iPhone (and iSync or Outlook is copying contacts and calendars). The phone will be activated remotely over the cell network.

"It will go much, much faster than the normal process of buying a cell phone," Allen said. "Apple has a reputation for streamlining the customer experience. I'm sure there's not going to be any box opening in the store. That's such an important part of the process of getting an Apple product."

Because Apple won't be opening boxes in the stores, Allen said buying an iPhone on Friday -- when big crowds are expected -- might be faster at Apple stores than at AT&T stores.

Google Maps will soon have street level views good enough to read the "no parking" signs by the curb. Check out the demo.

Via Wired.

UPDATE: Not everyone is happy.

I received a notification in my e-mail today about the Wolfram Demonstrations Project. If you don't have Mathematica 6.0, you will need to download the free player for the demonstrations. (I have version 5, and it doesn't play the files--you need the new interactive interface in version 6 apparently.)

There are some really nice demonstrations available, including many in for economics.

Take some time this weekend to download the player and experiment with some of the demonstrations. Many would be great for use in classes in math, economics, and the sciences. Enjoy!

Of course not.

Is Wikipedia a useful research tool for college students? Absolutely.

Middlebury College's history department sees students relying too much on Wikipedia and decides to nip it in the bud. (NY Times)

When half a dozen students in Neil Waters’s Japanese history class at Middlebury College asserted on exams that the Jesuits supported the Shimabara Rebellion in 17th-century Japan, he knew something was wrong. The Jesuits were in “no position to aid a revolution,” he said; the few of them in Japan were in hiding.

He figured out the problem soon enough. The obscure, though incorrect, information was from Wikipedia, the collaborative online encyclopedia, and the students had picked it up cramming for his exam.

Dr. Waters and other professors in the history department had begun noticing about a year ago that students were citing Wikipedia as a source in their papers. When confronted, many would say that their high school teachers had allowed the practice.

Well, encyclopedias have always been a staple of high school libraries. Any manual on writing term papers has a line about how to cite information from an encyclopedia. I always cringe a little at that.

But the errors on the Japanese history test last semester were the last straw. At Dr. Waters’s urging, the Middlebury history department notified its students this month that Wikipedia could not be cited in papers or exams, and that students could not “point to Wikipedia or any similar source that may appear in the future to escape the consequences of errors.”

With the move, Middlebury, in Vermont, jumped into a growing debate within journalism, the law and academia over what respect, if any, to give Wikipedia articles, written by hundreds of volunteers and subject to mistakes and sometimes deliberate falsehoods. Wikipedia itself has restricted the editing of some subjects, mostly because of repeated vandalism or disputes over what should be said.

Although Middlebury’s history department has banned Wikipedia in citations, it has not banned its use. Don Wyatt, the chairman of the department, said a total ban on Wikipedia would have been impractical, not to mention close-minded, because Wikipedia is simply too handy to expect students never to consult it.

Precisely. Consult it. Read it. Find out what people are saying. Learn some interesting facts and pick up some additional sources. Then confirm everything in more reputable sources before you put it in a term paper.

As for using Wikipedia to study for a test and then regurgitating what you read into your blue book, well, that's like skydiving without knowing who packed your parachute. It might work sometimes, but that doesn't make it smart.

At Middlebury, a discussion about the new policy is scheduled on campus on Monday, with speakers poised to defend and criticize using the site in research.

Rather low on my list of priorities for a campus discussion, but hey, whatever floats your boat.

Jimmy Wales, the co-founder of Wikipedia and chairman emeritus of its foundation, said of the Middlebury policy, “I don’t consider it as a negative thing at all.”

He continued: “Basically, they are recommending exactly what we suggested — students shouldn’t be citing encyclopedias. I would hope they wouldn’t be citing Encyclopaedia Britannica, either.

Mr. Wales gets it.

The Wikipedia phenomenon caught my attention some time ago. A few semesters back, I got deluged with Wikipedia citations in term papers. It took me a little by surprise. One semester it wasn't there at all, and the next semester it's everywhere. Henceforth, I have informed my classes that it is not an appropriate source for term papers or essays. It is, as I also tell them, a valuable research tool that can point you in positive directions.

That said, I've picked up a lot of interesting tidbits from browsing Wikipedia. It is useful for looking up obscure things like who the celebrity panelists were on Match Game. Not to mention the fact that the "wiki" concept is great for software support websites--let the users help each other. I'm all for creating open source communities. The Times article describes some positive uses of the wiki concept in higher education. The article also goes on to say,

The discussion raised by the Middlebury policy has been covered by student newspapers at the University of Pennsylvania and Tufts, among others. The Middlebury Campus, the student weekly, included an opinion article last week by Chandler Koglmeier that accused the history department of introducing “the beginnings of censorship.”

Oh, please. Much of what is in Wikipedia is also in reputable sources, so confirm it and cite the reputable sources. In fact, most entries have a reference list. Notable exceptions to this would be erroneous information and blatant opinion which is typically anonymous. Of course in a term paper opinions are fine to cite, but they should be attributable to someone, preferably someone of some authority on the topic. Most professors would look askance at citing an anonymous person off the street as an authoritative source in a paper. That's not censorship. That's teaching students the craft of scholarship.

But then there's this opinion,

Other students call the move unnecessary. Keith Williams, a senior majoring in economics, said students “understand that Wikipedia is not a responsible source, that it hasn’t been thoroughly vetted.” Yet he said, “I personally use it all the time.”

Leave it to an economics major. I'm of a similar mind. The part that Middlebury gets right was at the end of the 4th paragraph,

...students could not “point to Wikipedia or any similar source that may appear in the future to escape the consequences of errors.”

That's the right approach.

Note to students: If you cite Wikipedia in a term paper, you are working without a net. At this stage in your career, you are not qualified to do that. You may also be interested to know that your professors can spot errors, inconsistencies, and shallow reasoning in Wikipedia a lot more easily than you can. You have been warned.

I will never tire of walking around in the stacks, browsing the shelves and holding classic books in my hand. But digital resources are simply wonderful.

Last night, I was reading something at home when I suddenly got the urge to dig out an article by Jagdish Bhagwati on capital controls. I could not recall the year or issue, or even the journal--though I knew it was late 1990s and had a hunch that it was in Foreign Affairs.

I got up out of my easy chair and walked to the computer: estimated time about 30 seconds.

With one Google search, I had the issue and journal (May/June 1998 Foreign Affairs): estimated time about 15 seconds.

I click over to the university library, enter my library card number, go to the full text archive and I'm reading the article as a PDF in another 30 seconds.

It was not long ago that such things were only in our dreams.

If I want to be faster, the easy chair needs to be closer to the computer.

Roll the tape.... from Tuesday:

...I would be wary of a government managed plan for universal broadband access. My fear would be that they would adopt a 20th century solution to a 21st century problem.

Broadband technology (particularly the new wireless broadband) is still evolving. Once you make it a government program, you introduce a lot of rigidity. Better to keep some flexibility until we see which technology is superior. We can, I believe, afford to do that in this case because wireless is a low fixed cost operation compared to high fixed cost utilities such as the electrical grid, the copper wire laid down by Ma Bell, and even cable TV. There will be competition just as there is for wireless phone service--speaking of an industry that went from high class luxury to practically universal access in about a decade.

Friday's Wall Street Journal editorial has this to say:

Much of this [telecommunications sector] growth has been fueled by increased broadband deployment, which makes high-speed Internet services possible. The latest government data show that broadband connections increased by 26% in the first six months of 2006 and by 52% for the full year ending in June 2006.

Also noteworthy, notes telecom analyst Scott Cleland of the Precursor Group, is that of the 11 million broadband additions in the first half of last year, 15% were cable modems, 23% were digital-subscriber lines (DSL) and 58% were of the wireless variety. Between June 2005 and June 2006, wireless broadband subscriptions grew to 11 million from 380,000.

This gives the lie to claims that some sort of cable/DSL duopoly has hampered competition among broadband providers and limited consumer options. That's the charge of those who want "network neutrality" rules that would allow the government to dictate what companies like Verizon and AT&T can charge users of their networks. But the reality is that the telecom industry has taken advantage of this deregulatory environment to provide consumers with more choices at lower prices. Verizon's capital investments since 2000 exceed $100 billion, and such competitors as Cingular, T-Mobile and Sprint are following suit. So are the cable companies.

Memo to Congress: Don't mess this up.

Repeat after me. This is not another Y2K problem. Ok. Now get ready to do some manual time adjustments when daylight saving time starts because your software might not do it for you. Read this (MSNBC).

Daylight saving time arrives a little earlier — March 11 — and stays a little later — Nov. 4 — this year. And it’s bringing a problem along with it that could affect everything from stock trades to airline schedules to your BlackBerry.

Software created before the law mandating the change passed in 2005 is set to automatically advance its timekeeping by one hour on the first Sunday in April, not the second Sunday in March. Congress decided that more early evening daylight would translate into energy savings.

My hunch tells me that UNIX systems will take care of this easily since UNIX counts seconds since the UNIX epoch (Jan. 1, 1970) and that there should be a simple software patch to fix this at the operating system level. (A quick Google search reveals that my hunch appears to be correct.) Microsoft will release a patch soon according to the article. That will fix anything that depends on Windows' time keeping functions. So this is an operating system issue, which means it should be easy to fix on the vast majority of systems that handle important operations for businesses. UNIX folks have a lot more to worry about in 2038, but let's not go there.

So how about those energy savings? Don't count on it.

From Wilson Mixon at Division of Labour,

According to a report on NPR, "in a major disaster the government can spend $10-15 per gallon of water." (Link leads to a summary and an audio; LexisNexis has the transcript.)

Couldn't much of this water could be delivered at a fraction of this cost if price gougers were given free rein? Of course, the NPR report is hawking a gee-whiz technical fix, administered by FEMA, not a sensible market response.

It's worth it to listen to the audio. This is a really interesting idea. Basically, they've come up with a way to take water out of the air using a desiccant. An example of this would be those little packets of "silica gel" you find in various product packaging. The technological innovation described in the audio uses the same principle to generate around 1200 gallons of water per day for around 20 cents per gallon. (These numbers are mentioned in the audio.) The 20 cents per gallon is apparently the marginal cost for the fuel to run the machine. The machine itself costs $300,000. That is a lot of overhead. By my calculations, if the machine lasts 10 years and you run it 25 full days per year it will add about a dollar per gallon to the average cost of the water produced.

Even so, it is likely that the average cost from such a machine, if it is as successful as the report makes it sound, would be less than the $10-15 that the government sometimes pays to bring water to a disaster area. So it might be an improvement from the status quo. But what about Mixon's suggestion that letting "price gougers" take care of the problem by bringing water from outside the area and selling it at a higher price? Let us not forget that part of the reason that the prices the "gougers" charge are so high is that the government does a pretty effective job of discouraging them. If the government got out of the way, there would be more "gougers" and the price would be lower. The price differential should approach the transport cost. In such an environment, it may be the case that machines like the one in the NPR report would be considered too costly to implement. It would be interesting to do the detailed cost/benefit analysis on that one.

The fact that the government has taken such a stance to prevent "price gouging" actually helps to justify the purchase of the machine by the government and explains the incentive for private firms to look for this type of solution. All of this just goes to show that private individuals will try to capture the rents associated with various market interventions. One type of activity is praised and the other is vilified. Yet it is not immediately clear which activity yields greater benefits to society.

The NPR report also mentions that the military is interested in this technology. Now that actually makes sense. This could be a revolutionary way to provide water on the battlefield. Wouldn't it make sense to have National Guard units purchase these machines for deployment either in battle or in times of national disaster? Putting the machines in the hands of the National Guard would seem to have a much better cost/benefit calculus than putting them in the hands of FEMA. If allowing markets to work freely is asking too much, can we at least ask that the machines go where they will be used most efficiently?

From Wired: The real story behind the iPod.

Exercise for those who teach principles of economics (especially micro): Come up with as many principles level lessons as you can from this. Here's one:

Apple's team knew it could solve most of the problems plagued by the Nomad. Its FireWire connector could quickly transfer songs from the computer to player -- an entire CD in a few seconds; a huge library of MP3s in minutes. And thanks to the rapidly growing cell phone industry, new batteries and displays were constantly coming to market.

Complementary goods in production Lower input cost per unit of quality--Better and cheaper cell phone batteries and displays lowered the cost and increased the quality of Apple's innovation. (UPDATE: Lower input costs would be a more direct answer, but here is what I was thinking. Smaller batteries and displays were complementary with the smaller sized hard drive and other features of the iPod. What good is the small hard drive if everything else is bulky? What good is it to be mobile if the battery doesn't last? The iPod's value came from the combination of innovations. Only when the size and power requirements on all of these complementary inputs are met does the iPod become viable. Substituting a bulkier battery greatly diminishes the value. That is the sense in which I meant them complementary. Not in the literal production process, but innovative, value-creating process.)

You could probably spend a couple weeks in class on the intellectual property rights issues if you were so inclined. You are invited to post iPod teaching suggestions in the comments.

UPDATE: I also like to use the iPod as an example of invention (small hard drive, battery, etc.), innovation (assembling the various inventions into a music player), and diffusion (network externalities with iTunes, marketing, and adoption by the masses).

The 500GHz chip has arrived, but there is still one obstacle to getting it on your desktop--it operates at a few degrees above absolute zero.

Room temperature performance isn't that bad either...

Instead of relying on pure silicon, the foundation material of modern semiconductors, the chips in the demonstration were made of silicon and germanium, using an increasingly popular combination of elements to improve speed. Even at room temperature, the chips can run as fast as 350 billion cycles, or gigahertz. In contrast, silicon-germanium chips in the current generation of cellphones run at about two gigahertz.

Full story at the WSJ.

Wired calls it a "quantum leap".

I've been thinking about getting a tablet PC. I think I'll wait until Windows Vista ships. This sounds really nice. (NY Times)

With Microsoft's new Windows Vista software, to be available in January, virtually any newspaper, magazine or book can be formatted into an electronic version and read online or off. The software would allow The Times to replicate its look — fonts, typeface and layout — more closely than its Web site now does.

...

The Times said it would charge advertisers to appear on the new version of the newspaper, called Times Reader, but it had not decided whether to charge readers for the service. Microsoft would include the offering in the next version of its operating system.

The value added here is in the formatting and the organization. The content would be essentially the same. I'd say they should bundle it with Times Select. (End of free advice)

For today's demonstration, The Times was downloaded onto small tablet computers, about the size of a hardcover book, which are already commercially available for $1,000 to $3,000. But this printlike version of the newspaper could also be downloaded onto a home computer or a laptop. The electronic paper is displayed in columns and it formats itself to fit any size screen.

Mr. Gates said he had long wanted to make easier what he called "on-screen reading" and he had reached out to The Times to help develop that ability.

Mr. Sulzberger said the software combined the portability of the print paper with the immediacy of the Internet. Readers can in effect turn the page electronically. There is also a gauge that tells them how much of the paper they have read and how much more is left.

Tom Bodkin, an assistant managing editor of The Times and its design director, demonstrated Times Reader to the audience. "You can page through the entire paper in a natural and intuitive way," Mr. Bodkin said. Mr. Gates said that starting in January, new computers would come equipped with the software that would allow access to such newly formatted newspapers. He said he expected that by then, other publications would have developed electronic versions closer to their own styles and typefaces.

I'm not sure I need a gauge to tell me how much I've read and how much is left. Give me a book on a tablet PC that displays a "double page spread" the way that an open book in your hand does. Allow me to turn the pages by touching the screen. Make the pages turn as fast electronically as I can turn them physically in my hand. Give me tabs for chapters and an intuitive way to bookmark pages and keep my place. Oh, and a hypertext index would be nice too.

Give me all those things and I'll try it. An E-book (or magazine or newspaper) needs to replicate the physical presence of a book in a way that Adobe Acrobat Reader does not. But this sounds like the closest thing yet. I would like to see it demonstrated. Even if this product doesn't go all the way, I'm sure someone is working on satisfying all of my criteria.

It sounds crazy, but it just might work. From IN-FORUM (Fargo, ND--free registration required):

BISMARCK – Why put up more than 1,000 towers to spread cell phone service across North Dakota when a few balloons would do it?

So says former Gov. Ed Schafer, one of the backers of a plan to attach wireless re-peaters to weather balloons high above the state to fill gaps in cellular coverage.

“I know it sounds crazy,” Schafer said, “but it works in the lab.”

Extend America, a North Dakota wireless telecommunications company, and Chandler, Ariz.-based Space Data Corp. are developing the balloon-borne cellular technology, believed to be the first of its kind.

A trial balloon will be launched next month in North Dakota to test the theory, said Schafer, the chief executive officer of Bismarck-based Extend America. Schafer left office in 2000 after eight years as governor.

...

Jerry Knoblach, the CEO of Space Data, said although the balloon technology called SkySite is new to cellular, “the platform is very well proven.”

His company has launched thousands of the free-floating balloons in Texas, Oklahoma, Louisiana, Arkansas and New Mexico to track data for oil company vehicles, wells and pipelines over the past year, he said. And Knoblach is certain the balloons will work for cellular service in North Dakota – even in cold or stormy weather. He said balloons were launched even during Hurricane Katrina.

“It’s just like a weather balloon at the airport,” Schafer said. “There’s enough hydrogen in them to rise very rapidly.”

Up to 20 miles above the Earth, stratospheric winds would push the latex weather balloons across the state at about 30 mph. Each balloon would deliver voice and data service to an area hundreds of miles in diameter, Schafer said.

“Nine balloons would always be in the air, with some going up, some going down, and some in the middle,” Schafer said.

Once the balloons transit the state’s stratosphere, the electronic gear would be jettisoned remotely and fall to the earth with a parachute.

The electronic equipment, about the size of a toaster, would be recovered through the use of a global positioning satellite device.

“We’d pay some guy a bounty, put in a new battery pack and send it off again,” Knoblach said.

Schafer said a repeater could be used indefinitely “unless it lands in a lake or gets run over by a truck.”

Unless it drifts over the border into Minnesota, landing in a lake should not be a problem. By now you're probably wondering about the cost,

Knoblach said the hydrogen-filled balloons cost about $55 each. The balloons swell from six feet in diameter to 30 feet after they gain altitude. After the electronic equipment is released, the balloons expand with the drop in air pressure until they burst.

Winds at high altitudes are consistent, blowing west to east in the winter, and east to west in the summer, Knoblach said. The balloons would travel above the jet stream, and he said they would not be bothered by storms.

Schafer said it costs about $250,000 to build one cellular tower in North Dakota, and many remote areas don’t have enough customers to pay for it.

“To cover every square mile of North Dakota, it would take 1,100 cell towers,” Schafer said. “We can do the whole state with three balloons – and it won’t have problems with that line-of-sight stuff,” he said, referring to hills that can block signals from towers.

Sure sounds interesting. I'll keep my eyes open to see if they have a follow up on the "trial balloon". If their cost estimates are correct and it works technically then this could really take off. I guess my main technical question would be about reliability of the service. Do they expect that they could achieve "five 9's"? (99.999% uptime) That would probably be the make-or-break for the technology. Whatever the case, I give them credit for creativity. As an economist, I'm particularly intrigued by the attempt to overcome the barrier of high fixed costs of building a network with low customer density.

True genius is seeing the ability to improve on something that we assume to be so simple that it can't be improved. Here's an invention that I can't wait to see in action. (NY Times)

The best way to see the future of New York elevators may be to visit the Marriott Marquis Hotel, the behemoth in Times Square.

Visitors who get into the elevators there, expecting to press a button for their floor, are stymied: there are no buttons in the elevators. Instead, there are keypads in the lobby. Punch in the floor you want, and a digital readout tells you which elevator to take (each car is identified by a letter). "I call it the express bus system," said the hotel's general manager, Michael J. Stengel.

Because it knows where people are going before they board, the computer controlling the elevators can sort passengers, eliminating a pet peeve of elevator riders: doors that open at floor after floor even though the car is full.

Already, Mr. Stengel said, a system installed in "the back of the house" - the service zone used by employees - has drastically cut average elevator waiting time.

Beautiful. Anything to make the high rise elevator experience better is an idea worth supporting.

Check this out (CNN)

TOKYO, Japan (Reuters) -- Imagine watching a football match on a TV that not only shows the players in three dimensions but also lets you experience the smells of the stadium and maybe even pat a goalscorer on the back.

Japan plans to make this futuristic television a commercial reality by 2020 as part of a broad national project that will bring together researchers from the government, technology companies and academia.

Sounds like fun. It would be great if it works, but I have some doubts. Can you imagine the bandwidth and processor speed that would be required to convert live action to three dimensional virtual reality? Will Moore's Law get us there by 2020? Look at the improvement that HDTV is over analog TV. It took about as long to get us that incremental improvement that is nothing like what will be needed to get virtual reality TV (VRTV?)

But the end of the story had me wondering if they were serious.

The ministry plans to request a budget of more than 1 billion yen ($9 million) to help fund the project in the next fiscal year starting in April 2006.

Hands up all who think that kind of funding, even multiplied 15 times over, is going to do it. I thought so.

Google is amazing. I typed in "money spent on HDTV research" and found this from the Research Laboratory of Electronics at MIT. It was published in 1990. The words are those of MIT researcher William F. Schreiber.

People have tried to estimate how much the Japanese have spent on HDTV--some say $500 million, others say $1 billion. That may seem like a lot of money, but General Motors spent $3 billion to introduce the Saturn automobile. Any new model car from Detroit costs $1 billion to introduce. A Boeing 747 costs $150 million. Each TV network is a $4 billion operation, and all the networks together are smaller than NYNEX. So, in those terms, a billion dollars isn't much money.

A billion yen is even less.