|
|
|
|
|
|
|
|
|
Posted Monday, June 30, 2003 Where is Cometa?Am I the only one who remembers reading this?
I figured they must have made some progress by now, so I went back to their web site. They've changed their story a tad.
Three years seems like a long time. . . . . . .
Posted Friday, June 27, 2003 Do-Not-Call List is a Dumb ApproachI tried to get to the do-not-call registry, and so far the server is overwhelmed. Is this a surprise? It would be much simpler to manage a list of people who do want to be called by telemarketers. My guess is that you would need much less bandwidth and a much smaller database to manage that list. This is even more true for spam. I would not even sign up for a "do not spam" list, because spammers would just harvest that list for email addresses. If any sort of database should be set up for spam marketers, it should be a list of all the people who would like to receive spam. Unfortunately, I do not know of any way of restricting spammers to sending email only to registered spam-lovers. . . . . . .
Posted Wednesday, June 25, 2003 Medicine, Paternalism, and Social SoftwareSome discussions of distortions in health care markets were stimulated by stuff that I wrote here and here about prescription drug pricing. This led me to think about the whole subject of paternalism in medicine, which led me to think about whether reputation systems and social software could provide an alternative. My thoughts are disorganized, but here goes. 1. One reason that we have such a concentrated drug industry is because FDA regulations are so complex. The FDA protects big pharma, because you can't start a drug company without a huge cadre of lawyers and other experts in running the regulatory gauntlet. In the absence of the FDA, drug discovery, like other trial-and-error processes, would probably be done better by smaller, decentralized firms. 2. When it comes to public policy, doctors are very biased "experts." They have the gall to complain about the fact that drug companies devote resources to marketing to doctors, when the reason the drug companies do so is that doctors have a chokehold on the drug market. If you eliminated the requirement for a doctor's prescription for medications, you would not see drug companies marketing so heavily to doctors. But if you're waiting for doctors to recommend that solution, don't hold your breath. The point here is that paternalism is responsible for a lot of the distortions in the health care industry. We won't let unregulated drugs on the market. We won't let anyone but licensed professionals into the health care service market, including writing prescriptions. But is there any alternative to paternalism? Yes! Reputation systems! I could base my decision to use a drug on the reputation of the drug itself. If it's a new drug, I could base my decision on the company's reputation for research, as well as the opinions of scientists about the drug. I would want reputation systems evaluating the scientists. I could base my decision to use the drug on the recommendation of an expert (perhaps someone trained in medical school, but perhaps someone else). I want reputation systems evaluating medical experts. The point here is that a really thick layer of reputation systems might substitute for paternalistic regulation. Now we all know that people have incentives to game reputation systems. So the reputation systems would need to have mechanisms that resist gaming. See Howard's book for a discussion of that issue. Instead of thinking of social software as a way to mobilize street demonstrations, let's think of it as a way to make paternalistic regulations obsolete. . . . . . .
RIAA Threatens to Go to the MattressesThe RIAA is threatening to go after individual file sharers.
They'd better hope that the lawsuits aren't decided by juries. . . . . . .
No Such Thing as BroadbandOne way to describe Bob Frankston's latest must-read essay is that he is saying that there is no such thing as broadband. That is, it is misleading to think about "broadband" as a defined service. Let me stop trying to put words in Frankston's mouth. Here are a few of his words.
Now, I recommend that you read the rest yourself. . . . . . .
Posted Monday, June 23, 2003 Moore's Law and Car StereosUsing Moore's Law, I've predicted that something like this would be the future of music in cars.
Meanwhile, the clueless car companies are trying satellite radio. Losers. . . . . . .
Horizontal KnowledgeThe term comes from Glenn Reynolds. He never defines it exactly, but I think that what he means is a combination of decentralization and diffusion of knowledge. One consequence is that enables amateurs to compete with professionals in ways not previously imagined. In this follow-up, Reynolds discusses the role of weblogs in humbling the New York Times during the Jayson Blair scandal. Actually, I think a better example is the way that command-post.org outperformed the Times and other media during the Iraq war. Matt Welch reviews the book Moneyball, which describes how horizontal knowledge affected baseball.
'Mindles H. Dreck' notes Welch's review, and asks whether this phenomenon might have some application to the stock market. He quotes Michael Lewis on Jonathan Lebed, a teenager who was prosecuted for doing as an amateur what professionals appear to do all the time--traffic in rumors about stocks.
I would note, however, that baseball was a monopoly industry, and as such could resist improvements (including black talent) for a long time. Wall Street is more competitive, so it may be more difficult for amateurs to find unexploited informational advantages. . . . . . .
Posted Friday, June 20, 2003 Programmer Clientele EffectsBrother Brockmeier debates Rob Enderle. Enderle is making the case against Linux for the enterprise. One of Enderle's concerns is the personalities of programmers who are attracted to Linux.
What Enderle is doing is type-casting Linux users as rebellious and unwilling to compromise to conform to corporate needs. This may seem unfair. However, I think that programming languages do have a tendency to sort developers into clienteles. For example, in the late 1990's when I ran a web site, I shied away from Perl as a language, because every time I encountered a site powered by Perl it was the work of a single, arrogant programmer. The site's scripts would be brilliant, efficient--and completely undocumented and impossible to maintain by someone else. In contrast, Java programmers tended to be compulsive about documentation and maintainability. As far as I know, there is nothing about Java itself that forces a programmer to think in terms of a team setting or business continuity. There is nothing that prevents Perl from being used in a professional, team-oriented way. But programmers who were motivated to learn Java tended to be developers with a willingness to work in team settings in a business context. Perl tended to appeal to lone cowboys. I can imagine an IT manager who selects Linux running into these sorts of problems with clientele effects. When you're running a business IT shop, you have to have people who can pull together as a team and who can compromise when necessary. If Linux brings with it people who have a chip on their shoulder, or an uncompromising arrogance, then that is a very serious issue. My experience in software is that projects do not fail because of the wrong choice of programming tool. They fail because of breakdowns in communication among personnel. To encourage communication, it is essential to have programmers who show respect for everyone involved in a project, including people involved in non-technical functions such as marketing. To the extent that Linux developers are less inclined to show such respect, IT managers would do well to steer away from Linux. . . . . . .
Posted Wednesday, June 18, 2003 Email EfficiencyBob Frankston can say an awful lot in a snippet.
This is kind of like what I said in this essay, when I made a point of talking about basic email efficiency.
As a random thought, I would like to be able to classify my incoming email in more than one dimension. The folder captures one dimension. But another dimension is how long I want to save the email. Another dimension is when I want the email to pop up gain as a reminder. For example, if my Dad emails me his travel plans to San Francisco for July, I would like my email program to automatically bring that email back to my attention the day before his trip and automatically delete the email the day after his trip. There probably are better ways of thinking about this whole information-classification issue. But the point is that there is a generic problem here, and spam filtering is only part of it. . . . . . .
Posted Tuesday, June 17, 2003 Provocative Analysis of Technology TrendsI think that this is one of the most important essays that I've written.
Read the whole thing, and let me know if I'm overstating it. . . . . . .
When Spammers are ProfessionalsI think that the people who are worked up over spam are spending too much time worrying about the technical details and not enough time thinking about the economics. They should follow the money. For some recent background on spam issues, see Silicon.com's series, this Washington Post story on legislation, and this article from MIT Technology Review. The Tech Review story has some interesting comments.
What this suggests to me is that spammers are highly skilled professionals. In that case, to go after them you want skilled professional law enforcement agents, who can engage in infiltration operations. These agents should "follow the money" by setting up email addresses that attract spam, making purchases, and finding out who gets the money. Legislation, if needed, should target the repeat offenders. That is, penalties for being involved in taking money via spam should get progressively higher the more times that a person is caught. I think that the main point is that although spam is widespread and there are many victims, it is not necessary or even a good idea to give legal tools to each victim. Victim-initiated action, such as consumer lawsuits, will be haphazard and probably end up hitting the easiest spammers to find, without touching the true pros. Such efforts will be a huge waste of time and money. To fight spam in the legal system, we need pros fighting pros. Not lawsuits and other amateur tactics. . . . . . .
Posted Monday, June 16, 2003 New InterfaceVia the indispensible gizmodo, this looks cool.
. . . . . .
Gelernter's HammerDavid Gelernter describes his vision of a web-based newspaper.
Gelernter uses this index-card metaphor for anything having to do with computer interfaces. He is like the kid with a hammer, to whom anything looks like a nail. Bloggers also think that they have a hammer. The difference is that ours is out there working, and Gelernter's is still a concept. . . . . . .
Posted Sunday, June 15, 2003 Cell Phone as Smart CardNot long ago, people thought that it would be smart cards would be doing this kind of stuff.
But what this is talking about is technology proposed for a cell phone. It seems to me that they are a bit ahead of themselves here. It seems to me that getting cell phones to be able to routinely handle credit-card and debit-card transactions is the first step. Getting merchants to accept cell-phone-powered transactions is another stop. The agent powered by artificial intelligence can come much later. . . . . . .
Posted Thursday, June 12, 2003 Living with SpamI argue that the alternatives are worse. . . . . . .
Posted Wednesday, June 11, 2003 Trends to WatchPotomac Techwire points to a story about putting a barcode scanner on a cell phone.
This combines two important trends. One is the continuing migration of computing functionality away from personal computers and onto other platforms--in this case a cell phone. A second important trend is the focus on transaction capability rather than document production capability. The killer application in mobile computing is not transmitting stock quotes. It is not producing web logs on the go. It is executing transactions. . . . . . .
Robotic Home Health Care?This story finds Dr. Joseph F. Engelberger bullish on the concept.
Of course, if what old people really want is sympathy and company, robots may not be quite such a boon. . . . . . .
Yahoo's Challenge-ResponseLawrence Lee points to this story.
I would think that if you want to stop your users from sending spam, there would be a lot of good techniques available. Maybe this is part of a package. . . . . . .
Posted Tuesday, June 10, 2003 Lying to the Grand InquisitorThe front-page story in today's Washington Post begins,
Recently, we saw Martha Stewart charged not with a crime, but with lying to investigators about an alleged crime. This raised questions in some people's minds about the quality of the government's case. Even more questions should be raised about the firing of David Glenn from Freddie Mac. As best I understand it, Freddie Mac's alleged accounting problems consist of the following: As a mortgage lender, Freddie Mac faces the fact that the spread between the earnings it gets from a mortgage and its cost of borrowing to fund the mortgage tends to decline over the life of the mortgage. This decline comes at an unpredictable rate, depending on the level of interest rates (the profit spread can even decline and turn negative, as it has at times for Fannie Mae, a similar agency). Freddie Mac executes complex hedge transactions, which have the effect of reducing the spread up front and reducing the risk of decline over time. These transactions increase the safety of the company. Freddie Mac evidently attempted to have the accounting for hedges match the economics--that is, they wanted the hedges to reduce earnings in the near term. However, after Arthur Anderson imploded, Freddie Mac changed auditors, and the new auditor wanted a different accounting treatment. The result probably will be that Freddie Mac restates its earnings upward now, but its accounting earnings may be more volatile in the future, because the accounting treatment will no longer line up with the economics. In other words, my guess is that the accounting treatment that Freddie Mac was using all along probably is better than the one that the new auditors--PWC--wish to use. This "scandal" is the subject of an investigation. And Mr. Glenn apparently displeased the Grand Inquisitor because the personal notes that he provided were altered or incomplete. I was an employee at Freddie Mac in the late 1980's and early 1990's. I had limited dealings with Mr. Glenn, but my sense was that he is a straight shooter. My guess is that the Grand Inquisitor is not going to find a real crime. I do not think that we want to become a society in which anyone can be pilloried for failing to provide accurate personal notes to the Grand Inquisitor, regardless of whether or not there is an underlying crime. . . . . . .
Posted Sunday, June 8, 2003 Laissez-faire on SpamMark Hurst bravely makes the case for laissez-faire on spam.
Basically, it takes discipline to manage your email. If you're one of those people who lets everything sit in your "inbox" for weeks, then you are going to be an unproductive email user no matter how well spam gets blocked. I use a spam filtering system that is sort of like Mark's. However, I combine Popfile with a rule-based system. When the rule-based system and Popfile agree that a message is not spam, I read it. When they agree that it is spam, I delete it automatically. When they disagree, I scan the subject line. Usually, when they disagree, Popfile is right, but not always. UPDATE: In contrast to laissez-faire, we have Christopher Caldwell saying,
I'm sorry Mr. Caldwell, but the Internet's decentralized, hard-to-regulate structure is a feature, not a bug. Get over it.
. . . . . .
Posted Friday, June 6, 2003 Consumers Win OneA transfer of ownership rights from cell phone companies to consumers, thanks to Michael Powell's FCC.
. . . . . .
Good luck, ClayBrother Shirky tries to use economic-type arguments to suggest that some concentration is inevitable in media.
His essay makes sense to me. But I don't think that coherence is a priority among the people who think that easing the restrictions on newspaper/television ownership is a plot to turn us all into slaves of Rupert Murdoch. . . . . . .
VC's vs. Entrepreneurs
Entrepreneurs have a difficult time understanding this, and they have an even harder time accepting it. Moreover, as Zimran points out, entrepreneurs are afflicted with a bias that means that they under-estimate the probability of failure. The entrepreneur who really wants reasonable success with high probability belongs in what I call the "under the Radar" region. You need to accept the fact that you will get little or no funding for your business. If your idea requires a lot of funding but promises only a relatively small return even if successful, then tear up your napkin and come up with something else. If what you want to do is start a business with a high probability of success, then sift through ideas until you find one that does not require much funding. If you can't sift through ideas and discard the ones that don't work, then keep your day job. As far as I'm concerned, somebody who gets stuck on one idea is not an entrepreneur. He's a dreamer. . . . . . .
Netpreneur Post-MortemHere is a story about the final meeting of an organization that flourished during the dotcom heyday, To be fair, it was formed around 1995, and struggled for a few years. And it accomplished a lot--I benefited considerably from participating. I came away from the session with a bit of a sour taste. A lot of the speakers' companies have tanked, and yet their egos did not seem to me to have been marked down from their peak. But more than that, the locale for the session--an area called Reston Town Center--is one of those places that really offends me. It's a yuppie paradise sitting in what properly ought to be the middle of nowhere. Fancy restaurants, "upscale" shops, and "village-style sidewalks" which you can only access by driving there on a freeway and parking in one of several multi-story garages. As plastic fantastic as Cancun or Disney World. If Reston Town Center is what our society's wealth generates, then it's enough to make me stop writing essays like this and turn into a flaming socialist. . . . . . .
Posted Thursday, June 5, 2003 The Sky is FallingOrdinarily, I tend to denounce the "sky is falling" warnings from Lawrence Lessig and others about corporate power threatening to destroy the Internet. But this story bolsters their case.
This just makes me sick. It's search and seizure without a warrant by an organization that has no status in law enforcement. Ugh. . . . . . .
The Paper-thin ComputerA random, visionary proposal.
Thanks to Howard's Smartmobs blog for the pointer. . . . . . .
Posted Monday, June 2, 2003 Fundamental Attribution ErrorOne of the most important findings in social psychology is known as the Fundamental Attribution Error. People over-estimate the importance of individual character and under-estimate the importance of context as determinants of behavior. In this essay, I apply the attribution error to economic matters.
. . . . . .
Copyright 2002-2003 Arnold Kling. All rights reserved. Terms of use |
|
|
