Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
x220
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
23 ms
·
211.
▲
by
x220
8y ago
The best choice to make when you don't have birth control and don't want to have a child is to not have sex, since even having sex using condoms and similar protection are not completely effective. Are you saying that birth contro
212.
▲
by
x220
8y ago
>Society is generally okay with men having sex outside of marriage. "Okay" is a curious word to choose. There is a good reason why almost every culture on Earth has shamed female promiscuity: women get pregnant and men do not,
213.
▲
by
x220
8y ago
You have control over your own choice to oppress people or not oppress people.
214.
▲
by
x220
8y ago
>The application is specifically meant to track this information So a privacy risk is the cost of doing business. Knowing about it upfront and consenting to the data collection doesn't negate that. If H&R block stores all of the
215.
▲
by
x220
8y ago
>In practice, everyone has more than two levels of trust. That is true, but that doesn't mean that two levels of trust is not a privacy concern merely because everyone does it. >Someday isn't today I felt appalled after read
216.
▲
by
x220
8y ago
I would call it libertarianism if most of the people I've written to seem to actually believe in minarchist government and having rights limited only as much as absolutely necessary. I don't think that's what most people on h
217.
▲
by
x220
8y ago
In 2019 people just don't care as much about censorship or freedom of speech as they used to. Not saying it's right, but that's the way most people think now. I can't remember how many times I've had conversations w
218.
▲
by
x220
8y ago
So if one company has that information it's OK, but once two companies have it it's a privacy concern?
219.
▲
by
x220
8y ago
It looks like the company itself targets the ads, rather than selling the data to another company to target the ads. This shouldn't actually matter. Why do you care if your private information is in the hands of company A or company B?
220.
▲
by
x220
8y ago
>$2.3B every year At this level a couple billion dollars is hardly anything.
221.
▲
by
x220
8y ago
Yeah but don't you think having a raid array but not having a surge protector is missing the forest for the trees?
222.
▲
by
x220
8y ago
>I don’t know what happened for sure, but I think it may have been a power surge that fried the boards on both the Synology and the USB, as they were plugged in to the same socket. He didn't have a surge protector? Sweet Jesus. I do
223.
▲
by
x220
8y ago
You shouldn't bargain with people who are willing to screw you over this blatantly. You don't know if you can trust what they say or their motives.
224.
▲
by
x220
8y ago
>This event was the largest recorded creation of space debris in history with more than 2,000 pieces of trackable size (golf ball size and larger) officially cataloged in the immediate aftermath, and an estimated 150,000 debris particles
225.
▲
by
x220
8y ago
I was thrilled when I took my first microeconomics class in college. Rarely have I ever been so interested in something which I previously thought was boring and useless. I agree with Spolsky when he writes that programmers really should un
226.
▲
by
x220
8y ago
They are trying to screw you over.
227.
▲
by
x220
8y ago
>Cryptocurrency software is a large scale enabler for ransom, kidnapping, and money laundering, bribery, etc All of those were done before cryptocurrency even existed. It's not enabling those things if those things are clearly possi
228.
▲
by
x220
8y ago
>the service that TSA staff do, right now, is to let people continue to be able to fly, without (say) 5 hour lines No, they create the lines. TSA inspections are useless. Third party evaluations (government employees trying to sneak weap
229.
▲
by
x220
8y ago
I'm not sure if that evidence exists or is easy to find. From anecdotal evidence I can say that the people I've seen drinking in America either drink quite rarely (a couple times a week) and far too much at a time (vomiting) or ha
230.
▲
by
x220
8y ago
Not sure why you're being downvoted. You're right. These companies like workers who have no other responsibilities in their lives except their jobs. Edit: single men and exceptions to the rule, please keep downvoting me.
231.
▲
by
x220
8y ago
Why bother? The TSA hasn't demonstrated that it does more good than bad (they don't catch terrorists; they've been caught stealing). I don't pay people who don't do a useful service for me.
232.
▲
by
x220
8y ago
Not to mention that prohibition of a drug always makes the black market drug more concentrated and therefore more dangerous.
233.
▲
by
x220
8y ago
Look up the definition of anarchocapitalism, please, and you'll see it doesn't match the definition of what we are dealing with. Distributed internet also evades the ability of most single nations to censor or regulate it. Does th
234.
▲
by
x220
8y ago
It's a survey from more than a couple dozen drug experts. If citing one expert is a source, then citing a survey of multiple experts is also a legitimate source.
235.
▲
by
x220
8y ago
>It's possible that there's a cumulative effect on people from even tiny amounts of these and other pharmaceuticals in drinking water, but this hasn't been proven. And perhaps vulnerable populations (pregnant women, people
236.
▲
by
x220
8y ago
Here's my source. https://en.wikipedia.org/wiki/File:HarmCausedByDrugsTable.sv...
237.
▲
by
x220
8y ago
The drug scheduling regulations are delusional and do not reflect the actual research potential, medical use, or harm potential of drugs. I'd remind you that marijuana is scheduled more harshly than cocaine, and so are a lot of other r
238.
▲
by
x220
8y ago
This is a good case of statistics being insufficient without proper interpreting. There's a huge difference between the health of someone who has one drink with dinner every day and someone who has seven drinks in one night every Frida
239.
▲
by
x220
8y ago
>So it has been normalized for millennia Alcohol abuse is not normalized in most developed nations. America is a harmful exception.
240.
▲
by
x220
8y ago
Go look up studies done about the Big Five trait model. It's very standardized and reliable between different assessments on the same individual and across cultures.
More ›