Steve Reads

You don’t need to tune into the news about Virginia Tech. In fact you will never need to tune into the coverage from any shooting ever, or really from any tragedy ever. The meta-storyline goes like so (I had Virginia Tech and Hurricane Katrina in mind, but I think it works for many others):

1. Initial reports from the tragedy.
2. Later-refuted rumors from the tragedy.
3. “No one knew the guy was crazy, say neighbors,” or “Everyone knew he was a bad seed, say neighbors.” Or both.
4. Some recommended change in government policy (e.g., gun control, or its opposite)
5. Coverage of the candlelight vigil, or the Proud People Picking Up The Pieces
6. General gawking
7. Proposal, and/or passage, of a bill named after a (white, blonde) victim of the tragedy
8. Everyone forgets

You don’t even need to read the coverage from Virginia Tech. It wrote itself years ago.

I would also like to second Chris.

If I try to play a RealAudio URL in mplayer — say, the .rm file linked from On Point’s interview with George Howell — I can’t. I end up getting “Win32 LoadLibrary” errors. If instead I download the .rm URL with wget, e.g.,

wget -O - [.rm URL] 2>/dev/null 

and play the enclosed RTSP URL with mplayer, everything’s fine.

So why can’t mplayer handle the .rm container?

P.S.: Incidentally, you can use mplayer to dump RealAudio files to MP3s:

mplayer -vo null -vc null [rtsp URL] -ao pcm:file=[/path/to/file.wav] lame -V0 /path/to/file.wav file.mp3 rm file.wav 

This creates file.mp3 out of the RealAudio stream. There’s probably a way to do it using only one command (dump the mplayer WAV-file stream to standard output rather than to file.wav, then pipe that stdout to lame), but this is the lazier approach.

Paul Krugman’s most recent two columns — on how the Democratic Party ought to listen to its base more, and how the GOP has nothing to offer but “little lies” — are great. Highly recommended. They’re included below, in the order mentioned.

(more…)

Just in case you were curious: the slippery-slope argument is not a logical fallacy.

The Bush administration yesterday asked Congress to make more non-citizens subject to intelligence surveillance and to authorize the interception of foreign communications routed through the United States.

Currently, under the 1978 Foreign Intelligence Surveillance Act, individuals have to be associated with a foreign terrorism suspect or a foreign power to fall under the auspices of the FISA court, which can grant the authority to institute federal surveillance. The White House proposes expanding potential targets to include non-citizens believed to possess, transmit or receive important foreign intelligence information, as well as those engaged in the United States in activities related to the purchase or development of weapons of mass destruction.

The proposed revisions to FISA would also allow the government to keep information obtained “unintentionally,” unrelated to the purpose of the surveillance, if it “contains significant foreign intelligence.” Currently such information is destroyed unless it indicates threat of death or serious bodily harm.

And they provide for compelling telecommunications companies and e-mail providers to cooperate with investigations while protecting them from being sued by their subscribers. The legal protection would be applied retroactively to those companies that cooperated with the government after the Sept. 11, 2001, attacks.

The White House draft offered the first specifics of the proposal, which Director of National Intelligence Mike McConnell said Tuesday is needed to respond to “dramatic” changes in communications technology used by intelligence targets in this country.

The proposed changes do not address the controversial intelligence program, initiated in October 2001 and first disclosed in December 2005, that monitors communications between people in the United States and other countries when one party is suspected of having terrorist connections, according to senior administration officials.

The White House also threatened to veto a Senate version of the annual intelligence authorization bill, primarily over provisions that require a response within 15 days to Senate intelligence committee requests for particular documents, and reports to all committee members upon the initiation of extraordinarily sensitive activities, under threat of withholding funds. Under current practice, only committee chairmen and vice chairmen are told of such activities.

The White House, in a “statement of administration policy” sent to the Senate on Thursday, questioned the 4 percent reduction in funding that the intelligence committee applied to national intelligence programs and its threat of prohibiting funding for several classified projects pending reports to the panel.

Saying such provisions are “inconsistent with the need for the effective conduct of intelligence activities  . . .  and legislative-executive comity and cooperation,” the policy document said Bush’s “senior advisers would recommend he veto the bill” if it retains the provisions.

For a logical, economic, and legal treatment of the slippery slope, see Eugene Volokh’s wonderful paper on “The Mechanisms of the Slippery Slope”.

Attorney General Alberto Gonzales will be testifying before the Senate Judiciary Committee on April 17, explaining his and the DoJ’s position on the attorney-general firings. I can’t find the articles just at the moment, but I seem to recall that the White House initially pushed to delay the testimony as long as possible. At some point they thought better of that decision and tried to get an earlier slot; the Senate declined. In the intervening weeks, all sorts of delicious information has come out, proving one person after another to be either a liar or, at best, mistakenly self-contradictory.

The 17th is Tuesday. I for one can’t wait for the testimony. If I didn’t have to work that day, I’d gladly sit at home with a big bowl of popcorn and watch Gonzales get chewed alive.

One thing I’ve wanted for a while is for my computing experience to be identical no matter which machine I go to. My bookmarks would be identical, my browser history, cookies, etc. And that’s just the browser. When I set up, say, an RSS reader (like Akregator, the best Linux reader), I want it to pull down my list of subscribed feeds and a list of which posts I’ve already read.

Of course this all starts to sound very much like a bad attempt to think about new problems with old technologies — like my favorite old standby, namely the early clothes dryers. My friend Seth tells me that the earliest patents in this direction were for devices that automatically pulled clothes out of a washbasin and hung them up to dry; it took rather longer for someone to come up with the idea of a rotating heated drum. If I find a citation for one of these early patents, I’ll post it here.

In any case, hoping for machines to copy down all of our data from one location to another sounds insufficiently webby. If I want shared bookmarks, something like the seamless del.icio.us Firefox integration is closer to the mark. If I want shared RSS content, Google Reader is probably what I want.

But as I wrote in response to Adam Rosi-Kessel’s Google Reader exultations, I don’t believe that we should be planning for all-web-all-the-time. We should plan for a wifi-type world, where connections can fall away and everything still works fine.

The more fundamental reason to want this is that Centralization Is Bad. I have nightmares of what would happen if, say, Google’s machines blew up, or a few key routers (like MAE-East and MAE-West) were destroyed. We simply can’t be relying on central servers for everything.

The Internet is supposed to be radically decentralized, and in many important ways — technologically and philosophically — it is. To me, radical decentralization of news coverage is more important by far than decentralization of wiring, and we certainly have the former. Wifi and its ilk certainly count as radically decentralized wiring, so we have that too. What we should be aiming for is decentralized computing. In part this is because it’s safer: I want a world where 99% of the computers and wires in the world can blow up, and the remaining 1% can still talk to each other just fine. (If DNS disappeared right now, I’m pretty sure we’d all be screwed. This is frightening.)

But in larger part, we want decentralization because it’s more efficient. We have hundreds of millions of computers, each of which is something like a million times as powerful as the average computer from the late 70’s. Most of that computing power is going to waste. How full is the average hard drive? An efficient system would use all that available disk. So it just seems like a waste to turn all our computing power back over to centralized servers; it’s a return to the mainframe era.

That’s why I’d like to see a hybrid model: lots of computationally intensive JavaScript on everyone’s devices, thereby making good use of CPU; lots of caches of everything on local disk (which would require some slight advances in Ajax); networking that expects to fail and does so gracefully; and centralization only as much as is necessary to keep all the machines in sync.

Years ago a friend commented that Groove seemed like overkill, inasmuch as a set of open-source tools available then could do 90% of the same job for free. Of course open-source tools lack Groove’s polish, but I still agree with my friend. I’d prefer to see a world where we weren’t all relying on Google to keep our information in sync; instead we’d let hackers develop backend tools based on something like OpenSSH, sending our data around in the background while our applications are blissfully unaware that there’s a network behind them. This is what radical decentralization means: it means giving people the tools to do things that only corporations used to be able to do.

All of this, ironically, was intended to announce my discovery (at Adam’s prompting) of Google Browser Sync, which will soon be keeping Firefox in sync on all my machines. The growth of the GoogleMind continues.

Note that Google Browser Sync is only available for Firefox. This is, I think, a bit of evidence in favor of Paul Graham’s provocative claim that Microsoft is dead. Microsoft, of course, is not dead. It is highly profitable. It will be highly profitable for years to come. People (Don Dodge, for instance) have misunderstood Graham’s claim, which is mostly because Graham’s own language was inflammatory and the topic itself is so. The best synopsis of his claim is in his essay:

But it’s gone now. I can sense that. No one is even afraid of Microsoft anymore. They still make a lot of money—so does IBM, for that matter. But they’re not dangerous.

Microsoft is dead in the same way that IBM is dead. Which seems to me largely correct. The web is the thing. The web has been the thing for years. It’s probably too early to say that Microsoft has lost the battle for the web, but it’s getting there. Firefox, of course, doesn’t have IE’s market share, but that has to be measured against IE’s presence on every single Windows box in the world. If you want to get a measure of how IE will be doing in a couple years, you should probably look at what web developers prefer to build on. My own informal surveying suggests that IE just isn’t a factor. People test against it because they have to, but Firefox and Safari are what they want to build on.

And open source is quietly taking over. PHP, Perl, Python, and Ruby are the languages of the web now; MySQL is the preferred database backend. Lots of companies still install Exchange servers, of course, but IMAP and POP are where the world is moving. I’d like to see a rigorous look at the numbers on these things, because I am of course talking out of my hat. But I’d have to assume that open standards like JavaScript or PHP are waxing while proprietary technologies like ASP are waning. One good way to measure ascendancy would be to look at job postings requiring expertise in various technologies — say, MySQL versus Oracle, Perl or Python or Ruby versus ASP, IMAP versus Exchange, Apache versus IIS, LDAP versus Active Directory, Bind versus Windows DNS, etc., etc. The largest companies probably still install Exchange or Notes and get expensive support contracts, but I’d have to assume that small-to-medium-sized companies are moving to open standards. For one thing it’s cheaper. For another it avoids lock-in. For a third they’re better technologies.

Don Dodge misunderstands Paul Graham, and he has the numbers to argue his point. He’s not wrong. He’s just arguing the wrong thing. Thought leaders aren’t behind Microsoft anymore. The web is the thing, and Microsoft is a slowly receding force on the web. The world is moving to open-source technologies like Perl and Apache, and moving to open standards like IMAP, JavaScript, or PDF (did you know that Adobe is moving to make PDF an ISO-approved standard?). Whether or not people are aware of it, more and more of their everyday computing experience is conducted using technologies over which Microsoft has no control. Microsoft is fighting against commodity software, just as IBM fought against commodity hardware. Both fights will end the same way.

So says the Census Bureau. Any idea what that’s about? Is it just that people have moved out into the suburbs? Middlesex County’s population is up slightly. The state’s population as a whole declined by 150,000 or so between 2000 and 2005.

P.S.: The Boston-Cambridge-Quincy Metropolitan Statistical Area (M.S.A.) saw a 1.4% increase in population between 2000 and 2006. Interestingly, the Baltimore-Towson M.S.A. saw a 4% rise; I think The Wire has trained me to think that the whole area is emptying.

 . . . was initially the plan, but it was rejected as being “impractical.”

I’ve wondered since the beginning why they didn’t just do this. Because they knew they were going to bring down a shitstorm.

Did they ever resolve who wrote the provision into the Patriot Act that allowed new attorneys general to be appointed without Senate approval? Last I knew, even Arlen Specter didn’t know how that amendment got in there. Senator Feinstein’s bill would reverse the provision, but still: how did it get in there in the first place?

I’m going to go donate another $50 to TPMMuckraker. They’re doing great work.

P.S.: I like the thought that Marshall — bloggers generally, really, but Marshall is doing original reporting that not many others are — is today’s Woodward and Bernstein (all right, he’s just one guy, so let’s say he’s Bernstein). A boy can dream that a year from now, we’ll be hearing about how “bloggers brought down a president.”

Dear web,

I have a club-soda stain in my sofa. What do I do to get it out.

Thanks,
me

I’d like to start a list of journalistic clichés that are hereby forbidden from use by journalists, which means that they’re doubly forbidden for us plebeians.

I inaugurate the list with “firestorm.” Specifically “firestorm of controversy.”

More forthcoming as they arise.

Because this is an all-Cosma-recommendations-all-the-time blog, I should mention that I started reading the above-named microeconomics textbook last night, on Cosma’s recommendation. (To do so, I had to put down John Roemer’s A Future For Socialism for a moment. Note that Cosma’s review of AFFS trumps both its Harvard University Press homepage and Amazon’s page for it. I yearn for the day when I have that kind of Google cred.)

I have almost nothing to say about it just yet, given that I’m only through the prologue. It looks like it will be a good synthesis of the last 20 years of economic research, moving beyond the models of purely self-interested economic man and into something that accords better with experimental evidence.

It also looks deliciously multidisciplinary; the bibliography is quite hefty. I’ve had a post in my head for a while, wondering whether specialization — the academic sort in particular, but industrial and so forth as well — is likely to be counterproductive. On the one hand, it allows people to become domain experts, but on the other hand people are likely to ignore research from neighboring fields that might help them. And each discipline develops a jargon that quickly becomes impenetrable to outsiders, which makes it even harder for others to draw on its research. I hope Bowles’s book lives up to its strong multidisciplinary promise.

But actually, the whole reason I started writing was for a much more cosmetic reason: Bowles’s book is gorgeous. The cover is Ambrogio Lorenzetti’s “Allegory of Good Government In The City,” and all the text — including the titles on the cover, I believe — is set in Sabon. It’s a joy just to read it. I hope and expect that the text itself lives up to the design.

(Yes, I’m a book-design fetishist. So sue me. I read a lot of books.)

Incidentally, there’s another “Allegory of Good Government” that appears on the cover of the Penguin edition of Ruskin’s Unto This Last And Other Writings. The Wikipedia explains.

Somehow I still end up surprised every time I find a new Elvis Costello album that turns out to be a masterpiece. This is rather strange, given that I’m on at least my sixth such album by now. All of the following count as must-owns:

• The Delivery Man (loosely themed around an analogy between the Bible and a love affair, kind of)
• (with the Brodsky Quartet) The Juliet Letters (concept album — and the concept is supremely cool)
• Blood and Chocolate (angry post-divorce music, including the classic “I Want You” and at least two other lesser-known tracks that deserve to be classics)
• My Aim Is True (Costello’s first album — straight-ahead power pop with amazing lyrics)
• Spike (“Veronica” is probably the most famous track off this album, but eight other tracks are worth the price of admission)
• When I Was Cruel (harder rock; seven tracks on here are indispensable)

There are probably others, but that will do for now. It’s worth noting that although I don’t like Punch The Clock much at all, I think the tracks “Everyday I Write The Book,” “The World And His Wife,” and “Shipbuilding” are quite marvelous. The first one, in particular, seemed like cheese when I was growing up but actually is far more clever than a first listen would suggest.

P.S.: I forgot Almost Blue, which is a great country-music tribute album of sorts. It’s worth owning for “Brown to Blue” (“You changed your name from Brown to Jones and mine from Brown to Blue”) alone.

The common perception seems to be that statistics can say whatever you want them to say. In my experience, that is flatly untrue. Badly chosen statistics can say whatever you want them to say, but anyone with an elementary training in the discipline — which is to say, anyone who’s even read How To Lie With Statistics (142 pages long, written 53 years ago, available for less than $5 used on Amazon) — can spot the nonsense from a mile away.

The more accurate statement is that statistics can be misinterpreted by the misinformed to say whatever a liar wants them to say. Which shifts the burden from the statistics themselves onto the education system that allows charlatans to prosper. The problem is not the tools, but those on whom the tools are used.

Of course that brings in other problems. For instance, it shifts the burden onto educators to get statistical training out to more people. The burden falls somewhat on journalists as well, but they’re more often the problem than the solution. (An article in my hometown newspaper recently reported that Vermont’s consumption of gasoline is among the lowest — it may be in fact the lowest — of any state in the nation. Question: why is this not an interesting assertion?) Journalists need to be educated just as well as the rest of society, and often they’re not.

I’d like every high school in this country, before it tries teaching its students calculus, to teach them a class whose provisional title is Spotting BS In Newspapers And On TV. It would do a world of good.

Suppose, rather than solving the whole problem at once, that the federal government just provided a little health care for free — services that are preventative and more or less universally supported. E.g., the government will pay for one annual mammogram for every woman over 50, will pay for every child’s immunization shots, will pay for an annual wellness checkup, etc. Suppose the government sets a price that it will pay for each of these — the same price that the health-insurance companies already pay for them.

First, couldn’t a very easy case be made that preventative services pay for themselves?

The natural economic question is: would providing these services for free lead to long waiting lines? I of course don’t know the answer to that question offhand. But it doesn’t seem like it should be hard to structure the system so that this problem is minimized.

Rather than dive into the whole debate about “socialized medicine,” with its attendant horror stories of people waiting six months for a CAT scan, shouldn’t we start with the easy stuff?

P.S.: Sick : the untold story of America’s health care crisis, and the people who pay the price goes on the queue.

I think I may need to get one of these.

All geeks prefer writing dates in the form “yyyy-mm-dd”, so that they sort lexicographically.

I think the part that frustrates me the most about The Market System is that it’s leaving things abstract or simplified that ought to be concrete and detailed. E.g.,

The principal limitation on [the market system’s] domain, we already know, is that it can coordinate only those allocations that are voluntary. Market participants can obtain objects and performances only through offers of benefits, with both buyer and seller acting voluntarily. The market system can coordinate the construction of a skyscraper or the coordination of 10,000 workers to provide financial services to investors. But it cannot coordinate — purchaes cannot achieve — the acquisition of land for a highway. At least some owners must be compelled to surrender.

Is this really clear? If you’re a strong American libertarian — a viewpoint that Lindblom dismisses in a sentence somewhat early on — you’ll probably respond that, in fact, people would voluntarily cede their land if a government weren’t there to compel the takeover. You just need to pay them enough. If someone offered to pay me a billion dollars to build a vital segment of the Interstate Highway System through my house, I would certainly do it. I can see two responses to this:

1. There are some people who simply will not part with a plot of land at any price. These people need to be compelled.
2. Yes, you’d cede your land for a billion dollars, but that’s not an efficient price; society should pay less than a billion dollars for that road.

I can see the merit in these arguments, but they lead to a whole host of other questions, such as: is it really just to make people give up their land at a price below what it’s worth to them? Suppose society values that land at $100, but the owner values it at $1 million; who decides that the $100 price is the one that it will sell for? And who’s going to compute that $100 is society’s value? Presumably society gets a great deal of value out of moving from point A to point B at 65 miles per hour. Shouldn’t the landowners pick up some of that value?

So take this specific question of eminent domain. It could probably produce at least a decent-sized essay. Condensing it into a paragraph doesn’t satisfy anyone, it seems to me. There’s not enough theory to answer the question — only enough of a synopsis to frustrate the reader.

Lindblom’s goal, stated explicitly in the introduction, is to stand back and look at the system as a whole in a way that undergraduate economics classes tend not to do — to give us the forest where normally we’d get the trees. But it seems to me that looking very carefully at one particular tree, in this case, would give a much better view of the forest than Lindblom does.

If anyone has access to an electronic version of the journal Nature, I would most appreciate a copy of

Ziman, J.M. “Information, communication, knowledge.” Nature 224:318-324 (1969)

P.S. (8 April 2007): Many thanks to two readers for “hooking me up.”

I’m reading through the article and, per usual, wondering how the Internet changes what Ziman wrote in 1969. His argument is basically a conservative one, at least so far — namely that scientific journals are perfectly right to move slowly, and that doing so helps scientists filter out the wheat from the chaff. Attempts to turn less-formal communication — like scientific letters — into something more formal (e.g., journals devoted only to those letters) will only decrease the signal-to-noise ratio, says Ziman, and should be avoided.

But humans seem to be very good filterers. We certainly find ways to filter out junk on the web. Hundreds of millions of people, each making their own decisions about what’s wheat and what’s chaff, seem to spot all the good stuff.

I wonder how systems like the arXiv work. I’m not super-familiar with the arXiv, but my sense is that most anyone can post to it. As an empirical matter, I wonder whether even a low signal-to-noise ratio on the arXiv matters.

I’ll just be a whore and put it out there: if you’re one of the eight trillion people who will be buying Harry Potter and the Deathly Hallows, you can buy it off my site and I’ll probably get a commission. I myself am not going to buy it, but I know that many of my readers will. Why not send a little baksheesh my way? If not baksheesh, then how about a lagniappe?

(If the Harry Potter title above isn’t linked, it will be in a little while, owing to the magic of Amazon Context Links.)

On Cosma’s advice, I got Lindblom’s The Market System: What It Is, How It Works, And What To Make Of It out of the library. Cosma recommended it as an introductory work of economics, on par with Heilbroner’s The Worldly Philosophers — i.e., more biographical and narrative than a technical introduction to micro- or macroeconomics. (For the latter, he suggested Samuel Bowles’s Microeconomics: Behavior, Institutions, and Evolution. That’s on its way.)

About 1/3 of the way through it, I can’t recommend it if you’ve already steeped yourself in, say, Richard Posner — which is to say, you’ve been repeatedly smashed over the head with Coase’s “Theorem”. If you’ve done so, then you already understand the premise that a corporation arises when the transaction costs of buying in the open market become too great. Thus far, 80 pages in, that’s about the most complicated idea that I’ve encountered. The preceding 79 pages have tried to divorce the reader’s mind from a couple conceptions of markets — e.g., that market transactions are purely economic. Again, anyone who’s read Posner will have long since encountered Gary Becker’s work extending economic models to non-market transactions like choosing a spouse.

It’s trying to expand the reader’s mind to envision markets in a different way than we’re used to, but I actually find its scope rather limited. I’m not sure that Lindblom has much grasp on world history, or maybe he does and just doesn’t want to overwhelm his reader with what seems like abstruse detail. If he really wanted people to understand what markets can be, he’d scale out to societies from well before the Industrial Revolution, and maybe even use examples from ancient Greek and Chinese literature. Maybe that wouldn’t play well with his goal; it seems that he may be aiming at responses to the modern multinational corporation. Still, I’d appreciate some more stepping back; even my recent reading of Paul Krugman suggests that our view of the modern global economy is at odds with our history. If he really wanted to clear ground, Lindblom might be more of a historian.

All of that is said, though, with only 1/3 of the book done. I’ll finish it and see if it improves.