I worked for the IRS Chief Counsel’s office at a level where I had to be aware of budgeting. The coin of the realm was the “staff year.” A staff year was the salary of a single attorney at the grade 14 level, the journeyman level. Every category of budgeting was expressed in terms of staff years allocated to that category. X number of staff years for furniture, X number for travel, X number for renting office space, etc. Though I never looked into it, I expect that a similar situation obtains in the Commissioner’s office. That is, their budget is expressed in terms of an agent staff year at whatever the GS level is for their journeyman agent. So 87,000 agents probably doesn’t mean that 87,000 agents will be inflicted on the public. It’s just a way of allocating money to the IRS that will be spent on updating the computer system, hiring people to process returns & answer the phones, buying furniture, etc. I haven’t looked at the legislation, but I doubt that it actually specifies hiring 87,000 agents. Even if it does, this agents will be considered staff years and be converted into spending on other things.
I have this argument about Amazon fakes with a friend of mine on a regular basis, with him playing the part of Dan Luu and me playing the part of Zvi. I wonder what the different heuristics applied to search are?
I’ve ordered about 50 items from Amazon in the last year, and two of them were scams. Both times I got a refund without issue. Is that a high amount or a low amount? I think it’s kind-of high and have become more cautious about ordering things on Amazon, but I haven’t completely stopped or anything. But someone else might think that’s low, or just gotten slightly luckier and encountered 0 or 1 scams and think there’s no problem.
That's much higher than my experience - in my 20+ years on Amazon I can only think of 1 product I've bought that was an obvious scam (1TB USB drive that was really only 100GB). But maybe some people consider low quality products as 'scams'? But even then my experiences there are still probably much lower than 5%.
They were both outright scams where the correct product was never delivered (and the seller made attempts to obfuscate this, so the package wasn’t just lost). Until this year the number of scams I’d encountered was zero, so I’m not sure if this is just bad luck or things getting worse. In retrospect one of the listings had red flags that I could’ve noticed if I’d been paying closer attention, but the other had nothing obviously wrong with it, and in any case, having to carefully scrutinize each thing I buy to make sure it’s not a scam is something that makes me want to buy from Amazon less.
We've had some issues but I think all with third party sellers we knew about and also we order tons of packages. At core, when I order I don't worry I'm not going to get the right product. Friend bought all my computer's parts via Amazon and it was fine, etc. So this is all so strange.
On booster effectiveness, the paper cited by Zeynep says that previous infection reduces risk of future infection by 50%, and that a booster on top of previous infection moves the number to 80%. Those are suspiciously clean multiples of ten, looking at the numbers in their chart I see 54% and 76.3%. That the rounding in their discussion section had such a pro-booster bias isn't confidence inspiring. If you must round, and you absolutely do not have to, why not round to the nearest multiple of five?
I've been twice vaccinated and previously infected. Flipping the numbers around, this means I have a 46% chance of catching COVID again, but can reduce this to 23.7% with a booster.
The second dose kept me away from work for about three days. My infection knocked me out for about seven days, and was much more unpleasant.
If I roll the dice on not getting boosted, I have about a 50% chance that I'll spend zero days with any kind of sickness. With the booster I'll almost certainly be sick enough that I'd have to miss some days of work, and even then I still have about a 25% chance of getting COVID anyway and missing even more.
Neither option seems like a clear winner, but with these numbers I'm slightly against getting boosted. A 50% chance at avoiding any sickness at all seems pretty good weight against a near certainty of some vaccine-induced illness.
Maybe we should stop thinking about COVID vaccine effectiveness in terms of percent chance, and instead mean time between illnesses. The percent chance is always over some arbitrary study period, which makes it harder to reason about. Imagine instead being told that you'll get COVID every six months without a booster, or every 11 months with a booster.
Another consideration is that COVID symptoms and risks seem to get milder with each successive infection, while vaccine symptoms and risks stay the same or even get worse with each successive shot.
The Amazon knockoff problem is very real, but depends on the category of item. Small electronics are one frequent offender. You have to check the *recent* reviews (often requires several clicks) to see when a previously valid product has been taken over by a dishonest seller. There are certain things I absolutely won't buy on Amazon, e.g. replacement electric razors.
Usually you can still trust known brands if you buy from them directly via Amazon's site - but not always, because of inventory comingling. Hacker news has a thread on this topic every couple of months.
Conspiracy take: Amazon deliberately allows this behavior to drive customers toward Amazon Basics, which of course do not get comingled.
In Sweden, the vaccine is ”only” illegal under-12. Same thing for Paxlovid for under-65 (or for very limited risk groups) with the absurd claim that it doesn’t work for younger people.
Our IRS is superb though, and I do my taxes (including handling any additional tax payments) completely online and in under 15 minutes once per year, without having having to give it a thought for the rest of the year.
How good is it? It’s the most trusted part of the bureaucracy, even though it is *the IRS*.
I worked for the IRS Chief Counsel’s office at a level where I had to be aware of budgeting. The coin of the realm was the “staff year.” A staff year was the salary of a single attorney at the grade 14 level, the journeyman level. Every category of budgeting was expressed in terms of staff years allocated to that category. X number of staff years for furniture, X number for travel, X number for renting office space, etc. Though I never looked into it, I expect that a similar situation obtains in the Commissioner’s office. That is, their budget is expressed in terms of an agent staff year at whatever the GS level is for their journeyman agent. So 87,000 agents probably doesn’t mean that 87,000 agents will be inflicted on the public. It’s just a way of allocating money to the IRS that will be spent on updating the computer system, hiring people to process returns & answer the phones, buying furniture, etc. I haven’t looked at the legislation, but I doubt that it actually specifies hiring 87,000 agents. Even if it does, this agents will be considered staff years and be converted into spending on other things.
I have this argument about Amazon fakes with a friend of mine on a regular basis, with him playing the part of Dan Luu and me playing the part of Zvi. I wonder what the different heuristics applied to search are?
I’ve ordered about 50 items from Amazon in the last year, and two of them were scams. Both times I got a refund without issue. Is that a high amount or a low amount? I think it’s kind-of high and have become more cautious about ordering things on Amazon, but I haven’t completely stopped or anything. But someone else might think that’s low, or just gotten slightly luckier and encountered 0 or 1 scams and think there’s no problem.
That's much higher than my experience - in my 20+ years on Amazon I can only think of 1 product I've bought that was an obvious scam (1TB USB drive that was really only 100GB). But maybe some people consider low quality products as 'scams'? But even then my experiences there are still probably much lower than 5%.
They were both outright scams where the correct product was never delivered (and the seller made attempts to obfuscate this, so the package wasn’t just lost). Until this year the number of scams I’d encountered was zero, so I’m not sure if this is just bad luck or things getting worse. In retrospect one of the listings had red flags that I could’ve noticed if I’d been paying closer attention, but the other had nothing obviously wrong with it, and in any case, having to carefully scrutinize each thing I buy to make sure it’s not a scam is something that makes me want to buy from Amazon less.
We've had some issues but I think all with third party sellers we knew about and also we order tons of packages. At core, when I order I don't worry I'm not going to get the right product. Friend bought all my computer's parts via Amazon and it was fine, etc. So this is all so strange.
On booster effectiveness, the paper cited by Zeynep says that previous infection reduces risk of future infection by 50%, and that a booster on top of previous infection moves the number to 80%. Those are suspiciously clean multiples of ten, looking at the numbers in their chart I see 54% and 76.3%. That the rounding in their discussion section had such a pro-booster bias isn't confidence inspiring. If you must round, and you absolutely do not have to, why not round to the nearest multiple of five?
I've been twice vaccinated and previously infected. Flipping the numbers around, this means I have a 46% chance of catching COVID again, but can reduce this to 23.7% with a booster.
The second dose kept me away from work for about three days. My infection knocked me out for about seven days, and was much more unpleasant.
If I roll the dice on not getting boosted, I have about a 50% chance that I'll spend zero days with any kind of sickness. With the booster I'll almost certainly be sick enough that I'd have to miss some days of work, and even then I still have about a 25% chance of getting COVID anyway and missing even more.
Neither option seems like a clear winner, but with these numbers I'm slightly against getting boosted. A 50% chance at avoiding any sickness at all seems pretty good weight against a near certainty of some vaccine-induced illness.
Maybe we should stop thinking about COVID vaccine effectiveness in terms of percent chance, and instead mean time between illnesses. The percent chance is always over some arbitrary study period, which makes it harder to reason about. Imagine instead being told that you'll get COVID every six months without a booster, or every 11 months with a booster.
Another consideration is that COVID symptoms and risks seem to get milder with each successive infection, while vaccine symptoms and risks stay the same or even get worse with each successive shot.
The Amazon knockoff problem is very real, but depends on the category of item. Small electronics are one frequent offender. You have to check the *recent* reviews (often requires several clicks) to see when a previously valid product has been taken over by a dishonest seller. There are certain things I absolutely won't buy on Amazon, e.g. replacement electric razors.
Usually you can still trust known brands if you buy from them directly via Amazon's site - but not always, because of inventory comingling. Hacker news has a thread on this topic every couple of months.
Conspiracy take: Amazon deliberately allows this behavior to drive customers toward Amazon Basics, which of course do not get comingled.
In Sweden, the vaccine is ”only” illegal under-12. Same thing for Paxlovid for under-65 (or for very limited risk groups) with the absurd claim that it doesn’t work for younger people.
Our IRS is superb though, and I do my taxes (including handling any additional tax payments) completely online and in under 15 minutes once per year, without having having to give it a thought for the rest of the year.
How good is it? It’s the most trusted part of the bureaucracy, even though it is *the IRS*.
> But it sounds like only after the Omicron variant boosters failed did the push begin in earnest.
Wait, when did the Omicron boosters fail? Link please!
Hey Zvi, have you seen this interview in current affairs from Jeffery Sachs? I'm shocked it took me 2 weeks to see this
https://www.currentaffairs.org/2022/08/why-the-chair-of-the-lancets-covid-19-commission-thinks-the-us-government-is-preventing-a-real-investigation-into-the-pandemic
I had not.