The only alternative where companies don't have to hire private detectives in order to give you credit would be for the government to provide an easy way for lenders to check public records to see if people have gone bankrupt or have been sued for failure to pay a debt.
That theoretically might be better as there would have to be due process around what information is provided to prospective lenders, but it could also be even worse than the current system because the government is usually less accountable for its own mistakes due to sovereign immunity.
Besides reporting people's bad debts the other service that these central services provide is a way for lenders to see how much outstanding loans a person has taken out which is a warning to lenders not to lend additional money... again this could be a service the government provides as all loans might have to be registered with the government for them to be valid... as mortgages currently are.
But I think the current system (plus better execution) is probably about as good as it gets. I would like the government to continue policing the credit services and do a better job at it and the courts continuing to arbitrate disputes rather than the government managing the system itself.
Overall though you still need centralized credit reporting otherwise you can't have trade and transactions which involve debt... and that would undermine a lot of people's ability to buy more costly things and the ability to sell more costly things... setting us back to the days of "layaway" where people could lock in a sale price of something and make payments and then receive what they purchased after they had paid off the balance. Basically it would be a big negative to the current economy to reduce the role of credit.
I find the advantage of Amazon to be convenience and selection. Price is a secondary concern. I want it now, but driving for 15 mins each way and messing around in a store doesn't qualify as "now".
I think he was talking about how Amazon got its start... online books. Amazon used to be just an online book store that had cheaper prices and cheap shipping if you didn't mind waiting a week to 10 days because they negotiated lower prices in exchange for lower quality shipping. Talk about taking a niche market and using it to build out... which was the intent from the beginning not just to sell books for less than the competition. Now they have everything and you can have it either same day or a week later, or from third parties... or digital content, or.... everything.
But it used to be mostly just books... and then a few other things. Just as Google started as just an html input text and a button that connected to a few fast servers that did one thing indexing web pages for fast and relevant search results really really well.
I think it is point taken that start-ups should focus on those few things they can do well and keep their "vision" in check with the reality of what they can actually deliver. Heck Apple learned that lesson pretty well and doesn't even talk about new products until they are rolling.
Sometimes hype is necessary to get funding, but both investors and start-ups are worse off for it.
Yes. However, given that the key has to be the same length as the cleartext and can never be reused, that makes it an unworkable solution for two-way electronic communications.
It's just barely feasible for things like numbers stations.
These days you can fit 256GB on a microSD card. For point to point communication that's quite a lot. You could also smuggle two or more separate versions by different routes and XOR them together at the destination to guard against a single courier being intercepted.
It would be less secure, but easier to do say among a team that gets together every week or once a day in the morning (for a bit of coffee, a status update and a pad exchange)... if you periodically see people, then just have an app running in the background that does a one time pad swap in the background while you are in direct wireless communications range. Say transfer 300 Mb per person... ten people that is 3 Gb... which is doable.
Sure, people can be listening in on the pad exchange (or have a network of monitors in place to hoover all the wireless data up around a city or populated area) but if you are under that kind of intense surveillance already then there are already twenty different ways your communications are going to be intercepted almost regardless of what you do.
But yes, exchange of one time pads via a physical connection through removable media is very practical for gigabytes or even multiple terabytes of high value data as long as you can predict about how much data you will need in a given period of time before the next pad exchange.
It wasn't so long ago that gigabytes of data were best transferred on physical media anyway just due to physical limitations of network bandwidth and cost constraints.
Shipping data via delivery truck or hand delivery is still very much a current best practice for one time transfers of very large amounts of data. It probably should be considered best practice for high value one time exchanges of data too.
One time pads aren't unbreakable from a practical standpoint though, just theoretically unbreakable if you have perfectly random pad generation and perfect pad exchange. Would be good to see one time pad based encryption used more and then we can properly flesh out all those practical implementation issues.
Such a bullshit nonstory, such a bullshit headline. Fuck you, Beau.
Yes, I don't understand the editorial line that/. has taken - it is becoming more and more about inflating trivia to make it sound sensational, rather than real news with some thoughtful analysis behind. The thing is, this editorial line frustrates those of us who have been faithful readers for years, adding much of the comment that is actually driving the success of/. - when we submit comments, we do valuable work for the site in generating interest and starting cascades of comments etc, and we don't receive payment in any form. On that background, is it wise of the editors to constantly frustrate us with deceptive headlines? Every time I come across such a story and click on a link to an idiotic, vapid non-story, I get a little closer to simply abandoning/. as inconsequential. That is sad, I think - at on time this community gave name to the 'slashdot effect' (https://en.wikipedia.org/wiki/Slashdot_effect), but when has that last happened? Bad editorship is what has eroded the core contributers away - those of us that are still left, stay mostly out of habit.
I still like the user based moderating system better than most discussion websites. At this point do stories automatically get promoted to the home page based on the firehouse submissions? If so I could understand how things like this get through. Otherwise this appears to be an attack ad by Google's competitors rather than real news story.
I dunno. I see AI with decicion making powers happening at the tactical and theatre levels: semi autonomous weapons that are given a mission and the execute it with leeway to adjust along the way, or an AI coordinating troops and autonomous units. Enough options for a rogue AI to cause terrible damage, but not really something that will spark WW3 before humans can intervene.
At the strategic level, AI could well support decision making, but what would be the value of actually putting the AI in charge there? That makes sense only if you need to make split second decisions, or launch a counterstrike even if all meatbag commanders are dead. That's a cold war standoff scenario; I don't see it being really useful for anything else.
Decision support. That is the real risk. That we come to rely on our AI based modeling for decision support and suddenly it gets the scenario very wrong and outputs a recommendation to take a disastrous course of action that seems perfectly reasonable at the time.
That our black boxes become so good at predicting human behavior that we come to rely on them to help make decisions for example about what the repercussions will be if we preemptively strike a missile launch site in North Korea. That our models, with an AI trained on thousands of scenarios predict with 94% certainty that such a strike will achieve mission objectives and that there will be minimal retaliation... but in reality it causes a series of events that lead to World War III. Maybe it is bad data, (garbage-in-garbage-out) or simple the AI learning on a set of training data that doesn't apply to the new scenario.
In human terms we often make mistakes when we don't realize that our experience doesn't apply to a new scenario or a new phenomenon. The same thing is true of an AI trained algorithm. But if we as humans come to rely on an AI without understanding its scope and limitations then that to me is where the real risk hides.
Thugs can subvert an election by coercing the vote counters and elections officials far more easily than they can coerce a significant enough number of voters.
There is plenty of room in the middle to believe that the poor should have a good effective health care safety net paid for with an equitable tax but that we should also have freedom of choice in how we purchase health care and to allow people to save up their money in good health instead of purchasing health insurance. And who on either side supports the middle income cliff/wall where millions of people make too much for subsidies, but too little to afford adequate health insurance?
I believe you had to raise $250,000 from investors to qualify. So it would be option #2.
That seems like quite a low threshold and meant for creating a loophole....
Make the threshold $10 million, have effective government oversight to check and see that these are real start-ups intended on providing products and/or services (and not just to a related company or person) and then let's talk about whether it is a good thing to have wealthy foreigners come here to employ Americans.
Either way we should really focus on a reasonable number of green cards and new citizenships for people that want to come here and become American citizens.
The idea of sculping the iceberg to reduce drag is interesting... perhaps less like a sphere and more like a hull would be a bit more efficient. Though it might happen naturally in warmer waters as the iceberg is pushed North and the rougher edges melt away.
Also, the melting ice itself might reduce drag.
Would need to just go ahead and do it once with the ice berg as-is to practically baseline the efficiency.
If they can reduce costs sufficiently and get some additional uses out of the rockets that have flown multiple times, then this could be a brilliant way to create value while further demonstrating how many times their rockets can fly.
The whole point of the massive NSA datacenter in Utah is that they collect _everything_. The argument from the NSA and Federal Government was that they would only look at data where they had a warrant. Our argument back was that there is no way to ensure data is only viewed by warrant, especially when they were looking at ways of cataloguing data they could see, and trying to crack encryption on what they could not.
We were right, they were dishonest. Nothing new in terms of Government abusing power, and nobody should be surprised that the more we give them the more they abuse.
Since the hardware is already in place to copy all traffic to the NSA, law changes which impact collection of data would have to tackle that particular issue. Good luck with that. ISPs and Telecom providers get paid massive tax dollars to provide the service, so you know that they won't complain.
Well said. I would just add that beyond the practical privacy concerns of people actually snooping on other people for illegitimate purposes there is an important principle of constitutional law that the government is required to have a specific warrant to perform a search. And that making copies of data and in fact scanning that data in the first place to see if it is relevant to a variety of ongoing surveillance activities is itself a search.
It isn't merely the potential for search of the data without warrant after the government has collected it that is the issue. In fact if the data was legally collected and the government has it, then why shouldn't it be available for any legitimate investigation? Any legally collected data should be available to investigators. The problem is that it isn't legally collected data.
If it were just a practical privacy issue then you are already exposed to numerous companies that are collecting, analyzing, storing your communications for a variety of purposes that you might not want to specifically agree to, but might be somehow covered in a customer agreement.
The other big thing that I see as unconstitutional is that the government is effectively not allowing companies the option of an enforceable contract with their customers that it will require a specific warrant to divulge their communications to the government. The big telecoms got that as legal cover, but also to head off competitor companies marketing privacy as something they could legally deliver.
In the US, at least, privacy in your communications against unconstitutionally broad government surveillance isn't an option companies can even offer their customers and business partners because the agreement is made legally unenforceable with no opportunity to seek damages for contract violations against telecom providers under the Patriot Act. I should be allowed to provide customers with a privacy agreement in return for compensation that if I violate it would allow them to seek damages in court.
So the government is unconstitutionally interfering in what should be a lawful privacy contract between telecoms and their customers and business partners.
What really bothers me is that your "values" and sense of "fairness" are keeping a trillion dollars from coming back into the US economy.
Go ahead and tax the rich people with a higher progressive tax bracket... But the US economy needs all that capital from corporate profits. Let it be taxed as income when it gets put back to work in the US economy.
Bingo... We are the only country in the world that taxes money coming in and subsidizes money going out. We really need to bring that money back to the US and off set that by raising taxes on the top 5 or 10% while lowering taxes on middle income folks and the corporations that employ us.
Yes it appears the Obama administration abused it's surveillance powers... But I haven't heard Trump say anything about curtailing those powers in his administration. One thing to be indignant... And then another to tighten rules to ensure that the abuse of power won't happen again in any administration.
Good points. Perhaps ceding that non-citizens have no rights is not the right starting point.
Government should get a warrant.
Government should also respect people's right to privacy and end all drug prohibition so most searches are not needed in the first place and won't find any "contraband" being smuggled. But that is another discussion.
So we can dispense with constitutional rights as long as we declare US soil not to be US soil for the purposes of that particular constitutional right?...
The rule of law is untenable when so easily tossed aside when inconvenient to the government.
I think calling it a "liberal bias" in the first place is very misleading. It is a Democratic Party bias of the big city press. Born of cultivated political and personal relationships made in big cities controlled by the Democratic Party which also happen to be the centers of media markets.
I've personally seen this play out where reporters from the main press outlets are given office space inside city hall and the state government and clearly develop a symbiotic relationship with the political establishment. The only time reporters turn on the powers that be in City Hall is when they see blood in the water and see the prospect of a new patron taking over.
The only alternative where companies don't have to hire private detectives in order to give you credit would be for the government to provide an easy way for lenders to check public records to see if people have gone bankrupt or have been sued for failure to pay a debt.
That theoretically might be better as there would have to be due process around what information is provided to prospective lenders, but it could also be even worse than the current system because the government is usually less accountable for its own mistakes due to sovereign immunity.
Besides reporting people's bad debts the other service that these central services provide is a way for lenders to see how much outstanding loans a person has taken out which is a warning to lenders not to lend additional money... again this could be a service the government provides as all loans might have to be registered with the government for them to be valid... as mortgages currently are.
But I think the current system (plus better execution) is probably about as good as it gets. I would like the government to continue policing the credit services and do a better job at it and the courts continuing to arbitrate disputes rather than the government managing the system itself.
Overall though you still need centralized credit reporting otherwise you can't have trade and transactions which involve debt... and that would undermine a lot of people's ability to buy more costly things and the ability to sell more costly things... setting us back to the days of "layaway" where people could lock in a sale price of something and make payments and then receive what they purchased after they had paid off the balance. Basically it would be a big negative to the current economy to reduce the role of credit.
Like it or not, we have a debt based economy.
I find the advantage of Amazon to be convenience and selection. Price is a secondary concern. I want it now, but driving for 15 mins each way and messing around in a store doesn't qualify as "now".
I think he was talking about how Amazon got its start... online books. Amazon used to be just an online book store that had cheaper prices and cheap shipping if you didn't mind waiting a week to 10 days because they negotiated lower prices in exchange for lower quality shipping. Talk about taking a niche market and using it to build out... which was the intent from the beginning not just to sell books for less than the competition. Now they have everything and you can have it either same day or a week later, or from third parties... or digital content, or .... everything.
But it used to be mostly just books... and then a few other things. Just as Google started as just an html input text and a button that connected to a few fast servers that did one thing indexing web pages for fast and relevant search results really really well.
I think it is point taken that start-ups should focus on those few things they can do well and keep their "vision" in check with the reality of what they can actually deliver. Heck Apple learned that lesson pretty well and doesn't even talk about new products until they are rolling.
Sometimes hype is necessary to get funding, but both investors and start-ups are worse off for it.
Yes. However, given that the key has to be the same length as the cleartext and can never be reused, that makes it an unworkable solution for two-way electronic communications.
It's just barely feasible for things like numbers stations.
These days you can fit 256GB on a microSD card. For point to point communication that's quite a lot. You could also smuggle two or more separate versions by different routes and XOR them together at the destination to guard against a single courier being intercepted.
It would be less secure, but easier to do say among a team that gets together every week or once a day in the morning (for a bit of coffee, a status update and a pad exchange) ... if you periodically see people, then just have an app running in the background that does a one time pad swap in the background while you are in direct wireless communications range. Say transfer 300 Mb per person... ten people that is 3 Gb... which is doable.
Sure, people can be listening in on the pad exchange (or have a network of monitors in place to hoover all the wireless data up around a city or populated area) but if you are under that kind of intense surveillance already then there are already twenty different ways your communications are going to be intercepted almost regardless of what you do.
But yes, exchange of one time pads via a physical connection through removable media is very practical for gigabytes or even multiple terabytes of high value data as long as you can predict about how much data you will need in a given period of time before the next pad exchange.
It wasn't so long ago that gigabytes of data were best transferred on physical media anyway just due to physical limitations of network bandwidth and cost constraints.
Shipping data via delivery truck or hand delivery is still very much a current best practice for one time transfers of very large amounts of data. It probably should be considered best practice for high value one time exchanges of data too.
One time pads aren't unbreakable from a practical standpoint though, just theoretically unbreakable if you have perfectly random pad generation and perfect pad exchange. Would be good to see one time pad based encryption used more and then we can properly flesh out all those practical implementation issues.
Such a bullshit nonstory, such a bullshit headline. Fuck you, Beau.
Yes, I don't understand the editorial line that /. has taken - it is becoming more and more about inflating trivia to make it sound sensational, rather than real news with some thoughtful analysis behind. The thing is, this editorial line frustrates those of us who have been faithful readers for years, adding much of the comment that is actually driving the success of /. - when we submit comments, we do valuable work for the site in generating interest and starting cascades of comments etc, and we don't receive payment in any form. On that background, is it wise of the editors to constantly frustrate us with deceptive headlines? Every time I come across such a story and click on a link to an idiotic, vapid non-story, I get a little closer to simply abandoning /. as inconsequential. That is sad, I think - at on time this community gave name to the 'slashdot effect' (https://en.wikipedia.org/wiki/Slashdot_effect), but when has that last happened? Bad editorship is what has eroded the core contributers away - those of us that are still left, stay mostly out of habit.
I still like the user based moderating system better than most discussion websites. At this point do stories automatically get promoted to the home page based on the firehouse submissions? If so I could understand how things like this get through. Otherwise this appears to be an attack ad by Google's competitors rather than real news story.
Oh wait... would it end up being as effed up as China?
I dunno. I see AI with decicion making powers happening at the tactical and theatre levels: semi autonomous weapons that are given a mission and the execute it with leeway to adjust along the way, or an AI coordinating troops and autonomous units. Enough options for a rogue AI to cause terrible damage, but not really something that will spark WW3 before humans can intervene.
At the strategic level, AI could well support decision making, but what would be the value of actually putting the AI in charge there? That makes sense only if you need to make split second decisions, or launch a counterstrike even if all meatbag commanders are dead. That's a cold war standoff scenario; I don't see it being really useful for anything else.
Decision support. That is the real risk. That we come to rely on our AI based modeling for decision support and suddenly it gets the scenario very wrong and outputs a recommendation to take a disastrous course of action that seems perfectly reasonable at the time.
That our black boxes become so good at predicting human behavior that we come to rely on them to help make decisions for example about what the repercussions will be if we preemptively strike a missile launch site in North Korea. That our models, with an AI trained on thousands of scenarios predict with 94% certainty that such a strike will achieve mission objectives and that there will be minimal retaliation... but in reality it causes a series of events that lead to World War III. Maybe it is bad data, (garbage-in-garbage-out) or simple the AI learning on a set of training data that doesn't apply to the new scenario.
In human terms we often make mistakes when we don't realize that our experience doesn't apply to a new scenario or a new phenomenon. The same thing is true of an AI trained algorithm. But if we as humans come to rely on an AI without understanding its scope and limitations then that to me is where the real risk hides.
What's more likely... coercion of elections officials or coercion of a plurality of voters?
Far far easier to coerce a small number of elections officials.
Thugs can subvert an election by coercing the vote counters and elections officials far more easily than they can coerce a significant enough number of voters.
There is plenty of room in the middle to believe that the poor should have a good effective health care safety net paid for with an equitable tax but that we should also have freedom of choice in how we purchase health care and to allow people to save up their money in good health instead of purchasing health insurance. And who on either side supports the middle income cliff/wall where millions of people make too much for subsidies, but too little to afford adequate health insurance?
The US is a one party system with two parties.
Forget China.... the Clinton campaign worked like hell to make sure the Democratic Party primary was a coronation and not a serious primary.
How can this effort be serious when the people who worked to subvert a free and fair election in the US are given leadership positions?
I believe you had to raise $250,000 from investors to qualify. So it would be option #2.
That seems like quite a low threshold and meant for creating a loophole....
Make the threshold $10 million, have effective government oversight to check and see that these are real start-ups intended on providing products and/or services (and not just to a related company or person) and then let's talk about whether it is a good thing to have wealthy foreigners come here to employ Americans.
Either way we should really focus on a reasonable number of green cards and new citizenships for people that want to come here and become American citizens.
The idea of sculping the iceberg to reduce drag is interesting... perhaps less like a sphere and more like a hull would be a bit more efficient. Though it might happen naturally in warmer waters as the iceberg is pushed North and the rougher edges melt away.
Also, the melting ice itself might reduce drag.
Would need to just go ahead and do it once with the ice berg as-is to practically baseline the efficiency.
If they can reduce costs sufficiently and get some additional uses out of the rockets that have flown multiple times, then this could be a brilliant way to create value while further demonstrating how many times their rockets can fly.
Critically, they generally argue that it isn't actually surveillance until a human being reviews it.
Maybe it isn't "surveillance". But coming into my business and taking my electronic records sure as hell is covered under the 4th amendment.
The whole point of the massive NSA datacenter in Utah is that they collect _everything_. The argument from the NSA and Federal Government was that they would only look at data where they had a warrant. Our argument back was that there is no way to ensure data is only viewed by warrant, especially when they were looking at ways of cataloguing data they could see, and trying to crack encryption on what they could not.
We were right, they were dishonest. Nothing new in terms of Government abusing power, and nobody should be surprised that the more we give them the more they abuse.
Since the hardware is already in place to copy all traffic to the NSA, law changes which impact collection of data would have to tackle that particular issue. Good luck with that. ISPs and Telecom providers get paid massive tax dollars to provide the service, so you know that they won't complain.
Well said. I would just add that beyond the practical privacy concerns of people actually snooping on other people for illegitimate purposes there is an important principle of constitutional law that the government is required to have a specific warrant to perform a search. And that making copies of data and in fact scanning that data in the first place to see if it is relevant to a variety of ongoing surveillance activities is itself a search.
It isn't merely the potential for search of the data without warrant after the government has collected it that is the issue. In fact if the data was legally collected and the government has it, then why shouldn't it be available for any legitimate investigation? Any legally collected data should be available to investigators. The problem is that it isn't legally collected data.
If it were just a practical privacy issue then you are already exposed to numerous companies that are collecting, analyzing, storing your communications for a variety of purposes that you might not want to specifically agree to, but might be somehow covered in a customer agreement.
The other big thing that I see as unconstitutional is that the government is effectively not allowing companies the option of an enforceable contract with their customers that it will require a specific warrant to divulge their communications to the government. The big telecoms got that as legal cover, but also to head off competitor companies marketing privacy as something they could legally deliver.
In the US, at least, privacy in your communications against unconstitutionally broad government surveillance isn't an option companies can even offer their customers and business partners because the agreement is made legally unenforceable with no opportunity to seek damages for contract violations against telecom providers under the Patriot Act. I should be allowed to provide customers with a privacy agreement in return for compensation that if I violate it would allow them to seek damages in court.
So the government is unconstitutionally interfering in what should be a lawful privacy contract between telecoms and their customers and business partners.
What sorcery is this of which you speak? How does $250 billion cash turn into a trillion?
Make that $2.6 trillion... Companies are holding a $2.6 trillion pile of cash overseas that's still growing
What really bothers me is that your "values" and sense of "fairness" are keeping a trillion dollars from coming back into the US economy.
Go ahead and tax the rich people with a higher progressive tax bracket... But the US economy needs all that capital from corporate profits. Let it be taxed as income when it gets put back to work in the US economy.
Bingo... We are the only country in the world that taxes money coming in and subsidizes money going out. We really need to bring that money back to the US and off set that by raising taxes on the top 5 or 10% while lowering taxes on middle income folks and the corporations that employ us.
Because then the US economy would be good, wages would rise and you couldn't blame evil corporations... Now do you see why ?
Problem solved... Well done everyone.
Yes it appears the Obama administration abused it's surveillance powers... But I haven't heard Trump say anything about curtailing those powers in his administration. One thing to be indignant... And then another to tighten rules to ensure that the abuse of power won't happen again in any administration.
Good points. Perhaps ceding that non-citizens have no rights is not the right starting point.
Government should get a warrant.
Government should also respect people's right to privacy and end all drug prohibition so most searches are not needed in the first place and won't find any "contraband" being smuggled. But that is another discussion.
So we can dispense with constitutional rights as long as we declare US soil not to be US soil for the purposes of that particular constitutional right?...
The rule of law is untenable when so easily tossed aside when inconvenient to the government.
tends to have a liberal bias.
Get over it.
I think calling it a "liberal bias" in the first place is very misleading. It is a Democratic Party bias of the big city press. Born of cultivated political and personal relationships made in big cities controlled by the Democratic Party which also happen to be the centers of media markets.
I've personally seen this play out where reporters from the main press outlets are given office space inside city hall and the state government and clearly develop a symbiotic relationship with the political establishment. The only time reporters turn on the powers that be in City Hall is when they see blood in the water and see the prospect of a new patron taking over.