Stephen Laniel’s Unspecified Bunker

I always like to make sure that I’m not using a technological solution for a non-technical problem, but I wonder: doesn’t it seem reasonable that the Net’s easy access to information will make it harder to lie about demonstrable facts? I was thinking just now of Microsoft’s habitual spreading of FUD (“Fear, Uncertainty, and Doubt”) about its competitors, lately Linux. The meme they’ve been spreading is that no one will support Linux, no one will make sure that it has decent security, etc. Clearly this is false, and a quick Google will confirm that it’s false. Should we expect that false statements will be proved false more quickly on the Net than in, say, the print world?

One of the big dangers of offloading democratic government into the private market is that the market is under no obligation to be democratic. Indeed, quite often private companies have an incentive to be as closed with their information as possible.

I’m reminded of this as I read a New York Times article describing the robust sales of DVDs abroad, and the motion-picture industry’s refusal to disclose how well they’re doing.

Let’s look at the logic here, focusing in particular on one law. If you want to create a DVD player, you’ll need to find some way around the Content Scrambling System that encrypts that DVD. Normally this means getting a license from the DVD Copy Control Association. That license is expensive; a little googling suggests it’s in the tens of thousands of dollars.

So what do you do if you’re a lone hacker and you want to write your own DVD player for Windows, the Mac, or Linux? You can’t afford to pay the CCA. So an alternative is to hack your way around the CCA’s encryption. That’s not terribly hard: one MIT student who interviewed the MPAA’s Jack Valenti did it in six lines. (It’s called qrpff. I’ve also cached a copy.)

But forget about hacking your way around it, because that’s illegal under the Digital Millennium Copyright Act. You want to do something that ought to be really simple and cheap and unencumbered by the law, but you can’t.

Now the DMCA exists because the MPAA and other organizations insisted that they needed it. Without it, they said, our “digital millennium” would see the end of the movie industry. (They said similar things about the Betamax, incidentally. Jack Valenti is famous for having said that “the VCR is to the American film producer and the American public as the Boston strangler is to the woman home alone.” Moving on  . . . )

Tie it all back together: the MPAA has insisted on new onerous laws that forbid you from doing what you really ought to be able to do, because they swear that they’ll die without it; yet when asked, they refuse to offer any evidence that they’re suffering any harm from the Internet. They may in fact be thriving. In all likelihood, there’s no good reason why you and I are constrained.

The overarching point here, which Larry Lessig made most forcefully and eloquently in Code and Other Laws of Cyberspace, is that technology and law go hand in hand. Your rights are constrained by the DVD encryption just as much as they would be by the government itself. If we’re going to offload our laws onto private actors — say, we’ll let the technology do our copyright enforcement for us, and give legal protection to the technology — then we ought to insist on democratic accountability for the technology. I’d go further and insist that we minimize this sort of offloading, but openness in technology is an obvious first baby step.

Finally someone stated the obvious:

Indeed, the government cannot even articulate at this moment how it will determine when the war on terrorism has ended  . . .  The government has conceded that the war could last several generations, thereby making it possible, if not likely, that ‘enemy combatants’ will be subject to terms of life imprisonment at Guantánamo Bay.

Judge Joyce H. Green’s memorandum opinion (75-page, 3.1-megabyte PDF) and order (1-page, 37K PDF) are available online. I’ve cached both of them.

I was just looking through one of my old email folders, trying to figure out when one of my best friends left for China. (I was trying to remember the specific event of his going to the airport.) Along the way, I found an email message from him sending along the purchase order from one of his superiors, buying him a new laptop. The email read as follows:

ThinkPad T21
Processor type Pentium III
Processor speed 800 MHz
Processor cache 256 KB
Display (Std) 14.1 in XGA (1024×768) TFT Display
Hard drive type EIDE; 20GB
Memory 128 MB
Maximum memory 512 MB
Operating system Microsoft Windows 2000
Optical drive 8X-2.3XDVD
264787U $3,149.00 1 $3,149.00
IBM 8×4x24 Max CD-RW Ultrabay 2000 Drive (P/N 08K9569) 08K9569 $379.00 1
Carrying Bag w/ laptop sleeve $147.00
A total of: $3675

I would add some memory (about $100) and a copy of Microsoft Office Pro.

Total: about $4250.00

Without the carrying bag or Microsoft Office, that looks to be about a $3,775 machine. The date on the email was March 14, 2001. By contrast, I got a ThinkPad R51 with a 1.5-GHz Pentium M, 512 megs of RAM, a 40-gig hard drive, and so forth  . . .  for $1,378. Goddamn.

Even assuming that my new machine is only (1.5 GHz/800 MHz =) twice as beasty as the 2001 machine from above, that’s about a 5x drop in quality-adjusted price over three and a half years. If trends hold, a new laptop with the same quality as mine will cost something like $250 in mid-2008. I love progress.

From Randy Gallistel:

Adaptive specialization of mechanisms is so ubiquitous and so obvious in biology, at every level of analysis, and for every kind of function, that no one thinks it necessary to call attention to it as a general principle about biological mechanisms.

In this light, it is odd but true that most past and present contemporary theorizing about learning does not assume that learning mechanisms are adaptively specialized for the solution of particular kinds of problems. Most theorizing assumes that there is a general-purpose learning process in the brain, a process adapted only to solving the problem of learning. [ . . . ] From a biological perspective, this assumption is equivalent to assuming that there is a general-purpose sensory organ that solves the problem of sensing (Gallistel 1999:1179).

Discuss.

In response to an article on the Social Security ‘race argument’ in today’s Boston Globe (the race argument being the claim that Social Security gives fewer benefits to blacks because they tend to die sooner than whites), I started digging around for citations in response to that claim. Paul Krugman’s piece “Little Black Lies” from January 28th includes the quote

Mr. Bush’s argument goes back at least seven years, to a report issued by the Heritage Foundation — a report so badly misleading that the deputy chief actuary (now the chief actuary) of the Social Security Administration wrote a memo pointing out “major errors in the methodology.” That’s actuary-speak for “damned lies.”

In fact, the actuary said, “careful research reflecting actual work histories for workers by race indicate that the nonwhite population actually enjoys the same or better expected rates of return from Social Security” as whites.

Googling for that quote from the actuary brings up a paper from the Center on Budget and Policy Priorities entitled “Social Security and People of Color” from 1998 (my cache). It includes the text of the (February 4, 1998) memo from the Social Security Administration’s Deputy Chief Actuary, which I’ve pulled out of the CBPP’s report. Its summary reads

More importantly, the methods utilized by the authors bias expected rates of return downward to a far greater degree for African Americans than for the general population. This study thus erroneously indicates differences in rates of return by race that are greatly exaggerated. In fact, results from more careful research reflecting actual work histories for workers by race indicate that the non-white population actually enjoys the same or better expected rates of return from Social Security that for the white population. (See Duggan, et. al., “The Returns Paid to Early Social Security Cohorts,” Contemporary Policy Issues, (October), pp. 1-13.)

The original Heritage Foundation report, entitled “Social Security’s Rate of Return”, is available on the web (and I’ve cached it). Finally, the Duggan report is on the web; I’ve cached it as well.

I emailed the author of the Globe article to see if he knows of any good arguments in favor of the race argument. From my end, it looks like a pure conservative invention.

Microsoft has decided to publish its XML schemas for Office 2003. Does this mean that it will now be possible for OpenOffice to export to Office 2003 format with absolutely no loss of formatting?

Is this just an extension of what Microsoft did in December of 2003?

In honor of Captain Beefheart, from now on Monday will be known within Steve Laniel, Inc. as “Tropical Hot Dog Day Day.” All Steve Laniel Inc. communiqués will feature this new name. It will no longer be possible to get “a case of the Mondays.” One now gets “a case of the Tropical Hot Dog Day Days,” which I think we can all agree is much more enjoyable.

Spread the word.

I still can’t get enough of the 2004 Red Sox. I found a link to a Malcolm Gladwell article that I wanted to read, and when I clicked on “printer-friendly page” I got a Sox recap (my cache). What great fun to read again about how much the Sox dominated the final 56% of the ALCS. I can’t wait for the 2005 season to start.

Everything I’ve read about Michael Crichton’s newest screed against environmentalists, State of Fear, makes me loathe it. See Michiko Kakutani’s review and Bruce Barcott’s. State of Fear appears based on the premise that global warming is a hoax and that environmentalists are planning a climatological disaster to bring some sympathy to their cause. It just sounds bottomlessly self-righteous, in a really annoying way.

I hate to prejudge a book  . . .  well, no I don’t. If I were to compile a list of all the books that I want to read, even the list of those that existed before I was born would run to a few hundred pages. I can write off Crichton and not feel bad about it in the least.

 . . . live, in Brazil.

I remember reading about Jefferson’s complaints about the early White House. Ordinary people would knock on the door, and demand to see the President. Often they did. The presumption of that democracy lives in a sense here. And you never quite see how far from that presumption our democracy has become until you see it, live, here. “This is what democracy looks like.” Or at least, a democracy where the leaders can stand packed in the middle of a crowd, with protesters yelling angry criticism yet without “security” silencing the noise. No guns, no men in black uniform, no panic, and plenty of press. Just imagine.

(Via David Isenberg, though I’m not clear why I missed it the first time ‘round on Lessig’s blog.)

 . . . which kicks ass. If I see Billy Crystal again, I’m going to choke on forced neurosis.

One of the irritating bits about WiFi is that there’s such a limited number of channels available — ostensibly 11 such channels, but I understand they’re so close together that only three are really usable. And everywhere I go with my laptop, I see a bunch of open wireless access points using channel 6, which means there are probably a bunch of wireless customers getting frequently cut off in the middle of their surfing.

Isn’t there some smarter way around this? Maybe intelligent channel-hopping (I think that’s “spread spectrum,” though I’m not sure). Maybe assigning channels at random when the router ships. Maybe changing the channel at random periodically. Maybe paying attention to the used channels in the area, and switching as far away from them as possible (with some clever way around the potential infinite loop that this would create).

It seems that the technology wasn’t developed with any realization of how popular and cheap it would become. Is there any way to retrofit it? Could router firmware be hacked to use some of the ideas listed above, or are those the kind of changes that can only be grafted on by changing the hardware?

Because lesbianism will corrupt our youth, the new secretary of education has denounced a lesbian character in a PBS show. Ostensibly unrelated to her denunciation, PBS has pulled the show (for now?). PBS says,

Ultimately, our decision was based on the fact that we recognize this is a sensitive issue, and we wanted to make sure that parents had an opportunity to introduce this subject to their children in their own time

I’m really looking forward to the day that this statement sounds as absurd as if the subject were gravity: “We understand that the concept of two masses being drawn to each other in proportion to the product of their mass and inverse proportion to the square of their distance could be shocking to some, and we want to give parents time to teach it to their kids.”

What parents need to tell their kids is, “Some boys like boys more than girls, and some girls like girls more than boys. We call these people ‘gay.’ If you feel gay, it’s okay; we’ll support you because we love you, and we’ll love you no matter what. Though you might run into people who act mean toward gay people. Just ignore them; they’re not the kind of people you should be hanging out with.”

There, that was easy. Now on with the show.

Jack Balkin’s weblog has been doing a fantastic job keeping up with the nomination of Judge Gonzales. Today Marty Lederman continues, detailing Gonzales’s responses to the latest round of questioning (my cache). He’s apparently more candid, but also quite frightening. Not only can the White House choose to withhold documents from Congress whenever it deems such a thing necessary (thereby voiding much of Congress’s purpose), but it also continues to assert that torture is too squishy a concept to define clearly — even though the “shocking to the conscience” standard holds within the United States. I’ve been wondering the same thing that Lederman asks: why should American citizens be able to treat people inhumanely abroad when it’s illegal for them to do so domestically?

InfoWorld reports (via Slashdot) that IBM’s internal Linux migration isn’t going as well as planned, because a) the internal support staff don’t know how to support it very well; b) one of their own apps only works under IE; and c) Wine isn’t working so well for them.

I’d like to see IBM “eat its own dogfood,” combined with a strong devotion to getting Linux on the desktop — which means perfecting Wine and porting all their apps to work without IE. We’ll get by without IBM, but their help would certainly move us along.

Josh Marshall appeared on Al Franken’s show today, along with Frank “Famed Republican Pollster” Luntz. They discussed President Bush’s Social Security privatization plan, and they had a totally surreal discussion about the phrase “private accounts” (linked from Marshall’s blog). Luntz’s point is that the use of “private accounts” indicates a pro-Democrat bias, which may be appropriate in a debate but is not appropriate for journalists to use. Marshall asks, Well wait a second: Bush himself was using the phrase until just a few weeks ago. Does this mean that the media can only use a phrase as long as Bush himself is using it? That seems to be what Luntz is saying. I feel like I’m in the Twilight Zone.

I’d like to be able to set up some actual mathematical models of social structures like the Internet, so that I could answer questions like, “How long does it take before 90% of the world’s blogs and their readers have read a particular high-profile story from the New York Times?”

The way to do this, I think, involves computing some probabilities on graphs. A “graph” in this sense isn’t the sort of thing used to plot data; in the sense I care about, it’s a map of who’s connected to whom or which place is connected to which place. Graphs are used to represent, for instance, the distance between a number of cities: there are “nodes” representing, say, New York City and Boston, and an “edge” between them that might list the distance between the two cities (here the distance is often called an “edge weight”). You can use these structures to answer interesting questions, such as what the shortest path is between New York and San Francisco if you know a large number of road segments in between. (MapQuest probably uses Dijkstra’s Algorithm, if anyone’s interested.)

Anyway, so that’s the background. Now imagine that blogs (the nodes) are connected to one another by edges; specifically, blog A is connected to blog B if the owner of A reads B. (Note that the edge needn’t go in both directions: just because A reads B doesn’t mean that B reads A.) The more popular the blog, the more edges it will have joining it to other nodes. Now, if blog A posts on a topic, we might assume (for simplicity’s sake) that another blogger who reads A will link to that blog with some fixed probability. The more blogs that read A, the more likely the story is to move beyond A. Likewise, if A has more readers, the story is likely to spread more quickly.

All of that is straightforward and obvious. What I want to do is model it, so that I can get accurate estimates of the speed of “meme propagation,” for lack of a pithier phrase. This seems like it could be fairly tricky, not least because we’d want some way to stop cycles from forming (blog A mentions something, blog B picks it up, and blog A reads it in B, say; we don’t want the model to accelerate the pace of meme propagation just because A’s read it a second time).

A quick google for “probability on graphs” brings up a handy link to a free textbook by at least one very well-known prob/stats professor, and to a paper entitled “Random Walks and Electric Networks” (the latter of which has actually been GPLed!).

The whole point of my starting this post was to ask for references to probability-on-graphs books. If anyone out there can recommend anything, let me know.

The MGM v. Grokster case will be coming to a Supreme Court near you in a couple months, and a flurry of briefs has been filed in the past couple days. Over at Copyfight, Donna Wentworth posts a nice meta-blog entry (making this one right here, I suppose, a meta-meta-blog entry) linking to a lot of the good commentary on the case. I’ve read Felten’s post; it’s quite good. I expect Lohmann’s will be very good as well; he’s one of the legal techies like Eben Moglen whom you can count on to write well.

I just called Senator Kennedy’s office to suggest that he vote against Gonzales’s nomination to be attorney general. The guy answering the phone thanked me for my input and told me that Senator Kennedy does indeed plan to vote against the nomination.

If any of my readers’ senators are on the Judiciary Committee, you should give your senator a quick call to express your opinion. Whether the United States tortures human beings might well depend on your call and those of many like you.