@mwilbert.bsky.social
People were widely pointing out they were bubbles. The recessions were not as clear. It's easier to see a bubble than to predict the effect of it deflating.
Ending this bubble seems apt to cause a recession because of the magnitude of the spending, but it's still less clear than the bubble itself.
Ending this bubble seems apt to cause a recession because of the magnitude of the spending, but it's still less clear than the bubble itself.
November 30, 2025 at 10:22 PM
People were widely pointing out they were bubbles. The recessions were not as clear. It's easier to see a bubble than to predict the effect of it deflating.
Ending this bubble seems apt to cause a recession because of the magnitude of the spending, but it's still less clear than the bubble itself.
Ending this bubble seems apt to cause a recession because of the magnitude of the spending, but it's still less clear than the bubble itself.
Is this something anyone is worrying about? Do people think this will stop because the hyperscalers run out of money? I guess that's a possible theory, but it seems more likely that they (at least some of them) slow down because there's not going to be a decent return on their investment.
November 28, 2025 at 10:37 PM
Is this something anyone is worrying about? Do people think this will stop because the hyperscalers run out of money? I guess that's a possible theory, but it seems more likely that they (at least some of them) slow down because there's not going to be a decent return on their investment.
I don't feel reassured.
November 28, 2025 at 2:05 AM
I don't feel reassured.
If only you could tell. I've also gotten out of some good trades way too late. Nowadays I tend to scale out which may not end up being more profitable but which I definitely find less annoying.
November 27, 2025 at 4:28 PM
If only you could tell. I've also gotten out of some good trades way too late. Nowadays I tend to scale out which may not end up being more profitable but which I definitely find less annoying.
That's a huge penalty for the AWD. Wonder what the story is with that.
November 27, 2025 at 4:11 AM
That's a huge penalty for the AWD. Wonder what the story is with that.
And, if so, does gold then continue to a blow-off top around $8K in 2026? I really don't expect that to happen, but maybe it could. I remember 1979-80, and of course people didn't expect that either. I think miners should be better than metal unless the stock market tanks, but I'm holding both now.
November 27, 2025 at 12:34 AM
And, if so, does gold then continue to a blow-off top around $8K in 2026? I really don't expect that to happen, but maybe it could. I remember 1979-80, and of course people didn't expect that either. I think miners should be better than metal unless the stock market tanks, but I'm holding both now.
Maybe. I think it's just blowback from people who have some understanding of what poverty is. Certainly numeracy helps, but it's just a basic smell test to start with. I have the advantage of having a number of not-affluent-but-not-poor relatives.
November 26, 2025 at 11:07 PM
Maybe. I think it's just blowback from people who have some understanding of what poverty is. Certainly numeracy helps, but it's just a basic smell test to start with. I have the advantage of having a number of not-affluent-but-not-poor relatives.
Let's just say that if there are any aliens listening, they would do all our brains a favor by abducting Hassett as soon as possible. If they can time-travel, they should go back in time and do it several years ago.
November 26, 2025 at 2:04 PM
Let's just say that if there are any aliens listening, they would do all our brains a favor by abducting Hassett as soon as possible. If they can time-travel, they should go back in time and do it several years ago.
The obvious (and I think correct) answer is yes, they are now more reluctant to hire people. Continuing claims keeps going up, admittedly slowly, but up nonetheless.
November 26, 2025 at 2:01 PM
The obvious (and I think correct) answer is yes, they are now more reluctant to hire people. Continuing claims keeps going up, admittedly slowly, but up nonetheless.
My cranberry pie has to be made the day before. Maybe not if we ate really late, but we do not.
November 26, 2025 at 3:34 AM
My cranberry pie has to be made the day before. Maybe not if we ate really late, but we do not.
Yes, I think they want Starship to make Starlink more economic/more capable. But they aren't just spending money on the lift capability, but on the ancillary stuff required for Mars missions. That's a bunch of additional costs for a company whose revenue isn't currently that large.
November 25, 2025 at 4:25 PM
Yes, I think they want Starship to make Starlink more economic/more capable. But they aren't just spending money on the lift capability, but on the ancillary stuff required for Mars missions. That's a bunch of additional costs for a company whose revenue isn't currently that large.
Could be, but at this point it has such a dominant position in the satellite launch business that I don't imagine it's going away. Being private, we don't really know what the economics are, but they could improve them by ignoring Mars completely.
November 25, 2025 at 4:12 PM
Could be, but at this point it has such a dominant position in the satellite launch business that I don't imagine it's going away. Being private, we don't really know what the economics are, but they could improve them by ignoring Mars completely.
Of course I can't prove this, but my working theory is that he used to function better but decades of self-destructive behavior have eroded both his judgement and capability. I assume he'll still be very rich after Tesla implodes, and SpaceX isn't an AI play, but it won't be a happy time for him.
November 25, 2025 at 4:01 PM
Of course I can't prove this, but my working theory is that he used to function better but decades of self-destructive behavior have eroded both his judgement and capability. I assume he'll still be very rich after Tesla implodes, and SpaceX isn't an AI play, but it won't be a happy time for him.
Aside from the pre-existing issues with CoreWeave's business model, stronger competition for NVDA from Google TPU's is yet another risk. A risk people should have been thinking about before, but of course seeing it in action concentrates the mind.
November 25, 2025 at 2:19 PM
Aside from the pre-existing issues with CoreWeave's business model, stronger competition for NVDA from Google TPU's is yet another risk. A risk people should have been thinking about before, but of course seeing it in action concentrates the mind.
AI can be very popular and widely used and the space can still be significantly overpriced. To me, the question is what the margins end up being for the various players, and there are a lot of variables.
November 25, 2025 at 1:38 AM
AI can be very popular and widely used and the space can still be significantly overpriced. To me, the question is what the margins end up being for the various players, and there are a lot of variables.
It's a floor wax AND a dessert topping!
But yes, hard to manage a politics that at the same time condemns the existence of billionaires and is oriented to giving them control of everything.
But yes, hard to manage a politics that at the same time condemns the existence of billionaires and is oriented to giving them control of everything.
November 24, 2025 at 7:12 PM
It's a floor wax AND a dessert topping!
But yes, hard to manage a politics that at the same time condemns the existence of billionaires and is oriented to giving them control of everything.
But yes, hard to manage a politics that at the same time condemns the existence of billionaires and is oriented to giving them control of everything.
Maybe this will work, but I am skeptical they can scale this adequately before they run out of money. I'd call myself an early LLM/AI adopter, and I use AI constantly, but I'm not going to let agents do anything involving money for me for a long time, if ever. And Google's not going to do this?
November 24, 2025 at 6:30 PM
Maybe this will work, but I am skeptical they can scale this adequately before they run out of money. I'd call myself an early LLM/AI adopter, and I use AI constantly, but I'm not going to let agents do anything involving money for me for a long time, if ever. And Google's not going to do this?
It's a pain to get to on foot, but it's neat to see.
November 24, 2025 at 6:23 PM
It's a pain to get to on foot, but it's neat to see.
Or in this case, disservices.
This is simply a continuation of a trend that's been happening ever since the internet became widespread, but as they say, quantity has a quality all its own, and certainly there will be ever-increasing quantities.
This is simply a continuation of a trend that's been happening ever since the internet became widespread, but as they say, quantity has a quality all its own, and certainly there will be ever-increasing quantities.
November 24, 2025 at 6:20 PM
Or in this case, disservices.
This is simply a continuation of a trend that's been happening ever since the internet became widespread, but as they say, quantity has a quality all its own, and certainly there will be ever-increasing quantities.
This is simply a continuation of a trend that's been happening ever since the internet became widespread, but as they say, quantity has a quality all its own, and certainly there will be ever-increasing quantities.
I don't know whether I agree with this particular case, but whether I do or not, New York state never did anything to make this illegal after the Kelo decision ruled such takings constitutional.
I think there should be a high bar for a taking like this, but I don't think it should be impossible.
I think there should be a high bar for a taking like this, but I don't think it should be impossible.
November 24, 2025 at 4:58 PM
I don't know whether I agree with this particular case, but whether I do or not, New York state never did anything to make this illegal after the Kelo decision ruled such takings constitutional.
I think there should be a high bar for a taking like this, but I don't think it should be impossible.
I think there should be a high bar for a taking like this, but I don't think it should be impossible.
I don't think that's true, but in any case that is not a problem for me; I can evaluate it case-by-case regardless of intent. The problem is that I don't want to ingest low signal-to-noise data sources. I spend too much time processing information already, and people who emit nonsense waste my time.
November 24, 2025 at 4:39 PM
I don't think that's true, but in any case that is not a problem for me; I can evaluate it case-by-case regardless of intent. The problem is that I don't want to ingest low signal-to-noise data sources. I spend too much time processing information already, and people who emit nonsense waste my time.
I don't dislike Noah Smith as much as many on this platform seem to, but this was such a ridiculous take that I'm rethinking my position. In my opinion, he has a number of good ideas, but there's a limit to how much complete nonsense I'm willing to accept as an admixture.
November 24, 2025 at 4:26 PM
I don't dislike Noah Smith as much as many on this platform seem to, but this was such a ridiculous take that I'm rethinking my position. In my opinion, he has a number of good ideas, but there's a limit to how much complete nonsense I'm willing to accept as an admixture.
1) agree
2) it would be useful (and unlikely) to reach some kind of an agreement on what would constitute a minimally acceptable living standard. There's way too much "they have cell phones so they can't be poor" type discourse.
3) universal child care and health care would simplify this.
2) it would be useful (and unlikely) to reach some kind of an agreement on what would constitute a minimally acceptable living standard. There's way too much "they have cell phones so they can't be poor" type discourse.
3) universal child care and health care would simplify this.
November 24, 2025 at 4:23 PM
1) agree
2) it would be useful (and unlikely) to reach some kind of an agreement on what would constitute a minimally acceptable living standard. There's way too much "they have cell phones so they can't be poor" type discourse.
3) universal child care and health care would simplify this.
2) it would be useful (and unlikely) to reach some kind of an agreement on what would constitute a minimally acceptable living standard. There's way too much "they have cell phones so they can't be poor" type discourse.
3) universal child care and health care would simplify this.
In an educational context where the primary goal is learning stuff, of course it makes sense to do what maximizes learning. But that's not true in a work context--you have to balance employee development with labor cost and getting work done. Just what's needed to finish the task is often correct.
November 22, 2025 at 3:04 AM
In an educational context where the primary goal is learning stuff, of course it makes sense to do what maximizes learning. But that's not true in a work context--you have to balance employee development with labor cost and getting work done. Just what's needed to finish the task is often correct.
Right, but shipping water from the Mississippi might be easier than moving Tehran. Both seem hard.
November 21, 2025 at 7:08 PM
Right, but shipping water from the Mississippi might be easier than moving Tehran. Both seem hard.