3 ms·

OK, but conversely there were literally no mRNA vaccines mass produced a year ago, and now they are delivering hundreds of millions of doses this year - so they

by feral 6y ago

OK, but conversely there were literally no mRNA vaccines mass produced a year ago, and now they are delivering hundreds of millions of doses this year - so they've clearly being able to scale production quite quickly. Its a big claim to say another $100B wouldn't have scaled it faster. I'm sure they still have bottlenecks, and financial risks they weren't willing to take. $100B is a lot of resources - but a justified amount given the economic costs, which is the point.

epistasis 6y ago
If you're writing software, and want the project to be completed faster, does throwing another $100B at a project make it go faster? Usually, this will slow it down, not make it go faster. A classic of the software engineering industry, The Mythical Man Month, is all about this process. This is a similar sort of process. There's extremely specialized knowledge held by only a tiny number of people, and though there is a plan to vastly expand production, it's never been done before. The bottleneck is not money, the bottleneck is people. People who have undoubtedly been working themselves to the bone to get this done as quickly as possible.
- feral 6y ago
  >If you're writing software Yes, 100% yes. Adding substantial resources can hugely speed up a project. I will requisition the very best engineers, the best support staff, remove all financial bottlenecks, dedicate shadowing teams to work in isolation in parallel on the critical sections, etc. > A classic of the software engineering industry, The Mythical Man Month, is all about this process. First off, if i recall correctly, and it's probably been 20 years since I read it, that mostly applies to late projects that have already gone off the rails. The broader point isn't true anyway. We're way better at writing software than we used to be. I've personally been on quite a few software projects where they became more strategic or promising than they initially seemed, and we threw more engineers at them to speed them up, and they sped up excellently.
- fennecfoxen 6y ago
  They're not writing software. Insofar as the vaccine is like software, they wrote the software in about a week. The problem is deploying the software. This is a mechanical problem. It involves setting up assembly lines. These lines consist of expensive medical equipment, highly skilled installation experts, and skilled operators. $100 billion could definitely buy or lease more equipment, get it installed, get it inspected, and pay more operators to operate the equipment. It could pay for for scientists and technicians working in related med-tech fields. It could pay their current employers to lend them from their current jobs, guaranteeing their positions upon return. The money could likewise supply any needed training, and if necessary, it could pay them extra money and extra vacation to compensate for the disruption in their lives from working third-shift, so that the machines run overnight. Money to the tune of billions and hiring lots of the best people is specifically what makes it so people don't have to "[work] themselves to the bone".
  - bluGill 6y ago
    Money and time. You can't make the assembly lines overnight, but enough money could make them in a few months. Just copy the designs for the machines they already have. There is probably opportunity to make more efficient machines as well, there almost always is, but that takes longer than a blind copy. Better in the long run, but takes longer to work out the bugs in the new design.
- YZF 6y ago
  This is very different than software. We have plants producing this, we only need to replicate them, all the machines in those plants have drawings, the plants have drawings, the process is documented. It's not like some Ph.D. dude is mixing up liquids in a test tube. Now granted any money in the world won't have the plant replicated by tomorrow, but enough money will have it replicated much faster. And the process could have been started months ago. Pretty much any vendor will do things faster if you pay them more money and I'm sure this is true for Pfizer's various vendors here. You can order some machine part produced in 3 months or in 3 days (with the latter being like x20 the price, but who cares now).
- monkeypizza 6y ago
  Yet you presume that the current price is magically the correct one - even though that price is different by region. What would you say the effect of the government offering half as much would be? 1/10th? At one point does economics matter? I think once you admit money matters on the downside, you have a hard time proving that the point it zeros out of effect as it increases exactly matches the arbitrary point it was actually set at.

by feral 6y ago

epistasis 6y ago

If you're writing software, and want the project to be completed faster, does throwing another $100B at a project make it go faster? Usually, this will slow it down, not make it go faster. A classic of the software engineering industry, The Mythical Man Month, is all about this process. This is a similar sort of process. There's extremely specialized knowledge held by only a tiny number of people, and though there is a plan to vastly expand production, it's never been done before. The bottleneck is not money, the bottleneck is people. People who have undoubtedly been working themselves to the bone to get this done as quickly as possible.

feral 6y ago
>If you're writing software Yes, 100% yes. Adding substantial resources can hugely speed up a project. I will requisition the very best engineers, the best support staff, remove all financial bottlenecks, dedicate shadowing teams to work in isolation in parallel on the critical sections, etc. > A classic of the software engineering industry, The Mythical Man Month, is all about this process. First off, if i recall correctly, and it's probably been 20 years since I read it, that mostly applies to late projects that have already gone off the rails. The broader point isn't true anyway. We're way better at writing software than we used to be. I've personally been on quite a few software projects where they became more strategic or promising than they initially seemed, and we threw more engineers at them to speed them up, and they sped up excellently.
fennecfoxen 6y ago
They're not writing software. Insofar as the vaccine is like software, they wrote the software in about a week. The problem is deploying the software. This is a mechanical problem. It involves setting up assembly lines. These lines consist of expensive medical equipment, highly skilled installation experts, and skilled operators. $100 billion could definitely buy or lease more equipment, get it installed, get it inspected, and pay more operators to operate the equipment. It could pay for for scientists and technicians working in related med-tech fields. It could pay their current employers to lend them from their current jobs, guaranteeing their positions upon return. The money could likewise supply any needed training, and if necessary, it could pay them extra money and extra vacation to compensate for the disruption in their lives from working third-shift, so that the machines run overnight. Money to the tune of billions and hiring lots of the best people is specifically what makes it so people don't have to "[work] themselves to the bone".
- bluGill 6y ago
  Money and time. You can't make the assembly lines overnight, but enough money could make them in a few months. Just copy the designs for the machines they already have. There is probably opportunity to make more efficient machines as well, there almost always is, but that takes longer than a blind copy. Better in the long run, but takes longer to work out the bugs in the new design.
YZF 6y ago
This is very different than software. We have plants producing this, we only need to replicate them, all the machines in those plants have drawings, the plants have drawings, the process is documented. It's not like some Ph.D. dude is mixing up liquids in a test tube. Now granted any money in the world won't have the plant replicated by tomorrow, but enough money will have it replicated much faster. And the process could have been started months ago. Pretty much any vendor will do things faster if you pay them more money and I'm sure this is true for Pfizer's various vendors here. You can order some machine part produced in 3 months or in 3 days (with the latter being like x20 the price, but who cares now).
monkeypizza 6y ago
Yet you presume that the current price is magically the correct one - even though that price is different by region. What would you say the effect of the government offering half as much would be? 1/10th? At one point does economics matter? I think once you admit money matters on the downside, you have a hard time proving that the point it zeros out of effect as it increases exactly matches the arbitrary point it was actually set at.