Steve Reads

I would like to point out that if you want to find me, I will be in front of the TV for a total of 9 — or probably more like 12 — hours tonight and this weekend, watching the Sox/Yankees series. And like the ALCS and World Series, I intend to mute the television, bring a radio in, and have WEEI’s sarcastic, bitter baseball commentators speak where Tim McCarver would have.

Still the dumbest thing that I’ve heard anyone say during a baseball game was McCarver’s line that walking Hideki Matsui was as good as giving the Yankees a home run. The mind will go to excruciating lengths to make a chaotic world coherent; McCarver is the verbal equivalent of a Rorschach test.

P.S. (1 Oct 2005): The “Listen to WEEI and watch Fox” approach doesn’t work well: Fox is apparently on a 10-second-or-so tape delay, such that I always know about exciting base-stealing attempts (or what have you) before they happen. Not cool. So I’m just muting the TV whenever McCarver starts talking.

Offhand, wouldn’t it seem that some sort of biological process would stand a good chance of being the world’s most efficient energy-storage process? There’s tremendous evolutionary pressure to eke the tiniest extra bit out of input food or sunlight, whereas there seems to be nowhere near that pressure to build an efficient battery or nuclear reactor. (Another instance of “evolution is smarter than you are.”)

Does anyone know where to look for energy-efficiency statistics, measured in terms of (say) input grams of matter to output joules of energy? That unit will have to be tweaked somewhat, but you get the idea.

I just drove my mom to work so that I could have her car for the day, and on the way I heard two stories on VPR that are worthy of note:

1. A story about the Sweet Caroline (bah-bah-bah) / Good times never seemed so good (So good! So good!) tradition at Red Sox games, which makes me cringe every time it happens — which is to say, during every single Sox game in the eighth inning. However, hearing the tradition mentioned on NPR made me really badly nostalgic. I’m going to miss Boston. I’ll come back. I’ll certainly come back to watch Sox games. (I include the Globe’s explanation of the tradition below the fold.)

2. A story about Isaac Barré, after whom cities in Massachusetts, Pennsylvania (namely Wilkes-Barre) and Vermont are named. He was apparently an important figure in the American Revolution, yet no one knows about him. This story wouldn’t be noteworthy, were it not for an interview with a local historian from Barre, Vermont who sounds like she would have been perfectly cast in Waiting for Guffman.

(more…)

Friend Adam pointed me to a just unbelievably fascinating interview with Brewster Kahle of the Internet Archive. He’s an amazingly interesting guy, and a great speaker. I don’t know if I’ve heard such an interesting interview ever.

I encoded the interview into a 22-meg Ogg. Email me if you’d like a link to it.

P.S.: Kahle recounts the birth of the free-software movement; the story as he tells it is electrifying, infuriating, and sad. The way he describes it, the 1976 copyright act suddenly made everything copyrightable without having to register it or send a copy to the Library of Congress; this I knew. But he goes on to say that if that act hadn’t been passed, MIT would not have been able to do what they did, which was package up all the work on the Lisp machine that Stallman, Kahle and others had worked on, and sell it to Symbolics and a few other companies. I didn’t know that part. Could those of you with a better understanding of the relevant copyright history explain why it would have been impossible for MIT to do this before the ‘76 act? This doesn’t seem as much about automatic copyrightability as it does about the treatment of works for hire. Would MIT never have asserted copyright in the works of its employees and grad students before ‘76?

P.P.S.: Check out the Petabox. ::drool:: The Archive adds 25 terabytes every month. Damn.

P.P.P.S.: How can you not love a guy who ends an interview in this way?

If we end up with cell phones all being, like, a little personal AOL, where it’s all controlled by your provider as to what you see  . . .  what a tragedy. If we end up building another television out of this? What a shame. What a waste. We would’ve spent twenty years having a possibility of having done something great and having lost it. So I’m spending my time, my fortune, thanks to Steve Case, Jeff Bezos [who bought out his earlier companies] on trying to make sure that we have a future that we actually want to live in — something we’re actually proud of at the end of the day, having said, “Yeah! We built something as good as books.” We took the Library of Alexandria idea of having all information available and not only made it available in Alexandria, Egypt — which they did for five hundred years; we all say, “It burned!”; yeah, but it was up and running for five hundred years. They had five hundred years of great materials available.  . . .  But then taking the Library of Alexandria and making it available to anybody all over the world. That’s a worthwhile goal to get up and spring out of bed in the morning and go to work.

Dear god, I want to eat some of this. And the idea of Gruyère taking its sweet time to turn into a mouthwatering bit of perfection is  . . .  well, it’s killing me, I’ll be honest with you.

Once I’m back to steadier employment and I’m in a place with a raw-milk vendor, I will be Mr. Raw-Milk Cheese Man. Looks like Virginia might be for raw-milk lovers.

Incidentally, I think that Jamie should name his operation Jews for Cheeses.

The Sox have decided to die at the end of the season, I think. This makes me sad.

I don’t know why I forgot about this: no one needs to pay for TimesSelect, because BugMeNot exists. Thanks to Andy Gerard for reminding me.

Barring any major last-minute surprises, I’ll be moving to Washington, D.C. to take a very cool job within the next couple weeks. It’s still possible that the salary offer from the D.C. gig will be too low, and that an IT consultant in Boston who’s trying to woo me will offer a higher salary than I expect him to offer. But I doubt both those things, and in any case the D.C. position offers a lot of good learning opportunities; the biggest thing is that I’ll be working with enterprise-scale networks, so I’ll start learning how the big boys do networking and server admin. It should be a really good experience for me. Plus I’ll be moving near some of my favorite people in the world.

I’m very, very excited.

If there’s any one place where no thinking person could possibly dispute the importance of open-source software, it is in computerized elections. So it’s good to see that California’s starting to think about this stuff.

A few people have reported troubles with the new AdSense ads: apparently they cover the dropdown menus, even though I’ve set the z-index on the dropdowns so that they’ll always be above the ads. I’ve reproduced the behavior under Opera for Linux and Konqueror for Linux, but it doesn’t show up under Firefox, Deer Park, or IE. If anyone can see this behavior under other browsers, could you take the time to take a screenshot of this blog with the menus extended, then post a link to the screenshot in the comments here? Either that or email me.

Google’s AdSense tech support, by the way, has been very helpful on this. They suggest that it’s a browser-specific issue, depending on how each browser renders iframes.

I’m really astonished at the number of times that new Linux users complain about sudo, and about the absence of a root account in Ubuntu. So let me just go over this once and for all. Then the next time someone mentions this on the ubuntu-users or debian-user list, I can just point them here.

If you’re not familiar with Linux, sudo is a command that allows you to run a single command as the root user; it stands for “superuser do,” or at least it used to. Now I think it just stands for “sudo”. In turn, the root user is the first user created on any UNIX box, and is the all-seeing eye on any UNIX box. Root can change passwords, can look at everyone’s files, can run any program, and can delete any file on the disk. Hence the single command sudo rm -Rf / will delete every single file on the hard disk. (rm is the command to delete [it’s short for “remove”, I believe] a file, and / is the base of the filesystem. The -R argument means “delete files recursively,” and the -f argument means “Answer yes if the system asks me whether I’m sure I want to delete a file.”) Root is a powerful thing.

Now, if you wanted, you could sign in as root using the su command (which, again, used to stand for “superuser” but now just stands for itself) and happily execute lots of commands as root. However, this is dangerous and bad, and no one should do it. You might leave yourself logged in, and someone else could come along and wreak havoc. You might forget you’re logged in as root and execute a command that you think you’re executing as an ordinary user, thereby causing unintended mischief. You might edit a file as root and save it, thereby changing the owner to root and causing still more unintended consequences.

The point is that root is a very dangerous tool, and even the best of us can make mistakes with it. That’s why sudo exists: you run a single command as root, then exit. Sudo has other advantages as well:

• No one needs to know the root password. You use sudo with your own password, after the root user enters you in the /etc/sudoers file. If your company has you in the sudoers file, and they fire you, they can just take you out of the sudoers file; they don’t need to tell everyone to use a new root password.

• You can specify, in /etc/sudoers, what limitations sudoers have. That is, it allows finer-grained control than just “root v. not-root.”

Ubuntu actually gets rid of the root user: when you install it, you create a regular user, who immediately gets entered into /etc/sudoers. This leaves open a security hole, namely that if you want to sign into single-user mode, you can’t do so with the root password. Hence Ubuntu allows you to sign in with no password. This seems dangerous to me, but you’re only going to use it when you have local access to the machine — i.e., when you’re sitting in front of it. The common wisdom seems to be that anyone with physical access to the machine has basically arbitrary control anyway. This has never seemed an especially convincing argument to me: yes, given a long enough amount of time, someone with physical access to the machine could infiltrate any data on it. But suppose the machine were physically secured with a padlock; that the BIOS were password-protected (or, e.g., protected by the requirement that you plug in a USB key drive with your private key on it before you were allowed to log in); that as much of the disk as possible were encrypted; that you needed a thumbprint of some kind to access the operating system; that the hard drive would wipe itself after a certain number of failed password attempts; and so on and so forth. Yes, it’s still possible that someone with physical access to the box could access your data, but it’s unlikely. And if, for instance, the attacker only had a limited amount of time in front of the box — say, it’s the FBI trying to plant a keylogger on your disk while you’re at work — then these additional hurdles would be a real obstacle.

In any case, I hope this addresses the common objections to sudo. You want to use it. Really. I promise.

It’s nice to see sites that clearly were designed in 1998 and have not really changed much since then. It avoids some of the design disasters of early websites — see, e.g., the “center everything” philosophy of the EFF’s website c. 1996 — but it’s still stuck in the “now all I have to do is search for a really catchy background GIF” mentality.

We’ve come a long way since then. I like seeing, for instance, that Movable Type and Wordpress use default templates that are quite gorgeous and distinctive; you can always tell when you’re on a Wordpress blog. These are minimax solutions, in a sense: they minimize the maximum damage that anyone can do, with the realization that most people are not designers and shouldn’t have to take the time to be designers.

P.S.: I was actually just guessing on the “1998” bit, but the Internet Archive backs me up.

Congress Abandons WikiConstitution

WASHINGTON, DC—Congress scrapped the open-source, open-edit, online version of the Constitution Monday, only two months after it went live. “The idea seemed to dovetail perfectly with our tradition of democratic participation,” Senate Minority Leader Harry Reid said. “But when so-called ‘contributors’ began loading it down with profanity, pornography, ASCII art, and mandatory-assault-rifle-ownership amendments, we thought it might be best to cancel the project.” Congress intends to restore the Constitution to its pre-Wiki format as soon as an unadulterated copy of the document can be found.

A headline like “U.S. Says It Has Killed No. 2 Qaeda Operative in Iraq” brings up a question that I would have thought obvious, but which I don’t really see discussed much: under what conditions are we justified in using the military, rather than bringing someone back to be tried under the judicial system?

If the person we wanted to capture were within U.S. borders, I’m pretty sure it would be illegal to send the military after him, and it would certainly be illegal to kill him. My sense is that one legal doctrine — subject to common-law nuances, of course — is that when ordinary judicial procedure is working properly, the military is not allowed to overrule that procedure with military law. I’d have to assume that it’s even more cut and dried that the military cannot kill someone within U.S. borders. Presumably there are very precise laws about when the police are allowed to kill someone; I’d assume that even the threat of violence from that person at some unspecified future date is an unacceptable reason to kill him. Without knowing the case law, I’d have to assume that only the threat of imminent violence from that person could justify killing him.

Now, are there any statutes on the books constraining the military in a similar way, when it’s operating abroad? Is the military forbidden from killing someone unless a) it’s in a declared war, or b) there’s a threat of imminent violence? If there isn’t, why not? Shouldn’t we codify the oft-stated goal of only using the military as a last resort?

On the continuing “2.0” tip, we now have Identity 2.0. It’s a very interesting problem: you want to be able to prove your credentials to a number of different websites — Amazon, Craigslist, Flickr, etc. — so that you can, for instance, integrate a bunch of different applications on your website using XML-RPC. Dick Hardt gives a great example in that presentation: you want to be able to include your Amazon wishlist within your Friendster profile (I think he uses some other FOAF service as an example, but Friendster’s fine), so that you won’t have to retype that list every time it changes. As it stands, the only way to do this is to provide your Amazon username and password to Friendster. That’s silly, and it’s obviously not fine-grained enough. Hardt suggests that Web 2.0 (which is based around web services) will force Identity 2.0, and he sketches some of the traits that Identity 2.0 will have to have. It’s a great 15-minute presentation, well worth the time, and it has encouraged me to subscribe to Hardt’s Identity 2.0 blog.

(Via Larry Lessig)

I was banging my head against the wall today, trying to figure out why a particular rsync command line wasn’t working. Eventually the wall told me the answer.

Here was the problem: I was doing something along the lines of

rsync --include /long/path/name/ \ --exclude '' $localSource user@remote.host:$dest 

--include /long \ --exclude / \ --include /long/path \ --exclude /long/* \ --include /long/path/name --exclude /long/path/* 

--include /long/path/name \ --exclude '' 

shouldn’t work. It seems to me that that’s the more intuitive syntax — either that or

--exclude '**' \ --include /long/path/name 

— but in any case, I’ve figured out the problem. This note is as much for my own benefit as everyone else’s, because it seems to happen every few months that I get confused by rsync’s include/exclude syntax and have to re-bang my head.

On a trip to Boston a week ago, I brought along a whole bunch of CDs to keep me occupied during the drive. One was Zooropa, U2’s first album after Achtung Baby. I’ve always liked it, and at times I’ve thought it beautiful, but I’ve come to the realization that it’s probably the most underrated album in the last 20 years. Except for a couple dark spots (“Babyface” and “Some Days Are Better Than Others”), it’s an extended work of brilliance, every track of which tackles some new angle on the way the modern West works. Inside the liner notes is a photo of two people having sex while their heads are covered by televisions; that pretty well covers the theme of the album. (I think the image may be borrowed from Cronenberg; it certainly seems like something out of Videodrome.)

And I came to a realization about what “Stay (Faraway, So Close!),” Zooropa’s track 5, is about. First off, the song’s title is also the name of a film by Wim Wenders. I’ve not seen Stay, but I have seen the film — Wings of Desire — that precedes it and which apparently explores the same themes. In Wings, we see that the world around us is filled with angels whom we can’t see, who can hear all our thoughts; it’s a ghostly effect, with a continual low-intensity chatter filling up much of the movie’s space. The angels sit beside us and listen to us with calm, sometimes sorrowful expressions on their faces: they’ve been sitting in similar positions for all eternity, listening. They can only watch and listen; they cannot intercede. There’s one way in which they can affect human affairs, and that is to commit suicide in some way: fall from the sky, crash to the ground below, and wake up reborn as a human. (Wings was remade into a movie of stunning banality starring Meg Ryan and Nic Cage, but I repeat myself.)

Now, “Stay” — the U2 song — has made me completely rethink what the Wenders movie was about. The song talks about our being able to visit anywhere in the world with “satellite television” — places like London, Belfast, Berlin, Miami, New Orleans that we can see and hear, and whose sufferings we can empathize with, but whose sufferings we’re also powerless to fix. The final line in the song is “Just the bang and the clatter as an angel hits the ground,” which I’ve always taken to be the suicide of the song’s world-weary protagonist. But having thought about the connection to Wenders, my sense now is that it’s not a suicide at all — it’s a literal angel literally being reborn in a world that he can now change.

And I wonder whether the moral of the movie has to do with the act of watching: that our TV-saturated culture can look but not change, and that the only way to get out of that trap is to turn off the TV and rejoin the human world. I’ll have to go back and rewatch the films.

The New York Times did a pretty awful job a while back when they tried to give equal time to intelligent-design advocates and scientists. The Washington Post today weighs in with another piece on ID, this time being very clear from the beginning that there’s no legitimate doubt left about the prominent place that natural selection will have to hold in any theory of biology.

And yet the still give some time to the Discovery Institute:

Asked to provide examples of non-obvious, testable predictions made by the theory of Intelligent Design, John West, an associate director of the Discovery Institute, a Seattle-based ID think tank, offered one: In 1998, he said, an ID theorist, reckoning that an intelligent designer would not fill animals’ genomes with DNA that had no use, predicted that much of the “junk” DNA in animals’ genomes — long seen as the detritus of evolutionary processes — will someday be found to have a function.

(In fact, some “junk” DNA has indeed been found to be functional in recent years, though more than 90 percent of human DNA still appears to be the flotsam of biological history.) In any case, West said, it is up to Darwinists to prove ID wrong.

“Chance and necessity don’t seem to be good candidates for explaining the appearance of higher-order complexity, so the best explanation is an intelligent cause,” West said.

First of all, it’s not at all clear that the burden of proof runs in the direction West says it does. And secondly, if ID advocates are actually asserting that god would never fill humans up with useless appendages, then I’d like them to explain why humans have a tailbone. It’s pretty bloody obvious that the world’s extant organisms are filled with anatomical details that serve no purpose now, but did in the ancient world. Given that the evidence points strongly toward a non-intelligent design of living things in many particulars, it seems to me that the ID folks are the ones with some explaining to do.

Someone who clearly wasn’t qualified to interview Richard Stallman interviewed Richard Stallman, and drew appropriately contemptuous responses:

Sometimes we hear that a company violates the license conditions, but I think there is still a lot of confusion about what should happen next. Could you please explain what type of consequence it brings, and what “terminates the license” means concretely?

Legally, termination of the license means that that person or company no longer has legal permission to redistribute or modify the software in question. If it continues to do so without permission, it is copyright infringement, the copyright holders of the program can sue.

Copyright infringement is not necessarily wrong, but distributing software without respecting the freedom of the users is necessarily wrong. If we ever need to sue to enforce the GPL, the ethical justification won’t be “you disobeyed us” but rather “you are trampling other people’s freedom, and we are here to defend it.”

If the author of GPL says “copyright infringement is not necessarily wrong,” some people could take code covered by GPL and claim that violating GPL terms is “not necessarily wrong.”

I’ve addressed that point in the statement that inspired your question.

The GPL gets its legal force from copyright law, but that is not a source of moral authority, so none can come from there. Why then is it wrong to violate the GPL? Because that tramples other people’s freedom or puts it at risk.

Unrelatedly, Stallman is fond of this sort of language game:

The GNU GPL is based on copyright law. Due to a rather ill-conceived treaty, the Berne Convention, and an extremely nasty treaty officially called TRIPS but which I prefer to call TRIPES—“trade-restricting impediments to production, education, and science”—copyright law is basically the same in most countries around the world.

He brings up another example later on:

Palladium was Microsoft’s name for one particular scheme for Treacherous Computing.  . . . their principal motive is to make Digital Restrictions Management impossible to break with software.

Palladium was part of the “Trusted Computing” initiative. Stallman thinks it matters whether he uses phrases like “Treacherous Computing” or “TRIPES.” Not only do I think these language games (of which “free software” is one, as against “open source”) are silly, but I doubt Stallman is the best one to choose the names even if they weren’t silly. “TRIPES” and “Treacherous Computing” ring hollow.

Apparently I was unaware that there is such a thing as Web 2.0. But I guess that’s what Ajax is supposed to herald. See “The Web 2.0 is here” for links to new cool Ajax apps: BaseCamp, BackPack, Kiko and Meebo all seem cool.

(Via James Grimmelmann’s Del.icio.us feed)