That's a bit like a European saying they're avoiding the US beacause of all the spree killings and school shootings. The risk is minuscule for any particular person.
It's sad how much effect these isolated attacks can have. Hundreds of thousands dead from car crashes and smoking and whatever else, nobody cares, it's expected.
A few homemade bombs, and that causes people to change their whole outlook on a country? It doesn't make sense.
I wish there was some way we could not report on terrorist attacks, like we don't report suicides. Sensational news reports just blow things right out of proportion to their actual importance at the country scale.
It's hard to do, and maybe not realistic, but there has to be some solution where these asshole terrorists don't win. And they win as our freedoms are limited as well.
A few crazy people shouldn't be able to hold countries to ransom. I don't know the solution but it just sucks.
Yep. Quite simply, the 0.1% of the US population here on HN does not care about privacy nor protesting nor writing to their local or state government.
HN needs to stop living in a bubble. Only 1/3 of the US even bothered voting in the Trump v Clinton presidential election. I'll let you think about that for a moment. Now think about border searches. Has your phone been searched? Neither has mine. I'll go back to my company catered lunch now.
In our busy and overwhelming lives it is easy to procrastinate about civic responsibilities while still caring about living in a just and fair world.
My phone has not been searched, either, but the issue here is to stand up for your rights. If you have never been wronged by a person or organization in position of authority in US government then consider yourself lucky. I and most people I know living in the US have.
Actually you should count all Adults not just "eligible" for a few reasons
1. No one once reaching the age of majority should have their voting rights taken, for any reason including criminality
2. The figures also do not include people of age who are ineligible to vote or have not registered. The "Have not registered" is a key part because many stated in the last 2 election cycles have been passing a number of laws to disenfranchises segments of the populations making it harder and harder to register
3. The concept of our nation is "Consent of the people", that is all people, not just those voting. the majority of PEOPLE reject the 2 candidates put forth buy the corrupted and unethical political parties then the government can not claim to have the consent of THE PEOPLE...
Many many many Americans feel complete disenfranchised by the political systems that gives them 2 Choices that are equally terrible and corrupt. Forcing a defensive vote that is mainly against the person you do not want to win instead of voting for a qualified person you actually want to be in office.
As a life long Libertarian, I have no interest in voting for either a Republican or Democrat
Border guards were doing this stuff long before Trump was elected, and it was never a significant campaign issue one way or the other. Stop trying to make everything about Trump.
I know egress is one of the more expensive cloud services (e.g. compared to compute and storgage) at AWS, GCP, etc., but if I upload data to my learning system that's ingress AFAIK which is mostly free or less expensive. Btw. current Egress is like 0.1$/GB, so 100G ~ 10$.
Don't get me wrong, I am not saying you should always train in the cloud, but I do not think slower Upload or Ingress are the limiting factor.
K80s and K40s have a K. That's two generations old. We are currently on P and about to be V. If you think K80s are good, you are far far behind the times in machine learning. A single 1080Ti even in a 4U server outperforms 2 K80s.
Dell R720 / R730s, dual GPU (typically with K40m or K80) in there with 1100W dual redundant PSUs and the GPU enablement kit.
We also set our fans to constant 70% min, to keep the airflow good.
They actually work 15% faster on average compared to the Tesla K series (Remember it's a 5 year old card), but they just stop working after a few months, or return inconsistent results for some operations.
Now, we're not doing graphics with them. We have a GPU database called SQream DB - and we depend on the results to be correct. In the end, they didn't make a lot of sense for us to deploy in a production environment, so back to the Tesla series we went.
In a tensorflow job I find it hard to imagine that a memory error would make a jot of difference... wouldn't the error simply be removed by the network training? If the error was in the execution then I imagine that the epoch would fail and it would then just be a matter of restarting.
Major movie theaters and other retail stores in Phoenix, Arizona, accept bitcoin. It may not be 100% mainstream yet, but crypticurrency is going to explode in a big way
While I do agree that crypto-currencies might be a bit overblown, if you're right at 2% if HN readers having them, I'm actually more inclined to believe the OP is right. That's a lot of people.
To me though, the only thing crypto-currencies currently are good for is buying illegal drugs on the internet. There just isn't any reason to use them over regular money the rest of the time.
Still too expensive. Instead of biasing your user studies with big companies that already use your product, you should listen to the greater population of potential users. But hey, that's startup 101 and I don't expect you to know that.
> store them securely in a password manager and use different passwords for each account
You are about the 2% of the tech crowd (i.e, bay area software/data people). The vast majority of engineers do not use a password manager, let alone the entire US populace.
You severely overestimate the amount the average person cares about password security.
We're talking about best practices; I didn't make any claim about how many people use password managers.
The point remains - if you want to follow password best practices and optimize for user safety, don't enforce arbitrary password changes. You're right about ordinary users - we should provide them with fewer opportunities to shoot themselves in the foot. The lower the frequency they have to focus on generating passwords, the better.
True. But its not just the Bay Area. My wife uses LastPass and 2FA and tells others about it at work and she's a mental health counselor, not a software engineer. I guess my influence did have some influence. My parents are also on 2FA. We're in the Atlanta, Georgia, area, not the Bay Area ;-p
id much rather have a reasonable solution for widespread 2 factor authentication than this password mess.
with my bank i have a password, an app on my phone that generates a key and if i perform significant transactions, they call me to confirm before processing it.
the idea that i'm going to use a different password for every stupid site out there that i have an account with is a bit silly. if someone desperately wants to compromise some of them then so be it. hijack my twitter if it makes you feel better. im not going to waste mental energy on securing social media.
"just use a password manager" sounds cute. password managers are compromised, too. password managers are about as trustworthy as the people who operate them. theres no way im handing my passwords for bank accounts over to some random company and for passwords that protect pointless internet nonsense, im not going to use one either because its irrelevant.
you can invoke this whole "password managers are secure" hoohaa. if they ACTUALLY encrypt your passwords properly and ACTUALLY dont save them on their own servers for whatever they want to do with them later, then yes, they probably are secure. but theres no way to be sure that thats the case. Trusting a password manager introduces more uncertainty into your password woes than they will ever make you more secure, if you really think this whole thing through.
the other issue with a password manager is that in theory, they work across platforms. that ends rather abruptly when youre not in a browser and need to enter a password into an app on your phone.
You seem to be operating under the assumption that all password managers are netbased and commercially operated.
This is not the case for PasswordSafe, not the case for the various Keepass implementations out there, and not the case for several other, lesser known projects.
I am a happy, contended and reasonably safe KeepassX user since many years ago. While I do see the point of two factor auth in certain situations, I sincerely hop and pray it never takes off in a mandatory big way, uncalled for annoyance as it is in most cases. My 36 character passwords usually do the job just fine.
> if someone desperately wants to compromise some of them then so be it. hijack my twitter if it makes you feel better. im not going to waste mental energy on securing social media.
This is a bit short sited; if you use the same password for everything and your Twitter account gets broken into, then every site where you have the same email address is potentially also broken into
You don't need to use a web-based password manager.
I use the standalone KeePass app across all my devices, and can paste passwords into apps with no browser access required. I believe KeePass is widely regarded by those who know more than me as cryptographically solid. The only pain point is keeping the databases synchronized across all devices, but it's not that difficult.
Indeed, there are problems of this structure - individuals making stupid choices and shifting the costs onto others - in many areas of society. All of them should be addressed.
And strangely, a lot of people think we should both do that for free (at everyone else's expense) and that we should legalize the drugs that put them there.
I admit I am still a bit torn on the former. I do consider addiction to be a medical condition, and drug users in need of treatment. The question is where personal responsibility should factor in. At some point, if you do drugs, there was a good chance you were dumb enough to start doing them.
The drugs didn't put them there, the abuse of them did.
Should we make sugary drinks illegal, since we spend millions treating people for obesity related diseases? Or should we put speed limiters on cars, because of the cost of treating people injured in high speed accidents?
Let's make an analogy with alcohol. When alcohol isn't illegal, there is a quality supply of it. When you buy a bottle of 40% whisky, you're not getting something that might be 25% methanol due to some criminal doing an improper job with the distilling, because he only cares about money, wiping out his competitors, and evading the cops, with your health being at the bottom of his priorities.
(Which is not to say, of course, that legal alcohol eliminates the public health problem of alcoholism.)
> And strangely, a lot of people think we should both do that for free (at everyone else's expense) and that we should legalize the drugs that put them there.
> And strangely, a lot of people think we should both do that for free (at everyone else's expense) and that we should legalize the drugs that put them there.
Why should we pay money to enforce a legal regime of prohibition that subsidises the most dangerous and ruthless transnational criminal organisations -- to the point that entire fucking countries get destabilised from cartels and US military intervention? Why should my tax dollars pay to enforce criminal laws that turn the crank of the "iron law of prohibition" and cause drug use to be more and more deadly (which we all end up paying for, via taxes and medical costs)? https://en.wikipedia.org/wiki/Iron_law_of_prohibition
Why do people think it's fine to enforce a legal regime that immiserates drug users to the point that they're willing to risk their life (I don't own guns, but lots of USians do) kicking in my door and stealing my old laptop to pawn for some overpriced fentanyl-laced drugs (sold by criminals) that might overdose them in the street? There'd be less laptops stolen out of cars and houses and less people keeling over in the streets and less "gang-related shootings" if addicts could just buy some fucking precisely-dosed pharmaceutical-grade opioids at-cost from a pharmacy.
Didn't we already try this? Murder done by organised crime combined with ineffective/corrupt police and low homicide clearance rates, along with along with severe increases in incarceration rates, increases in overdose deaths from alcohol/drug consumption (caused by higher potency and more impurities) isn't a new syndrome, it also cropped up during the 1920-1933 alcohol prohibition. Ending that prohibition slashed the murder rate. Why can't we do the same thing for drugs?
Where's the "personal responsibility" that prohibitionists need to own up to? Will they take responsibility for how their policies helped trash the US's homicide clearance rate? Is "less solved murders" worth the wondrous results that drug prohibition has given us?
The harm that drug use/manufacture causes is a lot more a function of the relevant policies/laws than the actual chemicals themselves. Indeed, most of the harm from illicit drug use comes from:
1. dose/concentration uncertainty or potential contamination with higher-potency drugs -- like unexpected fentanyl inside heroin
2. contamination with unwanted or dangerous chemicals (from manufacturing / distribution processes) -- such as https://en.wikipedia.org/wiki/MPTP or cutting agents
3. Poly-drug use: combining any two of {opioids, benzos, alcohol} to exploit synergistic effects -- which can easily result in respiratory depression
4. Designer drugs: using a less tested / more dangerous compound because the better tested and safer one is prohibited.
All of these are either caused or significantly worsened by prohibition and can be much more readily avoided without prohibition. Those harms are not inherent to drug use; indeed, people who have adequate access to pharmaceutical-grade opioids can safely use them for decades without any major medical complications.
Here's another data point. I've never been to one of their shows. I don't plan on it. I'll go if I have time. Elephants or no elephants does not contribute to my decision.
