|
|
|
|
|
|
|
|
|
Posted Wednesday, December 18, 2002 Be Back January 6thI'm not leaving until Dec. 22, but we need to get ready for our trip. Our family is taking our vacation in Israel. So make a note to revisit this site on January 6th or thereabouts. Yes, I know, I know. No, I'm not a thrill-seeker. Somehow, our oldest daughter picked a study program there over her winter break, and the rest of us are tagging along (not for the study program). In today's Wall Street Journal, Claudia Rosett writes
I think I can answer that. The U.S. would have driven the Palestinians to someplace at least three time zones away. Land for peace? There's your land. Now leave us in peace. As an American, I make no claim to understand Israelis. From where I sit, it seems to me that the secular left in Israel expects the Palestinians to appreciate how decent and moral the Jews are and to leave them alone. It seems to me that the religious right expects the Palestinians to appreciate the biblical evidence for Jewish rights to the land and to leave them alone. Meanwhile, when I read something like this piece from David Warren, I wonder how anybody thinks that they are going to be left alone by the Muslims.
It reminds me of Mark Twain's novel A Connecticut Yankee in King Arthur's Court. The Yankee finds himself in a backward, barbaric world. He tries to give the medieval people his modern technology and moral outlook, but they do not appreciate what he offers. He winds up using his superior knowledge to inflict heavy casualties in the grisly battle that is the book's depressing conclusion. UPDATE: A reader writes to object to Warren's implicit dismissal of religions other than Judaism, Christianity, and Islam. My guess is that this is one of many unintended adverse reactions that I can get from straying into the topics of Israel, Islam, etc. All I was trying to say in this post is that you would think that since I am visiting Israel that I must be somewhat optimistic about the future there--but the fact is that I'm not. . . . . . .
Is Spectrum Scarce?Richard Bennett thinks so.
Unfortunately, I do not have the engineering know-how to evaluate this in relation to what others are saying. . . . . . .
Posted Tuesday, December 17, 2002 Generation GapsDavid Strom writes,
He is talking about how his teenage daughter seems more advanced than he is. I know the feeling. In May, 1994--less than 10 years ago--I started a Web business. It was one of the first, and it put me on the leading edge. Today, I feel as though I have at most a flimsy grasp of the latest technologies and trends. I certainly do not have the kind of first-hand knowledge that I had back then. When I first read Kurzweil's The Age of Spiritual Machines, my reaction was to doubt the ability of humans to keep pace with Moore's Law. It implies ever-widening generation gaps. UPDATE: Here are some folks' thoughts on whether there is a user-interface generation gap related to SMS (short message services). . . . . . .
The Costs of Pricing, ContinuedGlenn Fleishman sent me an email that led me to clarify my concern. He wrote
Well, more is always better, but that is not an argument against pricing. The argument sounds downright silly if it is applied to an industry that everyone understands: You would never tell an airline "Do not charge more for flights to Florida at Christmas--it would be better to add more personnel and planes." You would never tell the electric company company "Do not charge more for electricity use at peak times. It is always better to build bigger power plants." There may be something different about bandwidth (see the post below). But it would help if the Reed, Frankston folks could articulate the difference in a way that makes sense in economic terms. . . . . . .
Posted Monday, December 16, 2002 The Costs of PricingAn old debate that is unlikely to go away is whether or not the Internet needs some sort of pricing scheme (sometimes referred to as Quality-of-Service differentials) to deal with congestion. To tune in to some recent comments on this debate, go to SATN.org, although for me the most helpful article is this interview with Andrew Odlyzko.
The simplest pricing scheme would create two tiers for packets. You pay extra for top-tier packets (perhaps you pay a higher monthly fee and all of your packets go into the top tier). When there is congestion, a second-tier packet waits its turn until top-tier packets have gone. An interesting analogy might be with the problem of dealing with spam email. For spam, where there also is a pricing solution, I've been convinced by SATN.org folks and by John Gilmore that filtering at the ends is better than putting a complicated pricing scheme into middle of the email process. I think that we have the computer power to do good email filtering--I'm annoyed that nobody has a Bayesian statistical filter available commercially yet. I am persuaded that an endpoint-based solution is better than a network solution. I think that those who oppose congestion pricing on the Net are in a good position to the extent that there is an endpoint-based solution for congestion. That is, if you tell me that by buying a smarter receiver I can get packets more quickly, then there is a decent alignment of individual and social incentives. That may be the case in a mesh wireless network--as people buy better equipment, latency could go down for everyone. If the incentives don't align that way, then I would not throw out the pricing tool altogether. But my guess is that it needs to be incredibly simple and also rarely invoked. . . . . . .
Dissing MedicareThe main argument about Medicare in the last Congress was about how to add a prescription drug benefit. I have a somewhat different slant.
. . . . . .
Who Needs 'Creative Commons'?Lawrence Lessig's Creative Commons has its official launch today. It is supposed to provide alternative licensing mechanisms for creative works distributed over the Internet. I give Lessig credit for trying to put theory into practice. However, I do not agree with the theory. The theory behind Creative Commons is that what is broken is copyright law. Instead, I believe that what is broken are some traditional business models. There are people who think that the way to generate revenue on the Net is with bigger, jumpier pop-up ads. This tells me that the cluetrain has yet to reach the station. There are better business models, and I think that the Net will grope its way toward them. But the way I see it, this Creative Commons thingy is beside the point. . . . . . .
Posted Sunday, December 15, 2002 The Trackable SocietyOne use for tracking devices is to put them on your young children. But there are bound to be other uses. People on probation...people here on temporary visas...explosives...dangerous chemicals... . . . . . .
Posted Friday, December 13, 2002 O'Reilly Joins the ClubTim O'Reilly has some sensible comments on the economics and business of media on the Internet, particularly on the second page of his article.
I'm pretty sure that he would agree with this:
And with this.
. . . . . .
Posted Thursday, December 12, 2002 The Entertainment Industryis always fun to take shots at. In this essay, I write
Another shot comes from Glenn Reynolds.
. . . . . .
Posted Wednesday, December 11, 2002 A Collaborative Filtering ConceptI have a soft spot for numerical rating systems, like this idea for collaborative filtering of news. I think that there is huge potential in capturing data on rings of trust, and I was sorry to see the SixDegrees web site fail to take off--I'm hoping that it was just either ahead of its time or mismanaged from a business standpoint, or both. But when you offer me information, in the form of an email, a weblog entry, or othe communication, I think that the number of degrees of separation between you and me is likely to be a powerful indicator of how valuable I will find that information. In the blog world, you could measure degrees of separation by using blogrolls. For example, in the right-hand column of this page, my first degree consists of Zimran Ahmed, smartmobs, gizmodo, Brad DeLong, etc. Smartmobs in turn has a blogroll that includes some blogs that already are in my first degree, plus some new ones, such as Aula and David Brake. Those would be in my second degree. Following those links would take you to my third degree, etc. Similarly, one could create degrees based on email address books. Hylton (the Corante founder) would be in my first degree. The people in his address book would be in my second degree. I'll bet that if you had a fully populated database of these sorts of degrees, an anti-spam filter based on degrees of separation would be very powerful. Of course, the spammers would have a strong incentive to infiltrate and corrupt the database, but if you could stop that somehow... I also believe that a degrees-of-separation concept is the way to use a large database to identify potential terrorists. My guess is that terrorists have very high scores in measuring closeness to other terrorists and very low scores in measuring their closeness to decent people. . . . . . .
No More Moore's Law?Andrew Grove, the CEO of Intel, gave this warning:
Bummer. . . . . . .
Posted Tuesday, December 10, 2002 The Death of Newspapers, Con'tThe blog world is trying to take another bite out of their franchise by covering state and local politics. I think that the concept will work, although I am agnostic about this particular implementation. . . . . . .
Subsidize Open Source Software?Today's TCS has a point-counterpoint on whether or not government needs to tilt in favor of open source software in its procurement. Julian Sanchez makes the point.
I think that what Sanchez is saying is that for-profit software can fund lobbying activities more readily than open-source software, and something must be done to counteract the bias that such lobbying provides. I guess my view is that the government is inevitably going to make bad decisions about software. Any new regulations--such as "affirmative action" for open source--will add to the costs of the software procurement process without producing better results. In the government market, the buyers are committees. The motivation of the buyer is to minimize personal risk, not to maximize benefits relative to cost. The same holds true to some extent with large corporations, but in the private sector there is at least a profit discipline in the background. A lot of software companies stay away from the government market, because they find the process of dealing with government purchasers debilitating. I would advise the open source movement to adopt that attitude. . . . . . .
Two New PredictionsHere are some new predictions, based on the principle that sensory mismatches don't work. People have eye-hand co-ordination, and they have hearing-speech co-ordination. Trying to mix them does not work. 1. The media center PC will not take off. ('Jane Galt" worries that it might undermine free TV.) A media center PC allows you to manage your TV-watching and your music-listening more efficiently. However, for most people most of the time, efficiency is not the point. They want background noise. Listening to background noise should not require eye-hand co-ordination. Most people actually concentrate on the TV only a small percent of the time. The TV might be on for 10 hours a day, but it gets a person's undivided attention only 2 hours a week. The media center PC gives you management tools for when you really want to concentrate on the background noise--an oxymoronic function. 2. Speech recognition will always fail as an input mechanism for screen-based computers. Keyboard input (or handwriting) uses eye-hand co-ordination, which we have been developing for thousands of years. On the other hand, speaking to a screen is not natural. The form factor that works for speech recognition is a headset, where people get the input from earphones rather than from screens. . . . . . .
Posted Monday, December 9, 2002 Revisiting old PredictionsDan Pink points to an article saying that mobile phones have now passed landline phones in terms of numbers. My guess is that somebody saw that one coming several years ago (Nicholas Negroponte comes to mind) and deserves credit for foresight. Which led me to revisit some of my old predictions. For example, at a time when some people were wildly bullish on Java (George Gilder comes to mind), I was relatively bearish. In fact, I wrote,
I said that in late 1998, when XML was barely out of the oven. It was a very aggressive prediction, but now it looks almost safe. Web services are hot, and applets are not. A few months earlier, I made three calls. One was that Internet growth was in the process of slowing. In restrospect, this is clear, but many people were betting differently--hence the overcapacity and collapse in the telecom industry. A second call was about mobile computing.
The third call was that centralized portals (Yahoo! was the dominant firm at the time) would fail.
Again, I stand by that call, although I was ready to give up on single search engines altogether, even as Google was coming up with a clever Judo technique for turning the decentralized information on the Net into a useful centralized search engine. . . . . . .
Object OrientedIf you are a programmer, you probably know about object-oriented programming. Then you may appreciate Jakob Nielsen's riff on object-oriented...er...objects.
One of Nielsen's examples is this:
It turns out that Microsoft is developing Pensieve, although that is not what they call it. Thanks to Lawrence Lee for the pointers.
. . . . . .
Surveillance DatabasesIt turns out that somebody is in favor of them.
Let the flaming begin. . . . . . .
Posted Sunday, December 8, 2002 An EntrepreneurDoc Searls pointed to the story of the resignation of a state CIO, Phil Windey.
I'm with Doc that Phil's web site slogan "Organizations usually get the IT they deserve" is insightful. One way to think about working in a large organization is that you can either be someone who leads the horse to water or someone who gets it to drink. An engineer or other technical person is best suited to leading the horse to water--pointing out ways to make the organization more effective. Often, however, a different personality is better at moving the political levers to implement change in a bureaucracy. If you're the techie, it's best not to get too caught up in the overall organizational soap opera. Just make your recommendations and be done with it. Sometimes the horse drinks, and sometimes it doesn't. When it doesn't, try not to lose sleep over it--just move on to a different topic or idea. Of course, if the horse never drinks, then probably at some point it makes sense to find a different situation. Phil's problem is that he really wants the horse to drink. I know the feeling. The only solution for me was to become an entrepreneur. It's what I would recommend for Phil. . . . . . .
Posted Saturday, December 7, 2002 Pricing Wireless BandwidthI disagree with my colleague Dana Blankenhorn that price-per-bit is where we are or should be headed with wireless bandwidth. First of all, for the wireless last mile, consumers are the suppliers. In the Reed, Frankston vision, every device that is smart enough to interpret signals is also smart enough to relay them. In Frankston's metaphor, everyone owns a refrigerator, so nobody pays for someone to deliver ice. OK, but what about the fiber backbone? Just as we need water to make ice, our devices will need the Internet backbone to communicate. Here, the best analogy is not with water but with something like a large electric power plant. The marginal cost is close to zero, unless the plant is operating at capacity. There are two pricing issues associated with the Internet backbone. First, you need a scheme that enables the backbone providers to recover their fixed costs (the cost of the power plant). It seems to me that a flat monthly rate works better for that purpose. It creates a predictable revenue stream for both the supplier and the consumer. It is less costly to monitor and to bill. It might make sense to modify the flat-rate model to allow some crude price discrimination. For example, you could pay one flat rate for a certain average monthly usage rate, and a higher rate kicks in if you go above that. Sort of like picking your cell phone plan today. The second requirement is to provide a disincentive to overload the network at peak times. What I think that leads to is two-tiered service. One tier (call it second-class) is for people who do not mind if their transmissions are interrupted when there is a burst of activity. People will pay nothing beyond the monthly flat rate for second-class service. For first-class service, which will have first priority when the network is overloaded, people will have to pay extra--either a higher monthly fee or a higher rate for data packets that they designate as first-class. . . . . . .
Payment TechnologyMIT's Technology Review is always worth a read. This month, the highlight for me is the story on smart cards and other payment technologies. The situation is quite chaotic. What standards will be set for hardware, authentication, and communication? My own guess is that the communication standard ultimately will be Wi-Fi, which will have the greatest portability and adaptability. I think that the hardware will be very flexible--no one form factor will dominate. My guess is that companies that develop inexpensive, reliable authentication systems and achieve market penetration will have a short-run economic advantage. Every payment-receiving terminal will have a list of authentication systems that it will accept--just as today every store has a list of credit cards that it will accept. . . . . . .
Posted Friday, December 6, 2002 Paul O'Neill's EnemiesPaul Krugman is congratulating himself over the resignation of the Treasury Secretary. So is Lawrence Kudlow. Krugman and Kudlow are ideological opposites, but they have adopted almost identical writing styles--heavy on rhetorical sizzle and light on substantive steak. I honestly believe that nowadays the FCC may be more important than Treasury. The stakes in telecom are huge, and there are some really expensive mistakes that could be made by the regulators. I happen to like the current approach, which I see as encouraging competition from Wi-Fi. I think that this will do more to promote economic recovery than any tax cut. . . . . . .
Posted Thursday, December 5, 2002 Telecom Reform is ComingThis kind of telecom reform is so-o-o-o last century. This the kind of telecom reform that matters:
The twentieth-century telecom reformers wanted to use regulation to turn the telephone network into a common resource. What this Cometa thingy will do instead is turn the Internet backbone into a common resource, and that will make existing telephone networks worthless (George Gilder once termed them "copper cages"). With voice over IP, Wi-Fi can compete with cell phones and land line phones. I bet Wi-Fi wins. UPDATE: On Friday, December 6th, Paul Krugman trained his rhetorical cannon on Michael Powell for the failure of 20th-century Telecom Reform. John Markoff reported on 21st-century Telecom Reform. We report, you decide. . . . . . .
Why I think that Music CD's are in troubleHere is a hard disk which Gizmodo points out can store up to 125,000 MP3's. Suppose that we say that there are 50 years of recorded popular music, or 600 months. We could take over 200 songs per month from the history of pop music and store them on this hard disk. That is what makes the CD industry outlook so bleak. . . . . . .
Posted Wednesday, December 4, 2002 The Music Industry's Future'Jane Galt' has been thinking about the future of the music industry.
Indeed. The future of music looks like this: 1. The costs of storing and distributing music will continue to fall. I think this will lead to bundling vast quantities of music with stereo systems, much the way that software is bundled with computers. CD's will die. 2. Mass markets, which tend to go along with a scarcity of broadcast outlets, may give way to niche markets on the Internet. 3. Because so many people would like to make a living making music, a career in music will never be very lucrative. 4. The music industry as we know it will resist rather than take advantage of new technology. Therefore, it will collapse. I think that it is a mistake to equate intellectual property rights with the ability to make a profit by distributing music using CD's. What the music industry is fighting to protect is the latter. They will lose that fight. The music distribution market that emerges in its place will provide the benefits of property rights. But those benefits do not include high incomes for musicians. . . . . . .
Mobile Gaming Trend?Most of the time, I do not recognize a trend unless it is something that interests me personally. However, in spite of my lack of interest in it, I suspect that this sort of mobile gaming has potential. . . . . . .
Posted Tuesday, December 3, 2002 The Death of Newspapers, UpdatedWhen I wrote News of My Death in which I forecast the death of newspapers, a lot of dead-trees professionals emailed me to say that the the Internet will never be able to provide good local coverage. I think that blogs can definitely do a better job, and now John Hiler has put his money (or at least some of his time) where my mouth is.
Good luck, John! I think you've got a winning concept (which probably means that it will take a couple of years to catch on). PBS' Robert X. Cringely also is joining on the death-of-newspapers bandwagon.
Cringely argues that movies are safe from piracy because of the high fixed costs of production. I think that this is the wrong argument--if movies were easy to distribute and view over the Net, piracy would occur. I think that download time and loss due to compression are what is holding down movie piracy. I'm with Zimran Ahmed that Netflix looks like a good business model for quite a while. . . . . . .
The Game of GrowthMy latest essay, on economic growth, includes a game for the reader to play. Nonetheless, it is a sober philosophical piece on the problem of learning what works when trying to solve a problem that seems to have a nonlinear solution.
TCS waited until after the slow Thanksgiving week was over to publish the essay. Is it a useful overview, or just a leftover turkey? There is a place at the bottom of the essay to post your feedback. . . . . . .
Posted Monday, December 2, 2002 The Next Frontier in ComputingSome interesting thoughts from Zimran.
I think of this in terms of vertical market segments. Two of the most productivity-resistant sectors of the economy are health care and education. Both have done a pretty poor job so far of taking advantage of computer technology. Doctors' offices are just as bad as they were 4-1/2 years ago. My impression is that, at least for K-12, computers in education have actually lowered productivity, not even counting the opportunity cost of all the money that could have been spent on something useful. (I teach at two schools, and taking attendance is much more work at the one that requires me to enter data on the computer than the one that lets me use paper*). My guess is that at some point (soon?--I tend to be optimistic) the tide will turn, and we will see computers start to drive productivity improvements in education and health care. The key, as Zimran suggests, will be in getting the computers to adapt to the people who do the work in the industry, rather than trying to turn educators and health care professionals into geeks. (Yes, I know that there are individuals in both industries that could run circles around me as a programmer, but they are not the core component of health care or education. If anything, they are over-adapted to computers, to the detriment of their ability to stick to their core business.) *The system makes me go through a sequence of several selections, including scrolling through a drop-down list of every course in the entire school, in order to find my class. Now, I know that you are thinking that the paper-based system at some point requires somebody to enter data on to the computer. Fine, but that person is not a teacher. It is someone whose comparative advantage is in data entry. If you can't find a cheap clerical assistant here, you could fax my paper attendance sheet to India and let them enter the data. I mean, do you want your teachers futzing with computers or would you like them to teach? . . . . . .
The Trackable Society, UpdateA taste of what is to come.
What I've said is that I don't think our legal system can handle this. . . . . . .
Copyright 2002-2003 Arnold Kling. All rights reserved. Terms of use |
|
|
