Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
marcell
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
20 ms
·
211.
▲
by
marcell
6y ago
Charter schools are accountable to the parents. If the parents do not like the performance of a charter school, they can pull their kids out of it, and put them in their neighborhood public school.
212.
▲
by
marcell
6y ago
What you're describing may be a controlled burn approach under a different name. After a certain point, the virus infects the entire susceptible population, and then it stops because everyone susceptible has died or obtained immunity.
213.
▲
by
marcell
6y ago
Well if you take eradication off the table, then you have to assume the entire susceptible population will eventually get the virus. So I guess, I don't see the benefit of "stomping" the virus down to some small level. It was
214.
▲
by
marcell
6y ago
It's not really clear they are having better results. I don't know what the numbers are out of China, but certainly in Europe they had really big outbreaks in Italy, France, the UK, Spain, and other places as well. Their outcomes
215.
▲
by
marcell
6y ago
I'm not sure if it's worth it or not, but after you end the lockdown, you have the same situation you did at the start of the lockdown. I think there's a huge misconception in many media sources that lockdown will somehow eli
216.
▲
by
marcell
6y ago
> I don't understand the logic around the left's hatred of charters Charter schools are a threat to the public school system, so the public school system will naturally oppose charter schools. If you are a public school without
217.
▲
by
marcell
6y ago
> we didn't follow through with the lockdown nationally to actually stomp the virus This statement implies that there's a way to completely eradicate the virus, eg. if everyone in the world stays locked in their homes for 6 wee
218.
▲
by
marcell
6y ago
Interesting. If the study is indeed correct, and half the previous run-up to ~$20k was due to Tether, then the market has since correct as the price is a little below half that value. Of course there have been a lot of ups/downs since
219.
▲
by
marcell
6y ago
That makes sense, but is there evidence this is actually happening? I think for this chain of logic to be true, there would have to be cases of people attempting to convert Tether into USD, and failing. That did happen with Mt Gox, so it’s
220.
▲
by
marcell
6y ago
> there are only two options: default on the national debt(s) or devaluation I disagree that these are the only two options. There are quite a few other options: (1) Limit deficit spending and continue paying down the current debt (2) Ra
221.
▲
by
marcell
6y ago
I'm just not clear on how Tether usage specifically increases the bitcoin price. My understanding is it's used to transfer money between exchanges, and possibly to fund ICO's. Neither of those really lead to an increased bitc
222.
▲
by
marcell
6y ago
Not OP, but I would say: (1) Tether is largely orthogonal to bitcoin. It's a way to trade USD more quickly over the internet, which isn't really related to bitcoin, besides the underlying technology is similar. (2) I don't kn
223.
▲
by
marcell
6y ago
> If five years from now everyone in the world has become a bitcoin user, that could only be because something very very bad has happened to the regular monetary system If you ask a bitcoin “true believer,” many will tell you that someth
224.
▲
by
marcell
6y ago
Mainly, I'm surprised this was even allowed. My understanding is that if you're not a Han Chinese, you cannot really do business in China, that you're basically swimming against both cultural exclusion, and government exclusi
225.
▲
by
marcell
6y ago
Did you move there as a non-Chinese person? If yes, is that hard to do? Any challenges you've faced?
226.
▲
by
marcell
6y ago
You may find this analysis interesting as it looks at that exact question: https://unherd.com/2020/06/karl-friston-up-to-80-not-even-su...
227.
▲
by
marcell
6y ago
I working on a project called CloudPy that's a 1-line drop in for pandas to process 100 GB+ dataframes. It works by proxying the dataframe to a remote server, which can have a lot more memory than you're local server. The project
228.
▲
by
marcell
6y ago
Strong coder with 10+ years experience on all parts of the stack. Former startup founder and worked at lots of tech companies big and small. Email is in my profile, let’s get in touch. I'm in the SF Bay Area.
229.
▲
Ask HN: Who's Looking for a Co-Founder?
11 points
by
marcell
6y ago
|
4 comments
230.
▲
by
marcell
6y ago
Data shows the peak has passed. You can monitor ILI deaths (which includes covid) in the weekly flu report from the CDC: https://www.cdc.gov/flu/weekly/index.htm . The peak of deaths was a few weeks ago, which you
231.
▲
by
marcell
6y ago
Fwiw Instacart had the same problem, and it succeeded, even after Amazon acquired Whole Foods. It's really case-by-case, and depends on so many factors.
232.
▲
by
marcell
6y ago
That's an interesting and valid point. First, a disagreement. Your base assumption that police activity should scale linearly with population. This is just obviously not correct. There are huge, huge differences in countries and crime
233.
▲
by
marcell
6y ago
Sure, I think that's fair enough. I didn't mean to imply that there are 250,000 cases a year where a heart transplant is performed to cure pneumonia. But of those 250,000, some will be small mistake, and some will be epic fuck ups
234.
▲
by
marcell
6y ago
Huh? I didn't say racism isn't real, and I also didn't say cops aren't tear gassing reporters in the face. GP questioned how a profession can exist when it makes so many deadly mistakes. I pointed out that there is anoth
235.
▲
by
marcell
6y ago
There are lots and lots of mishaps in many professions, sometimes with deadly consequences. The United States is a country of 330 million people. Suppose an event is so unlikely it only has 1 in a billion chance of happening--it will happen
236.
▲
by
marcell
6y ago
> now best estimates are a 1-1.25% IFR Just FYI, the CDC gives their best estimate at 0.26% IFR. (source: https://reason.com/2020/05/24/the-cdcs-new-best-estimate-imp... , note that 0.26% adds in the estim
237.
▲
by
marcell
6y ago
I was very worried about Covid-19 when the initial data and charts were coming out of China and Italy. It seemed like it was going to quickly overwhelm all the hospitals in America. Since then, I've become much less scared about the im
238.
▲
by
marcell
6y ago
I mean... it's just not the responsibility of cities to redistribute income. In addition to people moving in, you have people paying high taxes who will move out to neighboring cities without this tax. I don't want to pay a 5% wea
239.
▲
by
marcell
6y ago
Cities don't really have budget to give out cash. If you took the entire police budget (about $50M/yr), you'd only have enough money for $10k/yr for 5,000 individuals. If you have a $10k/year rent subsidy available
240.
▲
by
marcell
6y ago
Which state are you in? Curves are exponential until they aren't. In most states, hospitals were never anywhere near being "overwhelmed", quite the opposite in fact. In my county (Santa Clara, CA), there are under 5% hospital
More ›