Not relevant. My laptop has 128gb but isn't the device that takes hundreds of high trees photos and videos each month.
When it comes time for me to download a year's worth of photos to process them, pick favorites, make a holiday card - I can't do it. Not enough storage. I have to page in only a couple of months at a time.
My phone also has 128gb. It doesn't have a big operating system, nor Office, nor Eclipse, not my source tree and all the binaries. Pretty much nothing but photos and videos. It ran out of space too.
You don't need to spend a few million on a study to provide data proving that texting, playing games, watching silly videos, and posting to social media all at the same time makes it hard to concentrate in a class room.
You also don't need to spend a few million to prove that a phone that's not used at all during class has no detrimental effect on concentration.
Now that we have the straw men out the way, the questions are (1) what outcomes do we wish to improve? (2) is a blanket ban, a blanket acceptance or some other more nuanced rule on cellphones the most effective way to achieve it?
Because robo-calls are illegal, and if you could identify the companies responsible you or law enforcement could go after them. Having the ability to change your number makes it exponentially harder to track down the origin of the call.
I don't think I follow that...
The telcos already know the companies responsible for each call, but I guess they're not able to classify whether each call is spam or not. Also it's not in their interests to pursue damages.
We end-users are already able to more-or-less the companies responsible for each robocall, because each robocall is selling us something, and we can just see what they're trying to sell us. But perhaps this chain of reasoning isn't robust enough for a lawsuit.
What you're describing is that robocalls will be stopped by empowering end-users to take it upon themselves to launch class-action lawsuits. Except that lawsuits require us to show we've suffered damages, and I think it'd be hard to argue that robocalls have much monetary damage.
What you said is that robocalls are *illegal*. That's a different matter, one for the government to prosecute. I can't see any mechanism by which "end-users knowing the originator of the robocall" would contribute to either (1) police or state prosecutors being more able to assess which robocall companies to pursue, (2) them being more incentivized to pursue them.
Maybe a different approach is to prevent caller-id-spoofing and then pass a new law saying "the damage caused by each spam call is deemed to be $100". That'd allow class-action-lawsuits to go ahead. But I bet this kind of law wouldn't be legal...
Not being in you contact list is something you can fix. It's what I do. If they're not on my list my phone goes direct to voicemail. They can leave a message. If it's from someone I do business with and expect to talk to again I add it to my contacts. It take literally 30 seconds. Bonus: that means when they call I know who they are before I pick up. If they won't leave a message that's on them. I've never had a legitimate business contact refuse to leave a message.
I'm still not getting it. Let's suppose "no-callerID-spoofing" goes through. In that future, when you refinance your mortgage then you'll get one call from Bill in accounting, another from Jill in brokereage, another from Dill in risk assessment, and they'll all have different numbers (rather than having been spoofed to look like they all just come from the main central InsuranceCo number). So for the first month or two you'll be playing each of your voicemails to check if it came from someone in the new business you did. Still 90% of those voicemails will be from robocallers.
Or you get a call from your kid's school. Is it your teacher calling? the school nurse? the counsellor? the deputy head? there's no way you're going to anticipate all those numbers. Some of them might be real emergencies that you should respond to immediately, rather than leaving to voicemail.
In addition to still having to listen to your voicemails, you've also got the extra administrative burden of entering each of those numbers into your contact list.
(not that it would even help me... I almost never speak live to business when they call me; I always call back at my convenience. So that work of adding into the contact list wouldn't even buy anything. But I understand it would buy something for people who do want to respond to businesses live).
Bad headline: "Schools Are Locking Students' Phones Away to Help With Concentration"
(1) Only one student reported that it helped with concentration. There was no indication that it actually does help with concentration.
(2) There was no indication that "helping with concentration" was the motive for the school doing it
(3) The article only mentioned one school; I don't know where the headline got the plural "schools".
(That said, I personally believe that phones are bad for concentration, and indeed my children's school also bans phones. I just want to see actual defensible data. Not a dumb article designed as click-bait to reinforce some people's prejudices and raise the ire of those who disagree with it.)
Hardly.....isn't it fully within the capabilities of the telecom companies to stop third-party caller ID spoofing?
How would that even help?
I get a phone call. I don't know if the number is from my dentist, my bank, my credit card, my doctor, my travel agent, my insurer, my mortgage lender, my employer, my employer's IT/security department, my dry-cleaner, a seller on ebay. Many of these legitimate calls are from out of state. I don't have their numbers in my contact list.
What difference would it make if they were forbidden from using caller-ID-spoofing? I'd still see an unrecognized number. I'd still have to answer it in case it's a message from one of these institutions.
As I continue to develop my new data management system in that archaic language C++. I am one of those Luddites who believes that 'scripts' are for doing once-in-a-while tasks that need to be written quickly or updated often. Real programs are written in Assembly or the next best thing...C or C++.
Continue to develop? Hey, maybe if you'd used Python or Java then you'd have finished your development already!
Isn't this the oldest argument? The less appealing it is to purchase something, the more likely pirating becomes.
I'm glad your conclusion was obvious to you, because the study actually says something completely different what you've characterized.
Study: "if you make it less appealing to pirate, then people will buy the product less." Your characterization: "if you make it less appealing to purchase, then people will pirate more."
The only tenuous thread by which I can relate your characterization to the study is if you assume that "pirate more implies buy less". If so, then your characterization is even more unrelated to the study....
Study: "if you make it less appealing to pirate, then people will buy the product less." Your characterization: "if you make it less appealing to purchase, then people will buy the product less."
Now the study and your characterization are a bit more related because of the causal mechanism in the study. They suggested that if you make it less appealing to pirate, then companies will find themselves unable to refrain from raising prices beyond the point where it's sensible. This makes the study a bit closer to your characterization:
Study: "if you make it less appealing to pirate, the it will become less appealing to purchase, and then people will buy the product less." Your characterization: "if you make it less appealing to purchase, then people will pirate more."
At last we've unified your characterization with the study. But we did so only by ignoring the entire point and chief finding of the study, namely "piracy often acts as a form of invisible competition". Which is non-obvious, and yes more or less new (or at least not widely understood), and yes it does need studies to back it up.
Q1. "Can we get someone to Mars before the end of my first term with infinite money?" Q2. "With infinite money, how soon could we get someone to mars?"
They're subtly different but one question is an intelligent question that identifies bottlenecks. The other is a vanity request.
I agree that they're subtly different. The first one has a concrete goal, one that's not achievable but also still admits reasoned answer as to why it's not achievable. Therefore it's more effective in identifying bottlenecks. The second one seems more likely to lead to answers that use more money than necessary. But I'm not understanding how Q2 would be a vanity request?
(I'm being obtuse. You of course meant to imply that Q1 is the worse request. I'm disagreeing with your assessment of the questions.)
Yes, Breitbart has a right-leaning bias, but it's like an antidote to the main stream media's false and biased reporting.
An "antidote" is a medicine you give to counteract a poison.
What you describe is countering one poison (what you call MSM's biased reporting) with another (what you call Breitbart's bias). That's not an antidote. I don't think there's any case of using one poison to cure another poison, other than homeopathy. In the absence of antidotes, I think the only thing we have us dilution -- i.e. counter biased reporting by clinging to news sources that are as unbiased as you can find.
You would be surprised by what lawyers can do with subjective terms like "insubstantial".
That's fair enough and seems likely. Maybe we should interpret this news story as instead saying: "Google views the current 'substantial' wording of the law as being too vague, and it makes them unable to assess their risk exposure in the EU. Their negotiators have been making this point to the EU legislators behind the scenes but that's a difficult process. They decided it was time to bring out their big negotiating guns by having a high-level publicly visible statement about turning off all of Google News in Europe, to make the legislators sit up and pay more attention."
I presume the fuss is over the supposed "hyper-link tax" which has to be one the most idiotic ideas I've come across in my adult life.
Why presume when you can actually find out? The article says:
https://www.bloomberg.com/news... Google is considering pulling its Google News service from Europe as regulators work toward a controversial copyright law. The European Union’s Copyright Directive will give publishers the right to demand money from the Alphabet Inc. unit, Facebook Inc. and other web platforms when fragments of their articles show up in news search results, or are shared by users.
The wikipedia page for the EU Copyright Directive explains:
The proposal [includes...] exemptions for either copying an "insubstantial" part of a work... The version of the directive voted on by European Parliament Committee on Legal Affairs contained explicit exemptions for the act of hyperlinking and "legitimate private and non-commercial use of press publications by individual users"
So it looks like this is specifically *not* about a "hyperlink tax", and either Google specifically wants to be copying substantial parts of a copyright work without paying the owners, or something more subtle is going on (and hence we can expect to see simplifications, distortions, and clickbait designed to inflame responses).
Tax fuel (gasoline, diesel, and/or electricity) at appropriate levels and you accomplish the sensible goal of taxing in close accordance with utilization. The bigger the vehicle and the more someone drives the more fuel they will use.
That's incorrect. Road damage increases as the fourth power of vehicle weight according to the US General Accountability Office https://www.denenapoints.com/r...
An 18-wheeler (~6mpg) causes as much road damage as 10,000 cars (~25mpg). The 18-wheeler is paying 4x as much tax but causing 10,000x as much damage. That is not "close accordance" at all.
Now it depends what you mean by "utilization". If you're referring to overall US expenditure to ensure a reliable oil supply then sure, that's directly proportional to the amount of gas you use. But if you're referring to all-up expenditure including roads and infrastructure then it's not.
There are sooo many Android apps that look nice - and free - at first, but then want to access every nook and cranny of your Android device, including the ability to look through your contacts directory and listen in/report on any phonecalls or other communications you perform with the device. My guess is that some of these apps are actually made by state-actors who want to eavesdrop on unsuspecting smartphone users all over the world.
My guess is just that the developer wants to make money from ads, and incorporated an ad SDK from a third party without thinking. And the ad broker who wrote that SDK obviously wants to scrape as much information as possible from the device so they can (1) target ads more precisely, (2) sell the data.
(I base this on having seen how universal it is to consume third-party SDKs without even thinking about how the SDK works...)
The Free Market has spoken. It doesn't like the finances of nuclear power. It considers it too risky, too long-term. (It does however like the finances of wind and solar).
That's fine! There are many things we do (such as nationalized health care and military defense) which the free market is bad at. Nuclear is another one. We should just be explicit that it will mean governments spending large amounts of taxpayer money to push it through.
There's not even standardized definitions of AI internals yet so I'm not sure how you write regulations that could easily be understood by those implementing the AI to begin with.
The concern is with ML which I think we can come up with a decent working definition. Here's my first attempt after 30 seconds "a decision procedure determined by a weighting system where the weights have been derived through a training data set and where a human is unable to predict/justify/explain the particular values that were assigned to the majority of those weights".
The concern is that for whatever reason - inexpressiveness of the neural network, or inadequate selection of the training set - the decision procedure ends up with decisions that are reasonable to it, but which a human when faced with this particular decision would evaluate that it violated some basic human right or similar.
I'd start by regulating it in specific areas, e.g. "if a person has been denied a loan, they are entitled to a written explanation for the grounds/criteria that went into that decision". The answer "our neural net said so" isn't an explanation and doesn't provide grounds/criteria. The answer "your credit score is below 500" would be fine. The answer "our neural net said so, and a human reviewed it, and based on the following reasons supports the neural net's answer" would also be fine.
WITH ALL DUE RESPECT: I think you're looking at this from the perspective of someone who has always had a roof over their head, has never gone hungry or thirsty a day in your life, has always had medical attention when you needed it, and has never had their life threatened by anyone for any reason, therefore you're having a problem imagining the places and people I'm talking about.
I did work for a year as a volunteer in the remote hilltowns of northern India so I have a slightly broader perspective. I'm thinking of existing proven benefits to (1) access micro-credit, (2) get market prices on cellphones in advance of walking a day to market, (3) transfer small amounts of money to friends and relatives from the wage-earner in the big town.
I think that what folks need to get out of grinding poverty is economic empowerment, and that's the only long-term means. Anything else - food, reliable water, reliable sanitation - will add up to little in comparison, because they'll vanish as readily as they came.
I think perhaps we should work more on 'universal access' to clean water, enough food to eat, and safe countries to live in for everyone, before we worry about 'universal internet access'. It's kinda hard to enjoy watching the box-centric antics of Maru on YouTube when you're dying of dehydration, malnutrition, or the local Warlord or Druglord is kicking in the door of your shack to steal your children, kill you, or both. Google, Facebook, and whoever else, can just wait their turn to monetize the rest of the 7,000,000,000 on this planet whose personal information they haven't been able to monetize yet.
Your reasons seem fair. But the opposite reason is that internet access is what will *enable* folks to do better - better market prices for their produce, better job opportunities, better oversight of their leaders, better community organization.
We already know that now, as according to the DEA's 2018 report, the most common way for drugs to enter the country in the south is via the points of entry already. After that is tunnels, light aircraft, and then marine vehicles.
Mexican TCOs transport the majority of illicit drugs into the United States across the SWB using a wide array of smuggling techniques. The most common method employed by these TCOs involves transporting illicit drugs through U.S. POEs in passenger vehicles with concealed compartments or commingled with legitimate goods on tractor trailers
[It specifically refers to "Ports of Entry", POEs, a more precise term than "point of entry". Ports of entry are the existing staffed border stations, that obviously a wall will do nothing to improve.]
Other cross-border smuggling techniques employed by Mexican TCOs include the use of subterranean tunnels... Mexican TCOs also transport illicit drugs to the United States aboard commercial cargo trains and passenger buses. To a lesser extent, Mexican TCOs use maritime vessels off the coast of California. Mexican TCOs also rely on traditional drug smuggling methods, such as the use of backpackers, or “mules,” on clandestine land trails to cross remote areas of the SWB into the United States. Mexican TCOs exploit various aerial methods to transport illicit drugs across the SWB. These methods include the use of ultralight aircraft and unmanned aerial systems (UASs) and drones to conduct air drops. Ultralights are primarily used to transport marijuana shipments, depositing the drugs in close proximity to the SWB. Currently, UASs can only convey small multi-kilogram amounts of illicit drugs at a time and are therefore not commonly used, though there is potential for increased growth and use. Mexican TCOs also use UASs to monitor the activity of U.S. law enforcement along the SWB to identify cross-border vulnerabilities.
The only one of these that the wall would stop are "traditional drug smuggling methods, such as the use of backpackers, or “mules,” on clandestine land trails". I'm curious what percentage this makes up.
Ok, still trying to understand. What law did they break. I'm by no means trying justifying what they did, but trying to have a conversation.
The law they broke is the UK Data Protection Act. Read the Guardian article. It's all there. Even if, as you say, Cambridge Analytica obeyed the Facebook terms of service, nevertheless...
(1) Cambridge Analytica held data on people; (2) Under UK law at the time - the "Data Protection Act" - if a company holds data on you, then the company is obliged to divulge that data to you personally upon your request; (3) someone did request their data and Cambridge Analytica ignored the request; (4) the UK's Information Commissioner's Office - which has authority under UK law relating to this part of the Data Protection Act - issued an enforcement notice which in the UK has weight of law, requiring Cambridge Analytica to comply with the request; (5) Cambridge Analytica declined the comply with the request arguing that it didn't have to; (6) a law court found that Cambridge Analytica did indeed have obey the enforcement notice, and had broken the law by not doing so.
Summary: the law they broke is the UK Data Protection Act.
They're either buying into the same marketing and media hype for the half-assed excuse for AI everyone keeps trotting out, or they've got something nobody else has, meaning general AI. The latter is highly unlikely, if they did we wouldn't be hearing about it at all.
Or the option you didn't mention -- they know that existing ML-based systems are really good at pattern-matching, and that many forms of pattern-matching don't require full general AI in order to be cost effective, and they're applying it (like the headline and article says) to another domain for pattern-matching.
if the economy only produces hotdogs and buns, automating hotdog manufacture gets you producing 33% more hotdogs with 33% less labor.
That's fair. But here's another example. If the service economy has fast-food-cashiers and trash cleaners, then automating fast-food-cashiers won't do anything. That's because the rate of service is limited by the rate of customers, which is already being met fine.
The solution to this is simple. Fines should not go into the government's general fund. They are punitive penalties for misbehavior against society. They should be put into a separate fund. Then on April 10 when you file you income taxes, the fund should be divided by the number of taxpayers, with each taxpayer receiving an equal share as a generic tax credit. In that way it's paid out directly to the people who were harmed by the initial violation. The government doesn't see a dime of it (giving it as a tax credit just reduces the need to send a check to every taxpayer), so the government's incentive for things like red lights is to time them to maximize safety, rather than to maximize revenue.
How are you going to "save the environment" by providing tax breaks for metal boxes so rich people have easier time commuting to work? It doesn't make a dent in anything. If you want to "save the environment" you shouldn't be driving around in ANY CAR.
It'd make a huge dent to switch over a country to electric vehicles 100%. And the only question is: what is the most cost-effective way for a society to achieve this transition?
I think the current model used in Norway and US seems like a really good way for society to leverage the free market - provide small targeted incentives which yes do benefit rich people because their real goal is to encourage innovation and this is the best place for leverage. In this case, we've seen that the benefits really do trickle down to all of society.
Slashdot Mirror
Not relevant. My laptop has 128gb but isn't the device that takes hundreds of high trees photos and videos each month.
When it comes time for me to download a year's worth of photos to process them, pick favorites, make a holiday card - I can't do it. Not enough storage. I have to page in only a couple of months at a time.
My phone also has 128gb. It doesn't have a big operating system, nor Office, nor Eclipse, not my source tree and all the binaries. Pretty much nothing but photos and videos. It ran out of space too.
I think bigger storage would be huge.
You don't need to spend a few million on a study to provide data proving that texting, playing games, watching silly videos, and posting to social media all at the same time makes it hard to concentrate in a class room.
You also don't need to spend a few million to prove that a phone that's not used at all during class has no detrimental effect on concentration.
Now that we have the straw men out the way, the questions are (1) what outcomes do we wish to improve? (2) is a blanket ban, a blanket acceptance or some other more nuanced rule on cellphones the most effective way to achieve it?
Because robo-calls are illegal, and if you could identify the companies responsible you or law enforcement could go after them. Having the ability to change your number makes it exponentially harder to track down the origin of the call.
I don't think I follow that...
The telcos already know the companies responsible for each call, but I guess they're not able to classify whether each call is spam or not. Also it's not in their interests to pursue damages.
We end-users are already able to more-or-less the companies responsible for each robocall, because each robocall is selling us something, and we can just see what they're trying to sell us. But perhaps this chain of reasoning isn't robust enough for a lawsuit.
What you're describing is that robocalls will be stopped by empowering end-users to take it upon themselves to launch class-action lawsuits. Except that lawsuits require us to show we've suffered damages, and I think it'd be hard to argue that robocalls have much monetary damage.
What you said is that robocalls are *illegal*. That's a different matter, one for the government to prosecute. I can't see any mechanism by which "end-users knowing the originator of the robocall" would contribute to either (1) police or state prosecutors being more able to assess which robocall companies to pursue, (2) them being more incentivized to pursue them.
Maybe a different approach is to prevent caller-id-spoofing and then pass a new law saying "the damage caused by each spam call is deemed to be $100". That'd allow class-action-lawsuits to go ahead. But I bet this kind of law wouldn't be legal...
Not being in you contact list is something you can fix. It's what I do. If they're not on my list my phone goes direct to voicemail. They can leave a message. If it's from someone I do business with and expect to talk to again I add it to my contacts. It take literally 30 seconds. Bonus: that means when they call I know who they are before I pick up. If they won't leave a message that's on them. I've never had a legitimate business contact refuse to leave a message.
I'm still not getting it. Let's suppose "no-callerID-spoofing" goes through. In that future, when you refinance your mortgage then you'll get one call from Bill in accounting, another from Jill in brokereage, another from Dill in risk assessment, and they'll all have different numbers (rather than having been spoofed to look like they all just come from the main central InsuranceCo number). So for the first month or two you'll be playing each of your voicemails to check if it came from someone in the new business you did. Still 90% of those voicemails will be from robocallers.
Or you get a call from your kid's school. Is it your teacher calling? the school nurse? the counsellor? the deputy head? there's no way you're going to anticipate all those numbers. Some of them might be real emergencies that you should respond to immediately, rather than leaving to voicemail.
In addition to still having to listen to your voicemails, you've also got the extra administrative burden of entering each of those numbers into your contact list.
(not that it would even help me... I almost never speak live to business when they call me; I always call back at my convenience. So that work of adding into the contact list wouldn't even buy anything. But I understand it would buy something for people who do want to respond to businesses live).
Bad headline: "Schools Are Locking Students' Phones Away to Help With Concentration"
(1) Only one student reported that it helped with concentration. There was no indication that it actually does help with concentration.
(2) There was no indication that "helping with concentration" was the motive for the school doing it
(3) The article only mentioned one school; I don't know where the headline got the plural "schools".
(That said, I personally believe that phones are bad for concentration, and indeed my children's school also bans phones. I just want to see actual defensible data. Not a dumb article designed as click-bait to reinforce some people's prejudices and raise the ire of those who disagree with it.)
Hardly.....isn't it fully within the capabilities of the telecom companies to stop third-party caller ID spoofing?
How would that even help?
I get a phone call. I don't know if the number is from my dentist, my bank, my credit card, my doctor, my travel agent, my insurer, my mortgage lender, my employer, my employer's IT/security department, my dry-cleaner, a seller on ebay. Many of these legitimate calls are from out of state. I don't have their numbers in my contact list.
What difference would it make if they were forbidden from using caller-ID-spoofing? I'd still see an unrecognized number. I'd still have to answer it in case it's a message from one of these institutions.
As I continue to develop my new data management system in that archaic language C++. I am one of those Luddites who believes that 'scripts' are for doing once-in-a-while tasks that need to be written quickly or updated often. Real programs are written in Assembly or the next best thing...C or C++.
Continue to develop? Hey, maybe if you'd used Python or Java then you'd have finished your development already!
Isn't this the oldest argument? The less appealing it is to purchase something, the more likely pirating becomes.
I'm glad your conclusion was obvious to you, because the study actually says something completely different what you've characterized.
Study: "if you make it less appealing to pirate, then people will buy the product less."
Your characterization: "if you make it less appealing to purchase, then people will pirate more."
The only tenuous thread by which I can relate your characterization to the study is if you assume that "pirate more implies buy less". If so, then your characterization is even more unrelated to the study....
Study: "if you make it less appealing to pirate, then people will buy the product less."
Your characterization: "if you make it less appealing to purchase, then people will buy the product less."
Now the study and your characterization are a bit more related because of the causal mechanism in the study. They suggested that if you make it less appealing to pirate, then companies will find themselves unable to refrain from raising prices beyond the point where it's sensible. This makes the study a bit closer to your characterization:
Study: "if you make it less appealing to pirate, the it will become less appealing to purchase, and then people will buy the product less."
Your characterization: "if you make it less appealing to purchase, then people will pirate more."
At last we've unified your characterization with the study. But we did so only by ignoring the entire point and chief finding of the study, namely "piracy often acts as a form of invisible competition". Which is non-obvious, and yes more or less new (or at least not widely understood), and yes it does need studies to back it up.
Q1. "Can we get someone to Mars before the end of my first term with infinite money?"
Q2. "With infinite money, how soon could we get someone to mars?"
They're subtly different but one question is an intelligent question that identifies bottlenecks. The other is a vanity request.
I agree that they're subtly different. The first one has a concrete goal, one that's not achievable but also still admits reasoned answer as to why it's not achievable. Therefore it's more effective in identifying bottlenecks. The second one seems more likely to lead to answers that use more money than necessary. But I'm not understanding how Q2 would be a vanity request?
(I'm being obtuse. You of course meant to imply that Q1 is the worse request. I'm disagreeing with your assessment of the questions.)
Yes, Breitbart has a right-leaning bias, but it's like an antidote to the main stream media's false and biased reporting.
An "antidote" is a medicine you give to counteract a poison.
What you describe is countering one poison (what you call MSM's biased reporting) with another (what you call Breitbart's bias). That's not an antidote. I don't think there's any case of using one poison to cure another poison, other than homeopathy. In the absence of antidotes, I think the only thing we have us dilution -- i.e. counter biased reporting by clinging to news sources that are as unbiased as you can find.
You would be surprised by what lawyers can do with subjective terms like "insubstantial".
That's fair enough and seems likely. Maybe we should interpret this news story as instead saying: "Google views the current 'substantial' wording of the law as being too vague, and it makes them unable to assess their risk exposure in the EU. Their negotiators have been making this point to the EU legislators behind the scenes but that's a difficult process. They decided it was time to bring out their big negotiating guns by having a high-level publicly visible statement about turning off all of Google News in Europe, to make the legislators sit up and pay more attention."
I presume the fuss is over the supposed "hyper-link tax" which has to be one the most idiotic ideas I've come across in my adult life.
Why presume when you can actually find out? The article says:
https://www.bloomberg.com/news...
Google is considering pulling its Google News service from Europe as regulators work toward a controversial copyright law. The European Union’s Copyright Directive will give publishers the right to demand money from the Alphabet Inc. unit, Facebook Inc. and other web platforms when fragments of their articles show up in news search results, or are shared by users.
The wikipedia page for the EU Copyright Directive explains:
https://en.wikipedia.org/wiki/...
The proposal [includes...] exemptions for either copying an "insubstantial" part of a work ... The version of the directive voted on by European Parliament Committee on Legal Affairs contained explicit exemptions for the act of hyperlinking and "legitimate private and non-commercial use of press publications by individual users"
So it looks like this is specifically *not* about a "hyperlink tax", and either Google specifically wants to be copying substantial parts of a copyright work without paying the owners, or something more subtle is going on (and hence we can expect to see simplifications, distortions, and clickbait designed to inflame responses).
Tax fuel (gasoline, diesel, and/or electricity) at appropriate levels and you accomplish the sensible goal of taxing in close accordance with utilization. The bigger the vehicle and the more someone drives the more fuel they will use.
That's incorrect. Road damage increases as the fourth power of vehicle weight according to the US General Accountability Office
https://www.denenapoints.com/r...
An 18-wheeler (~6mpg) causes as much road damage as 10,000 cars (~25mpg). The 18-wheeler is paying 4x as much tax but causing 10,000x as much damage. That is not "close accordance" at all.
Now it depends what you mean by "utilization". If you're referring to overall US expenditure to ensure a reliable oil supply then sure, that's directly proportional to the amount of gas you use. But if you're referring to all-up expenditure including roads and infrastructure then it's not.
There are sooo many Android apps that look nice - and free - at first, but then want to access every nook and cranny of your Android device, including the ability to look through your contacts directory and listen in/report on any phonecalls or other communications you perform with the device. My guess is that some of these apps are actually made by state-actors who want to eavesdrop on unsuspecting smartphone users all over the world.
My guess is just that the developer wants to make money from ads, and incorporated an ad SDK from a third party without thinking. And the ad broker who wrote that SDK obviously wants to scrape as much information as possible from the device so they can (1) target ads more precisely, (2) sell the data.
(I base this on having seen how universal it is to consume third-party SDKs without even thinking about how the SDK works...)
The Free Market has spoken. It doesn't like the finances of nuclear power. It considers it too risky, too long-term. (It does however like the finances of wind and solar).
That's fine! There are many things we do (such as nationalized health care and military defense) which the free market is bad at. Nuclear is another one. We should just be explicit that it will mean governments spending large amounts of taxpayer money to push it through.
For the people who check their phone 300 times a day, biometry saves them over half an hour a day. That seems a very reasonable tradeoff.
There's not even standardized definitions of AI internals yet so I'm not sure how you write regulations that could easily be understood by those implementing the AI to begin with.
The concern is with ML which I think we can come up with a decent working definition. Here's my first attempt after 30 seconds "a decision procedure determined by a weighting system where the weights have been derived through a training data set and where a human is unable to predict/justify/explain the particular values that were assigned to the majority of those weights".
The concern is that for whatever reason - inexpressiveness of the neural network, or inadequate selection of the training set - the decision procedure ends up with decisions that are reasonable to it, but which a human when faced with this particular decision would evaluate that it violated some basic human right or similar.
I'd start by regulating it in specific areas, e.g. "if a person has been denied a loan, they are entitled to a written explanation for the grounds/criteria that went into that decision". The answer "our neural net said so" isn't an explanation and doesn't provide grounds/criteria. The answer "your credit score is below 500" would be fine. The answer "our neural net said so, and a human reviewed it, and based on the following reasons supports the neural net's answer" would also be fine.
WITH ALL DUE RESPECT: I think you're looking at this from the perspective of someone who has always had a roof over their head, has never gone hungry or thirsty a day in your life, has always had medical attention when you needed it, and has never had their life threatened by anyone for any reason, therefore you're having a problem imagining the places and people I'm talking about.
I did work for a year as a volunteer in the remote hilltowns of northern India so I have a slightly broader perspective. I'm thinking of existing proven benefits to (1) access micro-credit, (2) get market prices on cellphones in advance of walking a day to market, (3) transfer small amounts of money to friends and relatives from the wage-earner in the big town.
I think that what folks need to get out of grinding poverty is economic empowerment, and that's the only long-term means. Anything else - food, reliable water, reliable sanitation - will add up to little in comparison, because they'll vanish as readily as they came.
I think perhaps we should work more on 'universal access' to clean water, enough food to eat, and safe countries to live in for everyone, before we worry about 'universal internet access'. It's kinda hard to enjoy watching the box-centric antics of Maru on YouTube when you're dying of dehydration, malnutrition, or the local Warlord or Druglord is kicking in the door of your shack to steal your children, kill you, or both. Google, Facebook, and whoever else, can just wait their turn to monetize the rest of the 7,000,000,000 on this planet whose personal information they haven't been able to monetize yet.
Your reasons seem fair. But the opposite reason is that internet access is what will *enable* folks to do better - better market prices for their produce, better job opportunities, better oversight of their leaders, better community organization.
We already know that now, as according to the DEA's 2018 report, the most common way for drugs to enter the country in the south is via the points of entry already. After that is tunnels, light aircraft, and then marine vehicles.
I guess the citation is this one: https://www.dea.gov/sites/defa...
Mexican TCOs transport the majority of illicit drugs into the United States across the SWB using a wide array of smuggling techniques. The most common method employed by these TCOs involves transporting illicit drugs through U.S. POEs in passenger vehicles with concealed compartments or commingled with legitimate goods on tractor trailers
[It specifically refers to "Ports of Entry", POEs, a more precise term than "point of entry". Ports of entry are the existing staffed border stations, that obviously a wall will do nothing to improve.]
Other cross-border smuggling techniques employed by Mexican TCOs include the use of subterranean tunnels... Mexican TCOs also transport illicit drugs to
the United States aboard commercial cargo trains and passenger buses. To a lesser extent, Mexican TCOs use maritime vessels off the coast of California. Mexican TCOs also rely on traditional drug smuggling methods, such as the use of backpackers, or “mules,” on clandestine land trails to cross remote areas of the SWB into the United States. Mexican TCOs exploit various aerial methods to transport illicit drugs across the SWB. These methods include the use of ultralight aircraft and unmanned aerial systems (UASs) and drones to conduct air drops. Ultralights are primarily used to transport marijuana shipments, depositing the drugs in close proximity to the SWB. Currently, UASs can only convey small multi-kilogram amounts of illicit drugs at a time and are therefore not commonly used, though there is potential for increased growth and use. Mexican TCOs also use UASs to monitor the activity of U.S. law enforcement along the SWB to identify cross-border vulnerabilities.
The only one of these that the wall would stop are "traditional drug smuggling methods, such as the use of backpackers, or “mules,” on clandestine land trails". I'm curious what percentage this makes up.
Ok, still trying to understand. What law did they break. I'm by no means trying justifying what they did, but trying to have a conversation.
The law they broke is the UK Data Protection Act. Read the Guardian article. It's all there. Even if, as you say, Cambridge Analytica obeyed the Facebook terms of service, nevertheless...
(1) Cambridge Analytica held data on people; (2) Under UK law at the time - the "Data Protection Act" - if a company holds data on you, then the company is obliged to divulge that data to you personally upon your request; (3) someone did request their data and Cambridge Analytica ignored the request; (4) the UK's Information Commissioner's Office - which has authority under UK law relating to this part of the Data Protection Act - issued an enforcement notice which in the UK has weight of law, requiring Cambridge Analytica to comply with the request; (5) Cambridge Analytica declined the comply with the request arguing that it didn't have to; (6) a law court found that Cambridge Analytica did indeed have obey the enforcement notice, and had broken the law by not doing so.
Summary: the law they broke is the UK Data Protection Act.
They're either buying into the same marketing and media hype for the half-assed excuse for AI everyone keeps trotting out, or they've got something nobody else has, meaning general AI. The latter is highly unlikely, if they did we wouldn't be hearing about it at all.
Or the option you didn't mention -- they know that existing ML-based systems are really good at pattern-matching, and that many forms of pattern-matching don't require full general AI in order to be cost effective, and they're applying it (like the headline and article says) to another domain for pattern-matching.
if the economy only produces hotdogs and buns, automating hotdog manufacture gets you producing 33% more hotdogs with 33% less labor.
That's fair. But here's another example. If the service economy has fast-food-cashiers and trash cleaners, then automating fast-food-cashiers won't do anything. That's because the rate of service is limited by the rate of customers, which is already being met fine.
The solution to this is simple. Fines should not go into the government's general fund. They are punitive penalties for misbehavior against society. They should be put into a separate fund. Then on April 10 when you file you income taxes, the fund should be divided by the number of taxpayers, with each taxpayer receiving an equal share as a generic tax credit. In that way it's paid out directly to the people who were harmed by the initial violation. The government doesn't see a dime of it (giving it as a tax credit just reduces the need to send a check to every taxpayer), so the government's incentive for things like red lights is to time them to maximize safety, rather than to maximize revenue.
That's a great idea!
How are you going to "save the environment" by providing tax breaks for metal boxes so rich people have easier time commuting to work? It doesn't make a dent in anything. If you want to "save the environment" you shouldn't be driving around in ANY CAR.
It'd make a huge dent to switch over a country to electric vehicles 100%. And the only question is: what is the most cost-effective way for a society to achieve this transition?
I think the current model used in Norway and US seems like a really good way for society to leverage the free market - provide small targeted incentives which yes do benefit rich people because their real goal is to encourage innovation and this is the best place for leverage. In this case, we've seen that the benefits really do trickle down to all of society.