If you use debian sid, and you upgraded to xserver-xorg 7.0.14, and you then found that emacs in graphical mode was useless due to broken fonts, the solution is this:

1. edit /etc/X11/xorg.conf (remember that you can use emacs within a terminal with emacs -nw)
2. Replace /usr/lib/X11/fonts with /usr/share/fonts/X11/
3. As root, run dpkg-reconfigure x-ttcidfont-conf
4. Restart X

While discussing the legislative battle for Internet Neutrality (in short, the phone and cable companies want to undermine the internet's potential to improve our lives, since it might come at the expense of their current business models), a TPM commenter made this poignant comment:

The more you hear about the breathtaking stupidity of politicians about the Internet, the more you have to respect Al Gore's role in pushing funding for it, and the more egregious it is that this simple fact was turned into a national joke by clever propagandists and stupid inept utterly clueless alleged journalists. —Taylor, Talking Points Memo Cafe

As you know, part of why I adore baseball is its intersection of statistics, random variation, and human interest give us a mirror to see how humans think. Greg Maddux's good start to the 2006 season is the latest textbook example. Maddux is the second-best pitcher of the last fifty years, behind only Roger Clemens, who himself is behind only 1910s pitcher Walter Johnson for the title of best pitcher ever. However, Maddux is forty, and has been in decline for several years. This season, without any appreciable change in his methodology, his results have been outstanding. So far this season he's the second best pitcher[1], up from roughly average last year.

Naturally, any notable fact requires an explanation. While us Maddux fans—and pretty much every nerd is a Maddux fan, since Maddux is a himself a slightly built thrower without great velocity, who has dominated with pinpoint accuracy and the ability to deceive and outthink most hitters rather than through raw power—would love to hear that he's back to full effect, the nerds have looked at the data. And the data says that Maddux is merely repeating last year's league-average performance, in terms of preventing hitters from hitting the ball, but fewer of the balls put in play have resulted in hits. That's generally not something under the pitcher's control. In other words, Maddux has simply been lucky.[2]

And Maddux himself agrees with this analysis:

... somehow -- and Maddux will swear he doesn't know how -- he's turned the losses of last season into wins.

"If he tells you that [he doesn't know], you gotta believe it. I believe whatever he tells me," Cubs manager Dusty Baker said last weekend in St. Louis. ... And from Maddux himself: "I wish I could explain it. Personally, I'm just getting ready to pitch like I always have. I'm just catching a lot of breaks. Sometimes you catch a lot of breaks in this game, and right now I feel like am."

But that's not satisfactory for baseball writers:

The real truth, or at least a major part of it, is simple: Maddux is in better shape than he's been in years. Maybe ever. After failing to win 15 games in 2005 for the first time since '87, Maddux approached Keith Kleven, a physical therapist in Las Vegas, and asked for help.

Kleven started Maddux on a new workout last November, a couple of months before he normally begins his offseason regimen. The results are speaking for themselves. (Jim Donovan, SI.com)

Random variation is often the best explanation for surprises, but humans seem to have a real problem accepting that. And explanations that include hard work and merit and cause and effect are always comforting. But "shit happens" is far closer to universal truth then you might want to believe.

Notes:

1. Maddux is second in both VORP and RA. VORP means the number of runs he has prevented compared to an average minimum-wage pitcher throwing the same number of innings. RA is simply the number of runs allowed per nine innings pitched. VORP is a cumulative statistic and RA is a rate statistic. The nerds prefer RA to the more familiar "ERA" for esoteric reasons.
2. "He’s benefitting from a very low batting average on balls in play (.192, fifth in MLB among pitchers with at least 15 IP). His peripheral stats don’t support a league-leading anything, and his groundball-to-flyball ratio is the lowest of his career, which is contributing to that BABIP. That he’s given up just one home run and three doubles on 18 fly balls is an accident, and can’t be expected to continue." —In Praise of Maddux, Joe Sheehan, 24 April 2006, Baseballprospectus.com

...

I volunteered to go to a local high school college fair on behalf of my alma mater, Pomona College. Last night I drove a flexcar to Fallbrook, about 110 miles round trip, through exurban rush-hour traffic. I was half an hour late (because Flexcar erased my reservation) and, in the opinion of the professional admissions counselers there, didn't miss much. The fair was a bust, probably, they said, because there was a much bigger fair the next day at the San Diego Convention Center. And, the counselors said, Pomona probably hadn't asked me to go to that one because it was so big that they would send professional staff, not a volunteer alumnus.

So, this morning I hopped on my bicycle and rode downhill to the Convention center (which, if a couple of condo buildings fell down, would be visible from my apartment) to hand in the one completed student interest card and give back the dozens of excess brochures. But it turned out that Pomona wasn't there.

I called the admissions office to find out why I had been asked to go 110 miles through rush-hour traffic to a sparsely attended high school gym college fair instead of one mile to a very large college fair. Maybe, I thought, The answer was, "huh, I'm surprised we didn't know about it."

I feel vindicated about never contributing to the alumni fund.

Postscript: I did get a very polite followup from the admissions office, where they explained that the big "NANAC" fairs are less cost-effective for Pomona to attend, and it was a deliberate choice not to participate rather than an oversight.

From my balcony, I can see thirty-nine cars parked on the streets. There are another thirty-six cars visible in the three parking lots directly under my balcony. An uncounted number of cars drove past on the streets while I was counting.

No wonder I hear so many car alarms all day and all night.

A great article on ESPN.com about Barry Bonds, baseball, and what it means historically and culturally:

... the past five years have been an especially depressing stretch to be an American ... it's the Era of Predictable Disillusionment: a half-decade in which many long-standing fears about how America works (and what America has come to represent) were gradually—and then suddenly—hammered into the collective consciousness of just about everyone, including all the people who hadn't been paying attention to begin with.

This will not be lost on future historians. In 50 or 100 years, they will search for events within the popular culture that supposedly embodied the zeitgeist of the time. Some of these people will use sports, not unlike the way contemporary historians might use Muhammad Ali as a means to define the 1960s. As these future historians try to explain what was wrong with the world in the early 21st century, I suspect they will use Barry Bonds.

[...]

A mound of evidence suggests that Bonds has been less than honest about steroids. But it seems like he's been honest about a lot of other things. "The last time I played baseball was in college," he said in his grand jury testimony during the BALCO case. He said almost the exact same thing to The New York Times Magazine in 2002: "The last game I played was in college. Ever since then, it's been a business. This is a business."

[...]

Early in "Game of Shadows," authors Mark Fainaru-Wada and Lance Williams try to illustrate what motivated Bonds to inject chemicals into his rump, and they suggest that his actions were driven by jealousy and, to a lesser extent, race. "They're just letting him do it because he's a white boy," Bonds supposedly said of McGwire's steroid-fueled run at the single-season home run mark. This statement makes Bonds seem as paranoid as Richard Nixon. How, one wonders, could unseen puppet masters be pulling the strings behind the home run race? It all seems crazy.

But, then again, nobody ever wrote a takedown book on Mark McGwire. I'm not sure anyone even considered it.

Nixon wasn't always wrong.

[...]

For all practical—and statistical—purposes, Ruth wasn't a real person. In 1927 he hit 60 home runs, exactly twice as many as NL co-champs Hack Wilson and Cy Williams hit; when Ruth retired in 1935, he had hit 714 homers, more than twice as many as Lou Gehrig, the man in second place. In 1925, he got a tummy ache, and it was one of the biggest stories of the year. He revolutionized the game, captivated radio audiences, built houses, inspired candy bars and was used as an epithet by Japanese soldiers during the war. Ruth has been dead for 57 years, and he is still substantially more famous than Barry Bonds.

But Bonds is going to pass him, and no one knows how to feel. Ruth was a troubling person, but he's a wonderful idea; Bonds is a troubling person who's an empty idea.

I'm pleased to announce that, after some study, a fair amount of paperwork, more than one outlay of cash, and a four-hour-long multiple-choice test, I am now a certified Project Management Professional. I was motivated in part by my belief in being professional about my job, and that entails continuing education, communication with other people in the same profession, teaching, and all that other good stuff. The Project Management Institute is the 900 pound gorilla of professional organizations in this line of work, incorporated in 1969 "to promote a unifying influence in the advancement of the field of Project Management ...." I'll have more to say about that, and about the certification process, in a later post.

The other part of a professional certification is of course its utility in hustling for jobs. So, if you know anyone—especially a non-profit— in need of the services of a certified project manager, be it a Technology Needs Assessment, Requirements Development, Usability Testing, or general-purpose Managing of Projects, drop me a line.

I saw this a few years ago and downloaded some parts; now the whole thing is available in a stream: "a recording of Ludwig van Beethoven's ninth symphony stretched to 24 hours, without pitch distortion."

The Box, Marc Levinson.

In chapter 1, Marc Levinson describes a shipping container and containerport, sketches out changes in port sizes resulting from containerization, outlines a few changes to the world economy that resulted, mentions the impacts on labor, and complains that the historical data doesn't exist to back up many of his claims. You could stop reading here and you wouldn't miss much. I wish I had.

The rest of the book is simultaneously very detailed and not especially informative. Although he throws in numbers in piecemeal fashion, the entire book is a collection of anecdotes, some very long, without much connective tissue or broad scope to convince the reader that the anecdotes should be taken as representational. More fatally, the prose is consistently mediocre. I didn't dislike it enough to abandon it, but I never enjoyed reading it. For a mass market book like this, that should be a fatal flaw.

After all, the author had as his subjects industrial technology, the biggest ships in the world, multi-billion dollar companies, the Vietnam war, organized labor, violent strikes and strike-breakers, organized crime, and a foundation of globalization itself, and still wrote a tedious book. Although the minutia of committee meetings to standardize container fittings might seem to be a dull subject: 1) it really isn't, when you think of the paradoxes and complexities that cause one-inch changes in pieces of metal to affect billions of dollars of commerce and lead to such absurdities as giant fleets of rubber duckies circling the world because a container tipped over in heavy seas, and 2) even if it were inherently limited, recently people have managed to write exciting books about salt, dust, longitude, and the number zero.

But when I read the dust-jacket and learned that Levinson had been an editor at the Economist, it made sense that he could pick a great topic, mention many interesting aspects, and still fail to produce a good read.

An apparently equivalent book, Box Boats, is due out soon; I think I'll try and skim that one in a bookstore before I buy it.

In my current project, I work with a lot of users who are very good at what they do (essentially social work on a hotline), but not especially technically inclined. To the extent that they can be trained to file high-quality bug reports that developers can act on directly, the project benefits. I often step in to rewrite the subject (for example, from "another big bug!" to "search in zip code 91101 returns fewer than expected results"), adjust the priority, or otherwise touch things up. But the users are doing a great job, and probably under five percent of bugs require conversations to clarify.

The key principle in a good bug report is that you have to provide all of the information necessary to reproduce the bug. Corollary to that, you have to have a good sense of what is germane and what is extraneous. Usually it doesn't matter that you encountered the bug on a Tuesday, but every now and then that might be the key to fixing the bug.

The third point I want to make is that one very exciting aspect of open-source projects is that the bug database is in full public view. Every piece of software has bugs, but in an open-source system you can often get the bugs that matter to you fixed sooner if you file good bug reports, make it easy to diagnose and fix the bug, and generally are nice and helpful to the developers. I've had paid Intel developers put in hours fixing a bug in the linux driver for my wireless card; I'm not a big corporate customer, but I was willing to jump through a few hoops for them as they diagnosed the problem. Sure, Intel gets my services as a tester for free, but I get personal bugfixing service that would cost hundreds of dollars otherwise.

With all that as context, check out this bug report from Firefox, an excellent example of how to use "Steps to Reproduce":

1. Create 2 unique user accounts (for steps sake, let's call 
   the two accounts Joe and Mary) in Windows XP Home.
2. Logout and sign-in under Joe.
3. Open Firefox and go to an e-mail site or to jdate.com or wherever.
4. Attempt to log-in to the site so that Firefox will ask 
   whether or not you want your password saved.
5. Choose not to save the password.
6. After successfully logging in and having selected the
   "never save password" option, logout.
7. Log-in as Mary and open Firefox.
8. Browse, browse, browse... but you don't really have to. 
   Just go to "View Saved Passwords," click on the tab that 
   will show you sites to never save passwords for, and you'll 
   see whatever painful site Joe denied to save a password for.
9. Break-up with fiancé.

This is the kind of news I like to wake up to:

The Dodgers win big and Cardinals fans get their hearts broken.

Remember, it's not enough to win. The other guy has to lose, too.

Sony: The Private Life, John Nathan
Something more than a reporter's notebook dump, less than a comprehensive history of Sony, this book traces the history of personal relationships between the paramount leaders at Sony. It seems fair to summarize the book as, "Sony's key leaders, particularly Ibuka and Morita at the founding, later Ogha, and in the US Schulhoff, have had close personal relationships that have often influenced their business relationships." In that light, the book is interesting but narrow, with competent prose but a slightly repetitive structure.

The author boasts of fairly comprehensive access, appears to speak fluent Japanese, and often follows interviewee's statements with his own assessment of their honesty. The key things I learned, if Nathan's is taken as completely correct in his reporting and in his speculation, include:

• Sony's paramount leaders made many business decisions in private conversation, and then committees, the board of directors, and even nominally superior officers, all served as rubber stamps. That's not exactly surprising news for any company, but the details of how exactly that happens were interesting and convincing.
• Sony knew that the purchase of Columbia at the asking price, and poaching of Peter Guber, were bad business decisions, and decided to abandon the deal, but then resumed pursuit when Akio Morita expressed his personal desire to own a studio; this motivation is the best explanation for why Sony took steps that led to $3+ billion writeoff only a few years later.
• Peter Guber took Sony for a multi-hundred-million dollar ride. I already knew this, and I was aware of a book about it that I had no plans to read. So I didn't really need hundreds of pages rehashing it, bulked out by self-serving interviews from all involved.
• Two executives are primarily behind Sony's futile quest for synergy, former Sony America president Michael Schulhoff and 1990s Sony chairman Nobuyuki Idei. If there's one single thing (in my opinion) that has ruined Sony, it's trying to unify the content business and the electronics business. The hardware people want to make great gadgets that people love. The content people, at Sony and elsewhere, have a long history of dragooning governments into supporting outmoded business models through laws and intimidation. If a line of business has opposed player pianos, radio, cable tv, vcrs, and MP3s, why would you want to merge it with a consumer electronics company?

  Sony invented the walkman; they should have Apple's market share in the digital player market and Sony's own music business is the reason they don't. I seem to remember a story about tails and dogs and wagging. What does Sony's latest CEO say? "While we press for stronger legal protection, we are actively developing a new and more sophisticated generation of copy protection. DRM—digital rights management—will allow us to safeguard content on devices in a manner that is easy for consumers to use and understand." (Howard Stringer, 2004). Sorry, Sir Howard, but it's not going well so far.

• Akio Morita was a lousy father and husband.
• Sony's founders were unusually entreprenurial for Japanese businessmen, but still relied from day one on extensive personal and family networks for money and access.

I enjoyed the fly-on-the-wall boardroom accounts, unreliable but convincing, and I was touched by the relationship between Akio Morita and co-founder and mentor Masaru Ibuku. The scene in which, having built a business empire together and now both felled by strokes, they sit together in a hospital room holding hands, is universal. (Well, universal for the set of all emperors and kings of kings who survive to infirm old age, but I think the rest of us can see something there.)

Overall, the book was good but limited, and I would recommend it only to those already interested in the subjects.

There's been a mini-revolution in baseball the last decade, which can be oversimplified to Traditionalists vs Statheads. Essentially the Statheads showed up and said, a lot of what you think is true about baseball is demonstrably wrong. The Traditionalists said, go to hell. The poster child of the Stathead movement is the Oakland A's general manager, Billy Beane. Despite a small payroll, Oakland has won its division or placed second for many years in a row, and has easily the best win per dollar ratio of any time since Beane started.

Two of Beane's staff have gone to other teams. One, Theo Epstein, led the Red Sox to their first World Series victory in 86 years. The other, Paul DePodesta, was hired as general manager of the Dodgers. (In baseball, general managers are responsible for signing and trading players, and managers are responsible for day-to-day oversight of players and tactics during games.) The Dodgers did well in 2004, his first year, and did poorly in 2005, when they lost $36 million in payroll (about a third) to injuries. DePodesta had never gotten along especially well with the Dodgers' well-respected Traditionalist manager, Jim Tracy. DePodesta effectively fired Tracy after the 2005 season. The owner of the Dodgers, Frank McCourt, a highly leveraged East Coast real estate magnate, fired DePodesta two weeks later. Given that both DePodesta and Tracy are both very good, and between the two of them exemplify the two different philosophies currently in play, letting both of them go is evidence that McCourt is incompetent.

Jim Tracy landed a job managing the Pirates, who have been lousy for over a decade. Today the Pirates played the Dodgers, and lost. Tracy's commentary is priceless:

"One-and-7 obviously isn't the way you want to start," Tracy said. "But the (Dodgers) team I managed last year started 12-2 and we won 71 games. So it's not a large enough sample size of games. It doesn't mean this is the kind of team we're going to be for the next six months."

Jim Tracy using the phrase "sample size". Awesome.

When fifty thousand people gather a block from your home on a sunny, comfortably cool Sunday afternoon to march for immigrant rights and human dignity, what else can you possibly do but join them? Si, se puede! ("Yes, we can!")

Bonus points: "Attendance for yesterday's march topped that of the largest known San Diego march, a 1994 'March for Jesus' that drew 25,000 to 40,000 people to the waterfront." San Diego Union-Tribune

The Wisdom of Crowds, James Surowiecki

Executive summary: Collective decisions are usually as good as or better than individual decisions, even expert decisions, provided that the component members of the group are diverse and independent. Wikipedia offers a detailed summary, which I recommend everyone read. Beyond that, the book offers anecdotes, details, and verbiage. It's moderately well written but quite skippable.

King Kaufman says of Barry Bonds' reality TV show, "It's impressive in a way that Bonds would approve of a show that leaves him, in the end, looking like a giant prick ... It would be more impressive if I could believe that Bonds realizes the show makes him look like a giant prick. I'm not sure he does." New York Times reviewer Charles McGrath similarly exposes himself as a giant prick. Although he acknowledges that one of Bonds' legitimate woes is "his daily barrage of ugly, racist hate mail," he still ends his review of Barry's show with, "... you realize that Bonds's greatest tragedy is that he was born too late. He would have been a hero back in the day when all players had to do was play, not give interviews."

Um. I guess McGrath forgot that black players weren't allowed in Major League baseball "back in the day."

I saw Barry last night here in San Diego. Hit by pitch, reached on an error, and walked. He also turned a single into a double, but it was a single by Adrian Gonzalez of the Padres, which Barry's steady jog was unable to cut off before it rolled to the wall. Bonds was pulled after the sixth for defensive replacement Steve Finley. (At 41, Finley is almost a year younger than Bonds, but has two healthy knees.) Personally, I'm cheering for Barry. He very probably used, but so have a lot of people; if Babe Ruth and his contemporaries wasn't taking steroids, it was only because they weren't available yet. Meanwhile they found plenty of other ways to cheat; Kevin Gross didn't invent sandpaper and Pete Rose didn't invent gambling. Until and unless Barry Bonds' records are proven to violate the rules in effect when he played, they shoud stand.

I also got to see my favorite player, Mike Piazza, in person for the first time in many years. He grounded out twice, lined out to center, and flied out. He's looking good, though, aside from the new goatee.

Today's good news....

You may be aware of the ongoing national campaign by phone and cable companies to get protection from competition. New Orleans has been running a free wireless internet network for locals and local businesses, reportedly set up for only a million dollars, and providing 512kbps service. The business interests had already gotten a "Fair Competition Act" passed, and so that free network would be illegal but for an exception granted for emergencies. The cable and phone companies are preparing to argue that the emergency is over and the network should be reduced in speed (so that they aren't facing unfair competition - remember that to telecom companies, all competition is unfair). The good news is that New Orleans technology chief Greg Meffert is ready to fight. "In the end, it takes a federal judge to issue a restraining order. Until that point, if that point ever comes, we'll keep running it. It's a lifeline to these people."