Recent submissions

Comment · 5 years ago
This article echos the common complaint I've seen from people hesitant to adopt TS.

"I always used to get some compilation errors which were hard to understand initially, and I scratched my head trying to figure out the problem. "

Which is like saying, "I got rid of my carbon monoxide detector because it kept beeping at me"

Comment · 5 years ago
The data wasn't to prove a point. The data was there as an example that you can construe different results from the same data. Given that data, you can make claims of "safety" quoting a lower murder rate, disregarding other factors such as improvements in emergency medical services and treatment of major wounds.

I'm not asserting whether things are safer or not. Im asserting that judging "safer" cannot be accurately done with metrics such as "murder rate".

Comment · 5 years ago
If Adam shoots Bill with the intent to kill him, leaves the scene, and paramedics rush Bill to the hospital, and he lives, Bill is not murdered. Bill was the victim of an aggravated assault

I can't vouch for the reliability of this source[1], but it will serve for my example here. The murder rate in the US was 5 and 5.1 in 1960 and 2019 respectively. The aggravated assault rate was shot from 86 to 250.

Life didn't get safer, we just got better at saving lives.

Then again, this looks like its probably based on UCR data, which is sent to the FBI from local/state law enforcement. I imagine there are certain demographics, especially in 1960s America, that didn't get crimes against them reported with full accuracy.

There are lies, damn lies, and statistics. Especially regarding crime. "Gun violence" can include suicide in some reporting systems, and not in others. Some will put a shooting at a bank robbery in with the same category as a single mother shooting an intruder in her home, and some wont.

My point is that the subject of safety is far more complicated than "See, less people are dead, it's safer". Imagine a special forces team of 20 conducts 100 hostage rescues in a year, gets in a firefight in every one, takes multiple injuries, but no one dies. Now imagine the logistics unit of 20 that performs maintenance, paperwork, administrative functions, etc, at a forward operating base has one of their members crushed by a vending machine.

1 in 20 people in logistics died, 0 in 20 in the special ops team died. Most people would not reasonably argue that the logistics team was more dangerous.

[1]http://www.disastercenter.com/crime/uscrime.htm

Comment · 5 years ago
Just wanna make sure I got this right. "ASS", automatic instant suspension with 0 human review.

"King's Throne: Game of Lust.", with advertised gameplay content such as "I, [the King], shall interrogate this female prisoner privately in my bed chambers", is one of the top advertised games on google play.

Got it.

Comment · 6 years ago
I dont have citations on hand, but my understanding is that this claim is largely attributed to a single/small amount of studies of dubious validity funded by US Beef industry, and that when you look at populations that have much higher soy intake, their estrogen levels are not elevated/are lower than similar US populations.

Grain of salt and all that, just offering a lead for those curious of the counter argument

Comment · 6 years ago
Comment · 6 years ago
Comment · 6 years ago
I'm not from Europe, but if I saw "BBC converts night images to perfect daylight in ~1 second", I would assume it meant the British Broadcasting Corporation. CNN is just as big of a name. His point is absolutely relevant.

Said differently: The percentage of people who are not from the US - but are aware of CNN as the Cable News Network, is higher than the percentage of people who are not machine learning experts - but are aware of CNN as a Convolutional Neural Network

Comment · 6 years ago
I 100% believe every anecdote you just presented. I'm sorry that those things happened, especially when unpunished. As you said, cops are just people. Most people are okay. Some are heros, some are monsters. Weeding out the monsters is notoriously difficult, especially when a profession like law enforcement has definite appeal to them. It also appeals to the heros.

I can give anecdotes about friends who are cops that risked their lives to save strangers, even the bad guys. I know more than 1 cop who arrested a guy, and had the suspects family thank them for how they saved their life in the process.

I can also bring up anecdotes about how people who were so passionate about security were also criminals hiding criminal deeds. I could assume, and assert, that since I've personally seen people use their phones' security to hide evidence of murder, and infant rape, that all people who care about their phones' security are the same.

But that'd be a disservice. It'd be a disservice to those who legitimately care about security for legitimate reasons, because freedom is important, and fragile. It'd be a disservice to others who aren't sold on either side of the discussion. And it'd be a disservice to myself in that it makes me seem very narrow minded and narrow viewed. Its letting fear overcome observation.

There are 700,000 cops in America alone. Undoubtedly some are unqualified garbage. Some are malicious monsters. Some are believe the ends justify the means. Some are paragons of truth and justice. Some aren't malicious, or dumb, but only care about their careers and are shortsighted with all else. We can't make sweeping statements either way, it does nothing to help.

There have undoubtedly been cases of cops using rubber hose decryption. There have been warrants falsified intentionally and unintentionally.

But cops wanting to be able to execute search warrants on phones isn't as simple as "We want more power, more control". There are countless legitimate cases of human trafficking, murder, and sadly worse. We, as a society, have to figure this stuff out. We have to find the balance between "Give us all your secrets" and "Do whatever you want without question".

But we can't have this conversation to find the balance, until we admit that we are on a scale.

Comment · 6 years ago
Do you have a citation that police are as _equally willing_ to beat a password out of someone as they are to use existing tool to image a phone in compliance with a lawfully obtained and executed search warrant?
Comment · 6 years ago
I was a cellebrite certified operator/analyst. The answer is a combination of "The iPhone is that secure" and "Androids usually aren't".

The tools that Cellebrite sells will handle all but the most recent iPhones and maybe the most recent Androids. The newer the iPhone, the less likely a law enforcement officer is gonna have the tools to get into it. Cellebrite offers a service where you can ship them the phone and let their techs go at it. Whether it is because of legitimate limitations in their technology, or purely a marketing/commercial limitation, is unknown.

In general, any non-i{hone is Android. Most Androids are garbage phones that are sold by the dozen in the bottom of a bin at a gas station. Security isn't something its creator or buyer is really concerned about.

I have been out of the field for a few years, so things may have changed slightly in either direction, but I imagine its the same general concept. Newer is harder to break into, brand new iPhone model is probably impossible for now.

Comment · 6 years ago
Comment · 6 years ago
Oh man, I'm glad i'm not the only one. Please let me know if you have a solution. It is really difficult to, in a kind way, tell my generous friends and family, "Please don't buy my kids anything. We have more clothes than we need, more toys than we need. They are just going to play with it today then throw it in front of the already overflowing toy box because we literally have no place to put it"

The closest solution we've gotten is to not do that, but instead use it as an opportunity to talk to the pre-schooler about charity and sharing with the less fortunate, eventually leading him to the "lets give our extra toys we don't play with very much to other kids" thought.

Comment · 6 years ago
Comment · 6 years ago
Proton is pretty great, but there are hardware limitations. Im running a pretty old video card. I can run games like Deep Rock Galactic on Windows, but it wont work on Linux as Proton needs newer linux drivers, which don't support my card, in order to run that game.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton