Month: March 2023

A brief Twitter exchange yesterday got me to thinking about one of the favorite former topics here in the discourse-o-sphere: neocons. If you're over 40 50 you remember them: the cabal of foreign policy intellectuals who were behind the invasion of Iraq and had visions of an American imperium that stretched across the entire Middle East. Their names were little known to average folks, but within the national security community they produced many of the ideas that led to the Iraq War.

But time passes and memory fades. Who were these men? There was Doug Feith, “the fucking stupidest guy on the face of the earth,” according to General Tommy Franks. There was his deputy, Paul Wolfowitz, who insisted that General Eric Shinseki was "wildly off the mark" when he said it would take several hundred thousand soldiers to keep peace in postwar Iraq. On the contrary, he told Congress, it would be a piece of cake:

In his testimony, Mr. Wolfowitz ticked off several reasons why he believed a much smaller coalition peacekeeping force than General Shinseki envisioned would be sufficient to police and rebuild postwar Iraq.

He said there was no history of ethnic strife in Iraq, as there was in Bosnia or Kosovo. He said Iraqi civilians would welcome an American-led liberation force that "stayed as long as necessary but left as soon as possible," but would oppose a long-term occupation force. And he said that nations that oppose war with Iraq would likely sign up to help rebuild it.

"I would expect that even countries like France will have a strong interest in assisting Iraq in reconstruction," Mr. Wolfowitz said. He added that many Iraqi expatriates would likely return home to help.

Who else? There was Paul Bremer, who was appointed czar of Iraq after the invasion and promptly disbanded the Iraqi army, got rid of every Baathist he could find, and laid the groundwork for the uprisings that were to come.

And there were the old talkers, people like Richard Perle and Elliott Abrams, who didn't play much of a role in the Bush administration but spoke and wrote a lot about neoconservative ideas.

Whatever happened to these guys? Well, Feith does a bit of speechifiying these days, but not much else. Wolfowitz was appointed president of the World Bank a couple of years after the war started but was forced to resign before the Bush administration even ended.

Paul Bremer paints and serves on a board or two. Perle doesn't do much of anything. Abrams wrote a book and then spent a couple of years in the State Department under Donald Trump.

In other words, not to put too fine a point on it, all of these folks dropped off the face of the earth after the Bush administration ended. Their influence is now about zero.

But that's not really what interests me. Since we're reminiscing about the Iraq War this week, I began thinking about the neocons and reevaluating how much influence they ever had. They certainly influenced the Reagan administration, but that was a couple of decades in the past when George W. Bush came to office.

With the benefit of hindsight, my conclusion is that they had almost no influence at all on the Iraq War. The architects of the war were Dick Cheney, Donald Rumsfeld, and George Bush, none of whom were neocons or anything close to it. They were just ordinary Republican hawks. They didn't care about democracy promotion or an imperial America or any of that. They just wanted to kick Saddam Hussein's butt and take his oil.

For them, the neocons were handy partners—true believers who could help promote a war with Iraq—but that was about it. Cheney was happy to have their help with that, but otherwise they didn't influence his thinking one whit.

So they never got what they wanted. Of course, Cheney didn't get what he wanted either: a quick demolition of Iraq followed by an auction of its oil to the highest (American) bidders. But unlike the neocons, who failed because they were deceived by Cheney and Bush, Cheney failed because events on the ground exploded in his face. Both the neocons and the hawks were stupid, but they were stupid in different ways.

The neocons offered a handy public face to attract the hatred of anti-war activists, but they never had anywhere near the influence we thought they did. They weren't quite useful idiots, but they were in the ballpark.

Brad DeLong continues to worry at the Chat-GPT4 issue as if he were a dog and it was an old shoe. In particular, he wants a chatbot that thinks like Brad DeLong, but it's not working out:

A human neuron would require some 3000 neural-network nodes to model it, with each node connected to some 30 others. That’s 100,000 parameters per neuron, and there are 100 billion neurons. That’s 10 quadrillion parameters for the within-neuron processing alone. And then each neuron is connected to perhaps 1000 other neurons. That smells to me like a network more than 30,000 times as complex as Chat-GPT4.

When you have a neural network 30,000 times as complex as Chat-GPT4 that has been trained by the genetic-survival-and-reproduction algorithm for the equivalent of 500 million years, I give you permission to come knocking at my door.

But this is not the right way to think about it. GPT-4 isn't meant to model Brad DeLong's brain. Or your brain. Or my brain.

Functionally speaking, our brains are made up of modules designed to do highly specific things. There are maybe a hundred altogether, and at most GPT-4 is meant to emulate one or two of them. The other 99 require different pieces of software.

In other words, Brad's arithmetic is off by a factor of about a hundred. All he needs is a network that's about 300 times more complex than GPT-4, which is probably not that far away. Maybe a few years, depending on how improved training interacts with improved software and hardware.

For what it's worth, I also think the piece of text that spurred this conversation is better than Brad gives it credit for. Is it as good as 100% pure Brad? No, but what is? Still, speaking as someone with a little more distance from Brad's brain than Brad himself, the "pretend you're Brad DeLong" answer that GPT-4 generated for him is not bad.

Sure, the real Brad is better, but 30,000 times better? Sadly for us overclocked and egotistical apes, probably not.

First we had the "Great Resignation." Then "quiet quitting." Now we have "Revenge of the Bosses."

The great trend these days seems to be that even CEOs and managers who once supported working from home are souring on it. The Wall Street Journal today has what feels like the hundredth piece I've read about this over the past couple of weeks:

“There’s a lightning-in-a-bottle effect that rallies people together,” says Allan Jones, founder and chief executive of Bambee, a human-resources software and consulting firm in Los Angeles. “But lightning strikes and then it dissipates.”

....Alone in the office when we spoke by video, he picked up his phone to check how many clients were signing up for Bambee’s HR software and management service. “Our new subscribers today are half of what they normally are at this time,” he reported, adding that a 30% drop is typical when everyone is at home. It isn’t principle or personal preference that sours him on remote work, he says. “It’s in the numbers.”

If Jones isn't exaggerating, this is an amazing stat: On days when everyone is working from home, subscriber signups drop 30%.

As regular readers know, I'm skeptical that the WFH revolution will last. There are several reasons for this. First, humans are social creatures and we like to congregate with each other. Second, managing workers remotely is difficult for most people.

Third, and most important, WFH is really, really hard on new employees if they have jobs that require them to work closely with others. An accounts receivable clerk, for example, might do fine. But a new product manager? I've been one, and I can't even imagine what it would have been like trying to learn what I needed to know if everyone I had to work with was available only via Zoom or phone or Slack. It's one thing for existing teams to continue working well from home; it's quite another to get a new member of a team up and running.

In the end, I think the pandemic forced WFH on a lot of companies, and some of it will work fine and stick around. But most of it won't work fine, and both managers and workers will end up wanting to get rid of it. A couple of years from now, my guess is that WFH will be maybe a third higher than it was in 2019—in other words, a rise from around 7% to 9%. We'll see.

Here's my latest brainstorm: we should set the federal minimum wage at half the average local wage, rounded up to the nearest dollar. Here it is for a random selection of places:

This would take care of inflation and geography all at once. And of course states and cities would be free to set higher floors if they wanted to.

NOTE: I'm using the mean local wage instead of the median because it does a better job of capturing high housing prices due to a concentration of rich people.

I just got back from a brain MRI. Apparently I still have one.

I thought I had done one of these before, but I guess not. At least, I don't remember it. Basically, they lock your head down in a little cage and then shove you into a tube only a couple of inches bigger than your head. I've always wondered why people complain about MRIs being claustrophobic, and now I know. I don't happen to be especially claustrophobic myself, but even so it was fairly uncomfortable. I can certainly see how it would seriously spook someone who was claustrophobic.

Anyway, it's easier if you close your eyes. That's my pro tip for the day.

An approaching train at the Vermont/Sunset station on the LA metro red line.

Tick tock. The Fed's rate hike decision is almost here! Can you feel the tension in the air?

POSTSCRIPT: Up a quarter of a point. Idiots. How tight do they think the economy needs to be?

Sam Quinones says he knows the solution to the fentanyl crisis. I was eager to hear it, but it turned out to be this:

It is hard to believe that two nations that have negotiated complex free-trade agreements cannot come to some deeper collaboration on drugs, upheld across presidential administrations and sustained despite distracting conflicts elsewhere in the world.

Quinones isn't even pretending to have a solution. He just can't believe that other people aren't able to come up with one.

But just to state the obvious, sustained crackdowns on illicit drugs have a dubious track record. And the whole problem with fentanyl is that it's used in microscopic quantities that make it a harder law enforcement problem than traditional opioids like heroin or cocaine.

I don't know the answer, myself. If I had to take a crack at it, I'd say that the only chance of long-term success is to reduce demand for opioids, which means . . . something. I don't know what. Because opioids are so addictive, the only solution is to stop people from even wanting to try them in the first place, and decades of efforts have produced no plausible way to do that.

Maybe science could save us? Is it really impossible to develop something with the effect of an opioid that isn't addictive? No one's done it so far, but that doesn't mean it's impossible. Maybe some future AI will do it for us.

Very pretty rainbow(s) over our house to end the day. I guess that's one good thing about endless rainshowers.

According to Datalytics, here's how things look on the homicide front so far in 2023:

Down 10% so far! And that's on top of last year's 5% decline. This isn't enough to make up for the 30% increase during the pandemic, but we're getting there.