453 karma · joined November 19, 2019
No, GP is averaging out across the whole day. If you get 100W during daylight and 0W during night, you can average to 50W for an entire 24-hour period. GP took 30W as a conservative estimate, which is entirely reasonable.
Estimated flu infections per season: ~25,000,000
Confirmed COVID cases last 12 months: ~22,000,000
Estimated COVID cases using 7x reporting ratio (CDC): ~154,000,000
Estimated flu deaths per season: ~40,000
Confirmed COVID deaths last 12 months: 360,000
So, even a modest estimate of projected deaths also confirms that COVID over 10 times deadlier than the flu in absolute terms.
First, GP is nowhere near "half-right". Most egregious is calculating a CFR which is one fifth of the true value (as best we know with the current data). This is much closer to "absolutely wrong" than "verifiably correct", so no, "half-right" is absolutely a stretch in my book. In terms of absolute severity COVID is roughly 10 times deadlier than the flu, so there is no comparison here. And this is before we even begin to talk about the lasting effects of the disease. Suffice to say that there is no argument in terms of severity; COVID is much, much worse than the flu.
Second, comparing four months at the beginning of the pandemic when the disease was early in its spread vs. four months of endemic flu season (where the flu virus already has a sufficient foothold across a wide swath of the population) is completely erroneous. A much better comparison (though still imperfect) would be to measure across the same time period (December-March) with the same societal mitigations. In the last month, we've seen nearly 80,000 COVID fatalities in the US, so if we want to assume a similar rate over the next 3 months (likely an underestimate given the current upward trend), that's 320,000 deaths. And that's with fairly substantial behavioral restrictions to reduce the spread of the virus. You also neglected to mention that the 2017-2018 flu season was the worst in the last 10 years, where most flu seasons attribute around 40,000 deaths in the U.S. each year. So, even if we decide to ignore the fact that COVID cases are already suppressed due to behavioral mitigations, and me being generous with all of these estimates, at the very least it is four times worse than the worst flu season we've seen in the last 10 years. In reality, COVID unabetted is probably on the order of 10 flu seasons wrapped into one.
Come March we'll have a better estimate of flu deaths for this season, but I wouldn't be surprised if the restrictions see the number of flu fatalities drop below 20k or even 10k. Granting some intellectual honesty, the stats would unfortunately be significantly confounded due to COVID-induced strain on our healthcare system, and the fact that many individuals who would have died from the flu are now dead from COVID instead. But as a loose comparison, I foresee that ~20 times more people will die from COVID this winter than from the flu.
Like you said, policy is a different matter. Obviously it's worth discussing varying isolation strategies so that low-risk populations aren't significantly affected by the lockdown measures, while high-risk groups are isolated and supported. But that is a completely separate discussion. My main concern is with speech that abuses numbers and proliferates a societally-harmful mentality without discussing any of the actual nuances and trade-offs of the issue. As a suggestion for future commenters, if you want to try to compare the harm of lockdown measures versus the harm of the disease, you better make sure that your numbers are accurate before moving on to cost-benefit analysis.
Indoor residential space: 400 square feet per occupant + 200 sf fudge factor
Indoor retail space: 25 sf per-capita + 25 sf margin
Indoor office/misc job space: 200 sf per office worker + 50 sf margin
Indoor schooling/daycare: 150 sf per student + 50 sf margin
Indoor supporting infrastructure: 20 sf per capita
> Hospital requirement is only around 1 sf, so I'm just adding an arbitrary margin to account for rarely-visited indoor locations (e.g. churches, community centers), storage for city vehicles (fire department), waste management, etc.
0-18: 24% ("student age")
19-64: 63% ("working age")
>65: 13% ("retirement age")
Residential + Retail + Office + School + Infra
[600 * 1000] + [50 * 1000] + [250 * (1000 * 0.63)] + [200 * (1000 * 0.24)] + [20 * 1000]
600k (Residential) + 50k (Retail) + 157.5k (Office) + 48k (School) + 20k (Infra)
875.5k square feet per thousand residents
Now, let's consider what a "walkable distance" is for an average person. I'm very happy to walk anywhere within a 10-minute radius (1/2 mile), but can probably live with 20-30 minutes if there is good bike infrastructure. For now let's just discuss the walking case. A 1/2 mile radius is 0.78 square miles, so a perfectly walkable space would contain all of the services I need within 0.78 square miles, or 21.7 million square feet.
How many people can we fit? Filling all of this space with single-story structures would allow us to fit 25,000 people. There are some obvious adjustments we need to make. First, we can realistically build 3 or 4-story structures for residential and office space. Next, we have to adjust for necessary road infrastructure (emergency services, delivery, waste management, etc). New York City dedicates around 50% of its land to roads and open spaces. Using suburban road densities, we can spend 10% of our land on simple 2-way streets. The remaining 90% can be spent on buildings and outdoor space. Let's say we dedicate 30% of the total land to buildings (6.5 million sf), which leaves 60% outdoor space (13 million sf). This is a fairly comfortable estimate based on looking at a map of my college campus. How many people can we fit? Let's assume some densities:
Residential: 4-story buildings
Retail: 1-story (could possibly go 2, but makes delivery logistics harder)
Office: 4-story
School: 2-story
Infra: 1-story
600k/4 + 50k + 157.5k/4 + 48k/2 + 20k = 284k square feet per thousand residents
This means our 60% open, fully-walkable space can support a bit over 23,000 people, for a population density of 29k per square mile. For reference, this is slightly higher than New York City.
So, what happens when we bring parking space back into the mix? A 150-car parking lot requires an acre of space, or about 300 sf of parking lot per car. Imagine your city needs two parking spaces per car (one for home, one for work). Even if each family of 4 owns just one car (this seems like quite an underestimate), the parking requirement for our city of 23,000 people is 11,500 spaces. This amounts to 3.45 million square feet of pavement alone, just for parking! In other words, 0.123 square miles of our 0.47 sq.mi. walkable area is now parking lot. Now, how much extra road do we need to facilitate moving all of these vehicles? I'd estimate we need to dedicate at least as much extra space for driving as we do for parking, so now 0.25 square miles (53%) of our outdoor space is now dedicated to personal vehicles. Sure, Amdahl's law of parking infrastructure allows us to reclaim half of this space back if we decide to build multi-story parking structures or hide them underground, but this is a huge waste of resources considering that our city doesn't even need passenger transport of any kind!
In summary: Using some very conservative estimates and medium density, a walkable circular area of 1/2 mile radius (0.78 square miles) can quite comfortably fit 23,000 people with adequate space for residence, office, retail, schools, and remaining city infra. If we dedicate 10% of the land for minimum-necessary roads (emergency services + transportation of goods and waste), we still have 60% of the land remaining for green open space and walking. If we want to add cars, we easily lose over half of this walkable space to parking and additional road. In all of these estimates, I tried to favor cars as much as possible.
Rant: Like many things that receive attention on popular and social media, agonizing over 5G distracts from issues that cause real provable widespread harm, such as UV radiation and air pollution. We rarely see articles rise to the top when they cover non-controversial points that would do well to be read by everybody, because how much buzz and discussion can you generate from "wear sunscreen", "don't drive while tired", and "get a prostate exam before you turn 50"?
One source was a video of election workers moving boxes around, with a conspiratorial narrative overlaid. This isn't evidence.
Several tweets showing "voting spikes" in favor of Biden as a result of regular counting operations. Nothing wrong here.
The remaining sources discuss "irregularities of expectation", yet none provide any plausible narrative of wrongdoing. The most plausible explanation is that it Democrat-favored voting methods were more streamlined, reliable, and less contested this year compared to years previous. Obviously this nets more votes for Democrats, but nothing practical is stopping a Republican from voting by mail, either. Also, I wouldn't be shocked if the 2020 election was actually more fair than in previous years, due to a large decrease in wrongfully-rejected ballots. I wouldn't be surprised if more Democratic votes were thrown away wrongfully in 2016 than fraudulent Democratic votes were accepted in 2020. If you're concerned about the legitimacy and fairness of U.S. elections, you can't talk about voter fraud while ignoring historical voter disenfranchisement. If someone thinks that mail-in voting was too easy this year, then their gripe is with the judicial systems that permitted this year's election rules.
Conclusion: The linked website doesn't seem to have very high standards of verification, as most items are obviously flimsy or easily debunkable.
I argue that, for most small projects, the net educational value of bringing an idea to fruition in public is far greater than any negative externality that such a project imposes. Imagine a world where every side project by a novice engineer is ruthlessly speared to the point of abandonment. I'm sure your counter-argument is that "trial by fire" is the most effective means of growth. You probably believe that forcing people to give up leads them to consider bigger and better things. This is, by far, not the case. Most people give up, and never come back. You are focusing far too much on the immediate value of the project, and completely dismissing its value as a means of creating a motivated, learned individual who can potentially make huge contributions in the future.
Anyway, if your most salient point is "the author should be more explicit about a support plan", then I don't disagree. But man, you could have presented it in a much more accessible and succinct way. If you really care about people taking your advice to heart, you have to be more kind.
States should allocate electors in proportion to their own popular votes. The +2 electors for each state is already a good mechanism for combatting pure majority rule. At this point, swing state influence is arguably worse for this country than majority rule.
http://archive.fairvote.org/e_college/reform.htm#proportiona...
The "nuclear football" is a completely separate authentication mechanism for verifying a presidential order, and as far as we know uses codes that rotate on a daily basis.
I also probably should have more charitably interpreted GP as "companies in a strong position have a better ability to provide good customer service."
But, point taken.
I agree that Google has a moral duty to take care of the people who choose to put their trust in their platforms.
The rest of this comment is going to be somewhat critical of Google criticism and seemingly apologistic. Please assume positive intent.
I agree that Google certainly could do more to help people on its platform. But this will always be the case until every single customer service issue is solved to the user's satisfaction. The important question that I never see anybody asking or answering is "How much is reasonably enough, and how do we know (subject X) has done enough?"
It all boils down to what we believe the moral obligation of people (and companies, by extension) ought to be. Any extra service that Google provides beyond those required for profit maximization (e.g. customer retention, maintaining reputation, being only slightly better than the second-best option) is purely charity. Then, how do we evaluate when a certain amount of charity is sufficient, let alone necessary? Public sentiment is fickle, and therefore not a great measure. I bet even if Google increased its CSR by 10x, we'd still see communities eagerly latch onto any criticism of the tech giant and shout "you have to do more!" without quantifying what the final demand actually is!
I take the moral position that we as a society certainly can and should do much more to take care of each other. Telling each other to be better is absolutely valuable and I will never argue against that. But I believe we ought to do a much better job of suggesting desirable and realistic end states, instead of perpetuating insubstantial discourse amounting to one of either "They need to do better!" or "Come on, they're doing the best they can."
Admittedly this places me in a non-committal position. I don't have any great suggestions for what the end states of YouTube, Twitter, et al should look like from a content moderation perspective. But I would love to hear input from actual experts in who can offer some reasonable ideas about the most cost-beneficial solutions for society as a whole.
A tRNA molecule at body temperature travels at roughly 10 m/s. Assuming a point-sized tRNA and stationary ribosome of radius 125 * 10^-10 m, the ray casted by the moving tRNA will collide with the ribosome when their centers are within 125 * 10^-10 m of each other. The path of the tRNA sweeps a "collidable" circle of the radius of 125 * 10^-10 m, for a cross-sectional area of 5 * 10^-16 m^2. Multiplied by the tRNA velocity, the tRNA sweeps a volume of 5 * 10^-15 m^3 per second. Constrained inside an ordinary animal cell of volume 10^-15 m^3, the tRNA would have swept the entire volume of the cell five times over in a single second. Obviously the collision path would have significant self-overlap, but at this rate it's quite likely for the two to collide at least once any given second.
Now, consider that this analysis was only for a single ribosome/tRNA pair. A single ribosome will experience this collision rate multiplied by the total number of tRNA in the cell, on the order of thousands to millions. If a ribosome is bombarded by tens of thousands of tRNA in a single second, it's very likely one of those tRNA will (1) be charged with an amino acid, (2) be the correct tRNA for the current 3-nucleotide sequence, and (3) collide specifically with the binding site on the ribosome in the correct orientation. In actuality, a ribosome synthesizes a protein at a rate of ~10 amino acid residues per second.
Any given molecule in the cell will experience millions to billions of collisions per second. The fact that molecules move so fast relative to their size is what allows these reactions to happen on reasonable timescales.
Anybody who is willing to pass (IMHO) bad-faith judgments on well-intentioned users should be also willing to answer these sorts of questions.