|
|
|
|
|
|
|
|
|
Posted Friday, May 30, 2003 Anti-trust OxymoronDid you see the latest achievement of our anti-trust laws? The main outcome of the settlement between AOL and Microsoft is that the two giants are now colluding.
As a consumer, you may feel like a loser in the deal. But you should look at it from the point of view of the anti-trust lawyers. They get their fees, and that's what counts. . . . . . .
The Media Merger IssueThe more I think about it, the more I'm convinced that the populist outcry against deregulating media mergers is foolish. What the the populists want is for some alleged Eden of diverse local broadcasters to be restored. But that depends on consumers, not on merger policy. If mergers are deregulated, there are two possible outcomes. [1] Suppose that there really is consumer demand for diverse local programming. In that case, local franchises will be valuable, and this value will be demonstrated by the large amounts that big companies pay to obtain smaller ones. That in turn will increase entry into local media, adding more new broadcasters than are lost through merger. [2] Suppose that local broadcasters are losing audience and have very little revenue. Then they will merge in order to avoid bankruptcy. I think that [1] is unlikely and that [2] will occur. But you cannot save a losing business by protecting it from mergers. If scenario [2] is correct, then local media is being killed by the cold indifference of the consumer, not the predatory attacks of larger companies. . . . . . .
Posted Wednesday, May 28, 2003 Converting to Voice over IPIf Sprint is the tortoise, then this Canadian telco is the hare.
Thanks to common-sense for the pointer. . . . . . .
Posted Tuesday, May 27, 2003 Lots of Challenge with Challenge-ResponseDeclan McCullagh lists some issues with the challenge-response approach to blocking spam.
My guess is that at some point the ISP's are going to figure out that the storage space and bandwidth wasted on implementing challenge-response exceeds that used to process spam. The human bandwidth cost-benefit calculation is not going to show a profit, either. . . . . . .
Alternative IntermediationJoshua Ellis talks about how the music business might evolve.
Meanwhile, the music industry screams "send lawyers, guns, and money" (in Warren Zevon's phrase) to try to enforce its outmoded business model. . . . . . .
Sprint's IP Conversion--why bother?Here is a story of Sprint converting to a packet-based network.
I feel like that is too slow. I just don't think that there is going to be much of a circuit-switched phone system left by the year 2010. Thanks to Reason for the pointer. . . . . . .
Standardized Testing and School AccountabilityI make a case against it.
. . . . . .
Centrifugal Force, Con'tIn my view, one of the most important trends in computing is what I call centrifugal force--the breaking up of the PC into separate pieces. In the 1980's and 1990's, we had centripetal force--more and more functions were being consolidated into the personal computer. Now we have centrifugal force--with functions being split up into stand-alone hardware components. For example, there is this story about a personal server.
Centrifugal force. Threatens to break Windows. . . . . . .
Posted Monday, May 26, 2003 Uses for Social Network MapsLet's start with this random quote from Jamie Lewis (which I got to from Dan Bricklin):
Actually, I get perfectly good email from people who I don't know. But imagine that there were a very complete social network map sitting out there, kind of the way Google's database is sitting out there. The social map would show my connections to everyone, and everyone's connections to everyone else. Every time I get an email, my email program could search that social map and figure out how the sender is connected to me. My guess is that good email is more likely to come from people who are fewer degrees of separation away from me than the senders of spam email. That would not in and of itself be a perfect spam filter, but it would really improve existing spam filters, particularly for reducing the number of instances where good email is erroneously classified as spam. I also think that this hypothetical social network map could help reduce terrorism. I'll bet that the social network maps of terrorists are very different from the social network maps of ordinary folks. I'm thinkin that if you're a terrorist, you have to spend a lot of time interacting with other terrorists and not so much time interacting with the rest of us. . . . . . .
Posted Thursday, May 22, 2003 What do Media Mergers Mean?Most people see media mergers as a sign of strength. For example, William Safire writes,
I think that media mergers are a sign of decline. The major TV networks have seen their shares plummet over the years. Newspaper readership is stagnant or declining. I haven't seen data on radio audience, but I would be surprised if it is robust. The market is saying that we do not need as many broadcast television stations, radio stations, and newspapers as we used to. I can't say why this is the case, but I suspect that competition from video games, movie rentals, cable television, the Internet, cell phones (instead of listening to the radio in the car, we are listening to one another) and even books (more titles are published every year) is eating into old media. Given my view of the causes of the decline of newspapers, broadcast TV, and radio, trying to save those media by preventing mergers is pointless. Big media is failing, not succeeding. As an economist, I say let it fail. . . . . . .
Posted Tuesday, May 20, 2003 Economics of Grid ComputingClay Shirky tries to do some economic analysis of Grid Computing.
My point is not to hold this up as an example of sophisticated economics. But if IBM did anything even that sophisticated before they decided to throw megabucks at Grid Computing I'll be surprised. . . . . . .
A Challenge for Challenge-ResponseAlso in the dead-trees Technology Review, Simson Garfinkel casts a skeptical eye on the "challenge-response" approach to fighting spam, in which as a sender your email would be blocked until you demonstrate a human response to a computer's query.
Other than filters, I have not seen an anti-spam solution that imposes fewer costs on the email system than spam itself. . . . . . .
Speed Hump Nation?In the current dead-trees version of Technology Review, Edward Tenner offers a Weinbergerian critique of hard-coding law enforcement into technology.
I agree that technology is making it possible to hard-code more laws, and that this is not necessarily a good idea. However, I am skeptical that the solution is to stop technology. What I think is needed is a re-examination of how laws should be written. If you wrote a law while never dreaming that it would actually be enforced as written, it's time to rewrite the law. . . . . . .
Posted Monday, May 19, 2003 The Two-State SolutionSince there appears to be something of a deadlock on economic policy, I propose dividing the country in two.
. . . . . .
Bad News for McKinseyThe strategic advice that they always give--sacrifice profitability in order to gain market share--turns out to be wrong, according to two recent books. For example, Stan Liebowitz writes,
Actually, my views on McKinsey may be out of date--I have not had any dealings with them since the Internet bubble popped. Maybe these days they believe in better products at lower costs. But that does not provide as much leeway for creative strategy as going all out for market share. . . . . . .
Posted Saturday, May 17, 2003 DeLong, Rheingold, and MeActually, Brad and I were talking about Howard behind his back. I'll stop doing that, and focus on Brad's latest comments.
Brad is much more erudite than I am on these matters. I have a simplistic picture. I don't focus on the problem of "structuring collective action." I focus on the problem as the writers of the Federalist Papers seemed to see it: how do you protect individual autonomy from tyranny? Their solution was a system of checks and balances, embedded in a Constitution. I see the genius of the American system as this system of checks and balances. Elections are part of that system. I judge the effectiveness of the system primarily on the basis of its ability to protect the autonomy of the individual from tyranny, not on the basis of its ability to make "effective collective decisions," whatever that means. My hope for technology is that it will allow us to live in political environments that are more diverse and more fluid. If physical location becomes less of a factor, then maybe I can live among a majority of left-winger neighbors but in a different political "region" that would be more receptive to, say, school vouchers. People who want to live in a "region" that legalizes marijuana might be able to choose to do so. People disagree about political issues. Often, this requires collective action to sort out. However, in my view, the best system for collective decisions is a system that gets the "collective" out of the decision unless it is absolutely necessary. . . . . . .
Amazon Deletes My ReviewAmazon deleted my review of Andy Kessler's book. It was a negative review, and it was deleted at the request of the author. Kessler is scum. He is trying to profit even more from the dotcom bubble that he shamelessly helped to inflate. Clearly, he has a lot of influence at Amazon. Next time you read customer reviews on Amazon, note that if the author is enough of a jerk, he can get negative reviews deleted. . . . . . .
Posted Thursday, May 15, 2003 Speaking of Brad DeLongI feel like I have to blog this quote.
I don't think Howard Rheingold should try to get on a panel discussion with Brad any time soon. . . . . . .
Posted Wednesday, May 14, 2003 Back to the 19th Century?Brad DeLong has a pointer to an in-depth survey of information technology by The Economist. One of the main themes of the survey is that information technology is becoming a mature industry, so we might need to look to the process by which 19th-century industries matured in order to guide our outlook, as this piece in the survey says.
I remember attending one of the early pep rallies for Java, with a rocking sound track and a public-address announcer introducing Sun executives the way you would the starters in the NBA championship game ("Chief Technology Officer ... J-o-o-o-h-h-n GA-AGE"). Schmidt was there, and he asked for a show of hands for how many people thought that the Internet was underhyped. No one raised their hand. Schimdt raised his, and he said "I think the Internet is actually underhyped, and here's why..." My guess is that he kept using that riff for a bit longer than he now thinks wise, so now he's making a U-turn. But the 19th century might not be the best model to use. My reading of it is that the pace of change was much slower then. My own feeling is that The Economist is wrong. In my opinion, the revolution lives. I still think that banks that try to develop software are going down. I think that IBM is going down, because the economies of scale that exist in mainframe computing are not there in consulting. Social software, with matching and reputation systems, is going to destroy any remaining size advantage in the consulting industry. I think that Microsoft is going down, because the centripetal forces that drove innovation into the PC have changed into centrifugal forces that are driving innovation into peripherals and other hardware. The Economist makes it sound like information technology is going to be boring. I think that the outlook is downright interesting. . . . . . .
Oxymoronic Battle Against Media ConcentrationIf webloggers launch a campaign against media mergers, isn't that sort of an oxymoron? I mean, how can we say that big media is a threat when we're cleaning their clocks? Just a contrarian way of looking at the issues in Sister Wentworth's call to action. . . . . . .
Building the ultimate poker playerBrother Lynch points to a DARPA proposal to create a system that can recognize human emotions.
. . . . . .
Why stand-alone hard drives threaten WindowsEvery time Gizmodo runs a story about a stand-alone hard drive, such as this one for wireless networks, I see the concept of a complex operating system being threatened.
What is going to happen is that the hardware innovation cycle is going to get too fast for bloatware. It used to be that people who made peripherals worried about writing drivers for Windows. That game may be going away. Inventors are going to start thinking of PC operating systems as something to work around, not with. . . . . . .
Copying != TheftThe famous David Weinberger writes,
The not-yet-famous Tom Bell writes,
Weinberger is arguing that DRM would "end copyright as we know it" by making it enforceable without taking legal action. Maybe that is a fair trade-off: get government out of the copyright enforcement business in cases where firms use DRM. You could argue that if authors can use DRM, then they don't need the "welfare system" of copyright.
. . . . . .
Go and VoteAfter I wrote a review on Amazon of a new book by Andy Kessler, he emailed me saying that if I do not remove my review, he will have my status as a top 1000 reviewer taken away by Amazon. Go read the review and vote on it. If most people say it's not helpful, I'll take it down. UPDATE: Your votes don't count, I guess. Amazon deleted the review, presumably at Kessler's insistence. . . . . . .
Posted Saturday, May 10, 2003 Oh, brother!My Corante compadres sometimes write things with which I beg to differ. That's fine. This is a big tent and all. But Brother Blankenhorn really hit the trifecta with this post.
Not really. See Oil Econ 101. Blankenhorn continues,
Not really. See Lynne Kiesling's analysis, which I featured here, with a follow-up here. Finally, Blankenhorn tops it off with this brainstorm:
Well, in a way Doerr would be the perfect guy to squander billions of dollars of other people's money by taking bad gambles. After all, that is what he did in the dotcom era, helping to float a host of bubble-companies where he cashed out handsomely before they were mis-managed into oblivion. If you think of late-90's finance in gangster terms, Jack Grubman is like the little street punk that the mob handed over to help the DA look good. Doerr is like the kingpin who got off scot-free. . . . . . .
Posted Friday, May 9, 2003 Advantage KurzweilWhen I first read The Age of Spiritual Machines, I really did not believe that Moore's Law could continue to hold much longer. But read this.
The article goes on to list technologies that IBM thinks will come on stream down the road, including nanotube, nanodisk, something called Molecular Cascade, and the quantum computer. . . . . . .
File-swapping, Intellectual Property, Etc.
. . . . . .
Posted Thursday, May 8, 2003 Emotional NoneconomicsBrother Lynch waxes enthusiastic for emotional economics.
I absolutely agree that people are emotional and make irrational decisions many times. However, the agenda of many of the opponents of classical economics is to replace the irrational decisions of free individuals with the irrational decisions of elitist intellectuals holding positions of power in government. As an economist, my world view does not rely on the assumption that people make correct decisions. Mistakes are part of life. Many of them end up costing people, and some of them end up paying off. However, I insist that we should be entitled to make our own mistakes. The freedom of people to make their own decisions--some rational, some not--is absolutely necessary if people are to live in dignity and prosperity. The last attempt to overthrow mainstream economics--Marxism--led to a totalitarian disaster. To repeat that mistake would be the most irrational move of all. UPDATE: Brad DeLong tries to pop some of the media bubble about behavioral economics. . . . . . .
Promising Social SoftwareRoss Mayfield is among those buzzing about LinkedIn, a referral network. This is an example of promising social software, because it is directed at a relevant problem, the one that I call matching. If you are trying to match a job opening with a candidate, the challenge is to get the word out to a lot of people without getting spammed by people you do not really want to hire. A solution is to have other people act as intermediaries in a referral process. But how do you know a referrer won't spam you? The more closely the referrer is linked to you, the more that your long-term interests are aligned. Brother Mayfield says,
Implicitly, he is dismissing it for non-business matching models, such as dating. But in fact, people set their friends up on dates all the time. Off-hand, I don't see why a web-of-trust approach couldn't work in the dating space. I'm encouraged by anybody that is trying to solve the matching problem. Moreover, a system that helps to automate a referral network sounds promising. In Theory. But it's not so easy to put together in practice. I haven't tried LinkedIn, but other services I've played with have turned out to be much more trouble to work with than they are worth. . . . . . .
Posted Wednesday, May 7, 2003 Big Challenge, Bad ResponseEarthlink announced a major anti-spam initiative.
My guess is that the cost of implementation will exceed the cost of spam. And I'd be curious how this software deals with, say, an automatically-generated newsletter that I sign up to receive. . . . . . .
Stand-alone Hard DriveVia Gizmodo comes this story.
This may turn out to be just a curiosity or niche product. But it also could be a harbinger of the unbundling of the personal computer and the end of Windows, which would make it a slightly bigger deal. We think of a PC as having a processor, a hard drive, a monitor, speakers, a mouse, a key board, and a DVD/CD something-or-other. All tied together by an operating system, which of course comes from the Evil Empire. But in a mobile world, we may want to leave some of that apparatus behind. Perhaps we can find a screen on the road. Maybe we'll go for a virtual keyboard. Maybe we'll take a picture on a digital camera and immediately send it over the Net to a stand-alone hard disk. Maybe we will execute complex applications by linking together disparate devices, each with a simple operating system, instead of requiring a single complex operating system to knit everything together. Maybe. . . . . . .
Choice of Web Site SoftwareDavid Strom passes along a decision diary of Tristan Louis, who became frustrated with a Microsoft server configuration and searched for an alternative. It is a useful discussion of the thought process involved. A sample quote:
I really like the way Louis approached the whole issue. For example, his decision to rewrite code rather than try to hang on to legacy code is a sound one. Unfortunately, at a large corporation you can run into a lot of bureaucratic resistance to doing code rewrites--and then have to deal with complaints that you cannot make simple changes. Rewriting code helps clean it up and prepare it for the future. The IT shops who used the Y2K scare as an excuse to go in and completely update their old code probably saved their corporations a lot of money in the long run. I faced a crisis in late 1996 when the Homefair site was running the Netscape enterprise server and crashing every 90 seconds. I wanted to develop using ASP, but I valued stability, and at that time Sun Solaris was much more crash-resistant than Windows NT. For server-side scripting, that left Perl or Java. I chose the latter, because I felt that it attracted team-oriented programmers who value good documentation, whereas Perl tended to attract documentation-allergic cowboys. Today, my guess is that I would go with Louis' choice of PHP. However, I would investigate one other issue, which is debugging. Java has a strict compiler, which ultimately speeds the coding process because it is harder to get buggy code past the compiler. . . . . . .
Posted Tuesday, May 6, 2003 Bearish on Palm-thingies, Wi-FiI was one of the original bears on Palm-thingies, and one of the original inventors, Jeff Hawkins, now has turned bearish.
It's interesting that he thinks that ordinary cell phone service is going to be able to hold back the assault from Wi-Fi. I think that a lot of people are betting the other way. His arguments are worth reading. . . . . . .
Does CSS Empower?Paul Philip watches a controversy break out over Cascading Style Sheets (CSS). I have a somewhat different perspective on CSS. I taught a high school class last year called "Basic Web Design," in which I spent a lot of time on CSS. The students did not like it. They knew, instinctively, that there were other techniques that made fancy-looking pages easier to develop. It was at that point that I realized that I should have called my class "Web Engineering" rather than "Web Design." CSS is an engineer's solution to web design, not a designer's solution. It has no appeal to the vast majority of people who design visually. It has the same problem that client-side Java had--many web coders who were getting along fine without it find it disempowering because it is so difficult compared to plain HTML. CSS happens to appeal to me because of the web site that I ran back in the 90's. We had hundreds of "co-brands," in which we adapted our content to the look and feel of other sites. Had CSS been available, it would have greatly reduced the effort required to create and manage these co-branded versions of our site. But for the vast majority of people with web pages, CSS is not an empowering technology. It is difficult to learn, and extremely difficult to de-bug. In a world where most of the web is created by non-engineers, CSS is a loser. . . . . . .
Posted Monday, May 5, 2003 Kass raises questionsLeon Kass, the head of President Bush's commission on bioethics, is notoriously antagonistic toward biotech. Here is a recent essay, in an interesting journal called The New Atlantis.
I tend to take the opposite side. Kass's position itself raises a lot of questions, particularly of the where-do-you-draw-the-line sort. But the essay is worth reading. . . . . . .
Google's Management PhilosophyFascinating interview with CEO Eric Schmidt.
Translation: we don't value McKinsey's employees more than our own.
Translation: we didn't by Blogger to aggregate eyeballs.
Translation: pay attention to Corante's Amateur Hour Pointer from Prashant Kothari, who in turn credits Rajesh Jain . . . . . .
Posted Sunday, May 4, 2003 Scientists vs. HumanistsMy Moore vs. Plato essay is not the only recent comment on this subject. You can read this article in the Chronicle of Higher Education (link from Arts and Letters Daily).
Or several posts from 'Jane Galt,' starting with this one:
And Steven Den Beste gets into the act.
And I successfully trolled David Weinberger, who mocks the whole issue.
. . . . . .
Posted Friday, May 2, 2003 Social Software BlogIn case you haven't seen it, I've got some new brothers and sisters over at Many-to-Many, a blog about social software. One of them, Clay Shirky, writes,
My bottomline perspective on social software is that I want to see it solve meaningful problems. I think that there are meaningful problems out there. Are the socialites pragmatic enough to focus on them? . . . . . .
More Important than Plato?I argue that Moore's Law is a big deal. . . . . . .
Copyright 2002-2003 Arnold Kling. All rights reserved. Terms of use |
|
|
