Anycast tutorial
I found the anycast tutorial by Bill Woodcock very interesting and helpful. It also makes me quite uneasy in my confidence in my understanding of Internet topology.
Support Bloggers' Rights!
I found the anycast tutorial by Bill Woodcock very interesting and helpful. It also makes me quite uneasy in my confidence in my understanding of Internet topology.
This is hilarious. A report says that Sony Music's bundled rootkit inadvertently defeats Blizzard's bundled spyware. (More precisely, the rootkit gives people a tool to use to hide things from the spyware. See "Lowcost superb RING0 rootkit developed(payed) by Sony BMG".)
It's the latest incarnation of real-life Core War on the PC.
P.S. Lessig says that "code, law, norms, and architecture" regulate; if so, getting mad (and getting other people mad) about rootkits and spyware may serve a concrete purpose. As J.B. Nicholson-Owens suggests in a comment on Ed Felten's blog, we can imagine a world in which people don't put up with software that undermines its user and has to fight with other software.
P.P.S. There are more points of view about this than you might think, because there is a seemingly large and growing community of security people in the business world who neither want users to be able to modify software in ways that its publishers disapprove of nor want software to be able to conceal itself, spy on other software, or attack other software. Those people support the idea of Blizzard preventing people from modifying Blizzard games (or even from writing software like bnetd) but oppose the idea of Blizzard using (or having to use, or being allowed to use, or being able to use) spyware to accomplish this objective.
John Gilmore recently persuaded me to get a San Francisco Muni monthly pass. I first suggested that it might not be a good deal for me since I don't use transit to commute to work every day, but it turns out that it's often a good deal if you use transit about an average of once per day. I told John that I wasn't sure whether I use transit even that much, and he pointed out that having a flat rate makes people much more willing to use transit -- if they see a bus, they can just get on without needing to pay a fare, even if they only want to travel a short distance. So you can be a more impulsive transit user, because you never experience a marginal cost from choosing to use transit.
Sure enough, I've already used my pass for about $10 of transit service in just two days, and it's already making me see Muni as a more practical and ubiquitous option. Having the pass probably will increase my use of transit to an average of more than once a day. And it works on BART and even cable cars.
When I lived in Berkeley, I had a BART Plus pass. It's interesting to contrast the Muni fast pass with the BART Plus pass.
Muni fast pass:
BART Plus:
The bus pass feature of both passes expires at the end of the period of validity of the pass, but the BART Plus ticket can still be used for for-fee travel on BART after that period.
BART Plus became a drastically less good deal for people who live in (or often travel to) the East Bay when, a few years ago, it stopped being valid on AC Transit. AC Transit support is still a glaring omission for BART Plus.
At present, I see the Muni pass as a much better deal. I think my experience will show that it can be useful even for people who aren't regular transit commuters. I doubt that the same can be said for BART Plus. And people who really ride BART across the Bay frequently might well be better off with a high-value ticket, which provides a slight discount without also serving as a bus pass.
Congratulations, Anirvan and Charlie!
Now the Bay Area Cena Latina, one of my favorite things to do, has a home page.
Venite omnes!
At the Rusticatio Californiana, we had a fun tradition of having an "adverb of the day". One of the best-loved adverbs of the day was promiscue, and it was that adverb that came to mind when I thought about how to describe the vast and secret collection and retention of data through National Security Letters described by the Washington Post in a front-page article this week.
Best line: "What happened in Vegas stayed in Federal data banks."
This is said to be a common error, but it turns out to be so common that it even affects Latinists:
http://la.wikipedia.org/wiki/Interrete:
Interrete (-is, n) (Anglice: internet) totum mundum computatris connectat. Multi sunt servi, ubi paginae domesticae (in forma .html) et data varia locatae sunt. Computatra clientes vocata data a servis accipent, saepe ad paginas domesticas inspiciendam. Ductus electronici multa data transportant.
Interrete inventus est Genavae, in aedibus institutionis Europaenae CERN vocatae.
That is:
The Interrete (genitive Interretis; neuter) (in English: Internet) connects the whole world through computers. There are many servers where home pages (in HTML format) and various data are located. Computers called clients receive data from the servers, often in order to view home pages. The electronic paths carry much data.
The Internet was invented in Geneva at the facilities of the European institution called CERN.
This kind of error could seriously undermine the credibility of the Latin Wikipedia.
This weekend, I went to two very different cultural events in San Francisco.
On Saturday, courtesy of the Craigs, I went to the San Francisco Symphony to hear a performance of Carl Orff's Carmina Burana, which was one of my favorite things to listen to in high school. I've only heard it performed live once before, at UC Berkeley in 2000.
The San Francisco Symphony's version was fantastic, and it was especially amusing due to the informality and joking around of the conductor and soloists. During the lead-in performance of La Noche de Los Mayas, several audience members mistakenly applauded after the second movement. (At classical concerts the social convention is that you applaud only at the end of a composition, not at the end of a section of a composition. This means much less frequent applause than at other kinds of concerts. Often, a few people in the audience don't know this or are confused about when a composition has ended; usually the social pressure of the other audience members' silence gets them to stop right away.)
The San Francisco Symphony tells first-time concert-goers that
[i]t is considered proper concert etiquette to clap only after a piece is complete. This means, for example, if you're listening to Beethoven's Symphony No. 9, which has four movements, it is appropriate to clap after the last movement. You can look at your program book to find out how many movements a piece has. Usually, there is a 15- to 30-second pause in between movements. So, in the case of Beethoven's Ninth Symphony, you know you're hearing the Finale after three pauses. If all else fails, you can always wait for the rest of the audience to clap before applauding.
At the concert on Saturday, some people started clapping in between movements, and the conductor turned right around and exclaimed -- to the majority of the audience which had remained silent -- something like
Listen, we all know there's not a musician in the world who doesn't like applause and don't let anybody tell you otherwise.
This was hilarious and was greeted with much laughter and applause.
Later, the soloist singing "Olim lacus colueram" actually acted out (with great enthusiasm) the whole story of being a swan about to be eaten, and all the soloists really got involved in their parts and played them up humorously. And I think that's interesting because I've heard discussion of the phenomenon that, since Carmina Burana is in Latin, some people will approach it very solemnly, like an ancient religious text. This is funny because Carl Orff's own title for his piece is Cantiones profanae cantoribus et choris cantandae comitantibus instrumentis atque imaginibus magicis, which starts out by telling us that they are "secular songs" or even "profane songs". And certainly their content is that -- so these particular soloists really seemed to appreciate that point thoroughly.
One interesting thing was the number of different ways of pronouncing Latin. The several choruses all pronounced intervocalic c as ts, whereas I think two of the soloists pronounced it as ch and the other as ts. (I pronounce it as k.) One of the soloists, on the other hand, pronounced at least one intervocalic t as ts. (I should compare this with modern spoken German; I have a theory that that's where it came from, because some classical singers may be extremely familiar with German phonetic rules.) It's funny to hear people pronounce Latin differently from the way I do, but it's even funnier to hear different musicians in the same performance pronounce it differently. I guess it's hard for the conductor to get them all to override their years of training and practice...
On Sunday, at Zooko's recommendation, I went to the ComBots robot battle tournament at Fort Mason Center. Here I saw various robots smashing, tossing, and grinding each other up.
The most exciting thing was when a "spinning" robot, which was either spinning as a unit or else carried a rapidly spinning cutting blade, clashed with another robot of whatever kind. This invariably produced sparks and sometimes produced flying shrapnel. The robots fought each other inside an enclosure of bulletproof glass. Not only was this repeatedly proven necessary as robots and robot parts slammed into it at high speed, but we ever sometimes wondered whether some of the spinning robots would break through it if they were thrown up against it at the right angle. After all, the walls of the arena inside the bulletproof glass enclosure were made of steel I-beams (the kind used in construction), and the spinning robot Megabite had repeatedly torn enormous gashes straight through the I-beams on contact. If a fighting robot can cut through a steel I-beam, perhaps it can cut through bulletproof glass, too.
The robot builders followed various strategies: essentially, these were the spinner, the wedge, and the hammer. (A spinner either spins its body or a separate cutting blade at extremely high speed; if it has a separate blade, it could be mounted either horizontally or vertically, which makes a big difference in combat. A wedge tries to flip its opponents over, or pick them up and dash them into the walls, or sometimes pick them up and flip them out of the arena, which may involve a moving arm. A hammer tries to smash the other robot from above with a heavy moving part.) There is a great deal of variation within these categories, including with regard to weight distribution, whether there is a robot arm present, whether the wheels are shielded, what kind of wheels and motors are used, etc. The spinners definitely produce the most drama, with sparks constantly flying everywhere, but some people described a rock-paper-scissors situation where each strategy had strengths and corresponding vulnerabilities.
I had a great time at both Carmina Burana and ComBots and am happy to live in a city that offers them on a regular basis.
Upon reading in Wikipedia this evening that George Boolos practiced constrained writing by delivering a speech on Gödel's Incompleteness Theorem in words of a single syllable, I pulled out Boolos's Logic, Logic, and Logic, where I found the speech. It's pretty good, although it only explains Gödel's Theorem, without proving it. Still.
There's some fabulously fun stuff at the end of Logic, Logic, and Logic, including a new typography for a limerick about W. V. O. Quine, a refutation of Penrose on humans' noncomputational ability to "see" that Gödel sentences are correct, and the thing I want to mention here, which Boolos calls "the hardest logical puzzle ever".
When I read it, I realized that someone some years ago had presented me with a slightly easier version of the puzzle, that I had worked on it for about three days, and that I had come up with a proof that the puzzle was impossible to solve. (My proof turned out to be wrong!) The original version that was presented to me was Raymond Smullyan's version; Boolos explains in a footnote that John McCarthy (yes, that John McCarthy) proposed the change that makes it even harder (but still solvable).
This puzzle is an extreme twist on the familiar liar/truthteller puzzles, which have already been elaborated in all directions by Raymond Smullyan. In Boolos's view, this puzzle is at the apex of Smullyan's work in the genre.
Here is how Boolos phrases the puzzle (omitting his footnote on McCarthy's contribution):
Three gods A, B, and C are called, in some order, True, False, and Random. True always speaks truly, False always speaks falsely, but whether Random speaks truly or falsely is a completely random matter. Your task is to determine the identities of A, B, and C by asking three yes-no questions; each question must be put to exactly one god. The gods understand English, but will answer all questions in their own language, in which the words for "yes" and "no" are "da" and "ja," in some order. You do not know which word means which.
My impossibility argument came back to me as I was thinking about this puzzle. Here it is: Each question you ask will yield at most one bit of information. The number of bits necessary to express who is who will be ld(3 P 2) (the dual logarithm of the number of permutations of three elements taken two at a time), which is about 2.58 bits. However, when you ask the first question, since you don't know who is random, it is possible that the answer you get will be random, and hence independent of the assignment of names to gods, and hence not yielding any information whatsoever. (That is to say, the answer may not decrease your uncertainty about which god is which in any respect, because the answer could be "ja" completely independently of which god was which, and the answer could be "da" completely independently of which god was which.) Even if you had a way to get one bit of information from each of the two remaining yes/no questions, the total amount of information you could get would be 2.0 bits, which is not enough.
My argument is wrong and the puzzle can actually be solved in a deterministic way. When I first heard the puzzle, I was attached to my impossibility argument and could not be talked out of it until someone told me the solution, whereupon I agreed that it was right.
This is a great puzzle, and I wish I could remember who first asked me about it.
Sony should, and I hope will, pay dearly for its transgressions in this matter, but I hope we don't lose track of the real point. The real point is that DRM is intolerable. This was an especially intolerable form of DRM, but any DRM is intolerable; we, as users, should not tolerate DRM, period. A kinder, gentler DRM with the most serious holes of this one fixed would NOT be okay. There can be no acceptance of DRM at all.
The first rule of Security Theatre is never to talk about Security Theatre.
(Ruidh on Schneier's blog)
I went to Cirque du Soleil's show Corteo with Michelle and Nicol in San Francisco last night.
I was totally amazed and would be eager to see their other shows. This was the first time I've ever seen Cirque du Soleil in person, and the first time in about twenty years I've seen any circus (if I exclude the aerial tissu artists in a theatrical performance I saw in Seattle).
For me, one of the most special moments was seeing someone actually floating on a bunch of helium balloons -- since one of my favorite books as a child was The Twenty-One Balloons, a children's adventure/utopia by William Pène Dubois. I liked it so well that I'm sure I re-read it at least three times. Among the many fabulous things in The Twenty-One Balloons was an image of a child hanging, for fun, from the bottom of a helium-filled weather balloon and bouncing high into the air -- almost flying. Thanks to Cirque du Soleil, I've finally seen it happen.
I also greatly admired the skill with which the Cirque acrobats jumped up and down on beds, and Uzeyer Novruzov, the ladder expert, who would climb (and walk around on) ladders that weren't balanced against anything, while being chased by friendly robots and a flying angel. I also enjoyed the symphonic performance by a whistling ringleader, a table full of water-filled wineglasses, and some whirling players of glowing bells. I would also mention the Cyr wheel as one of several acts that I wouldn't have imagined were physically possible for people to perform, and the cleverness of having the high-wire walker walk up a slanted wire. The whole show was full of wonderful things.
Corteo will be here in San Francisco until January 8.
Contact: Seth David Schoen