A site could disallow access from problematic countries. For example, allowing East Asian countries read access but not post (to forums) access might cut down on sock puppet and spam replies.
Of more interest is the NSA angle. Suppose your website disallows visitors from within 50 miles of Washington, DC. Or better yet, shows sanitized links to visitors known to be associated with the government.
Any IT person will know that this is trivial to circumvent, but look at it from their point of view: Nothing they use locally will see the links they need, anyone outside the radius can't send a link into the circle for review, and setting up a tunnel (VPN &c) to a location outside the radius is a pain, and all the effort could be invalidated by the website adding the tunnel exit to the block list.
It wouldn't be hard to keep a global list similar to the SPAM blocklist sites that have lists of IPs used by government. You could download a blacklist that includes the local police station, state police, and FBI building. People could "report" access from government agencies like they currently report spam activity. It would be much easier to hold that demonstration without the police knowing your plans in advance.
Again these are not difficult to circumvent, but it makes it harder for the criminals to get in, and economy of scale is on your side: one blocklist would have to be circumvented by each agency addressed. One action on your part needs actions from multiple parties to compensate.
If there were a simple implementation of this - say, an Apache plugin that periodically grabs the blacklist - it would be a big headache for the overlords.
The US can kill an American and his teenage son, yet no one can challenge the action because they were not directly affected. If all the relatives are taken out in one action, then the US is free and clear.
We can't just protest to have unconstitutional laws removed, we have to prove they were used on us. Simply keep quiet about parallel construction and you're good to go. If the defendant says "yes they did" and the US says "no we didn't", then the constitutionality of the law makes no difference, the US is free and clear.
This thing about not challenging a law because it doesn't affect you is bullshit.
If a law is unconstitutional, then it should be possible to challenge the law on its face.
In the case of Lavabit, the government demanded, and was given, a warrant for the HTTPS private key to monitor the online actions of a couple of defendants. This would allow the FBI to monitor not only the specific defendants, but all Lavabit customers.
And I want to be totally clear about this: The government asked to install a pen trap device *and* have the private keys which would have allowed it to monitor all Lavabit customers.
(Unlike phone companies, E-mail providers are under no legal obligation to make surveillance easy, or even possible, by the government.)
Third parties have a duty to assist law enforcement, but that duty does not extend "regardless of the burden involved". The ACLU argument is that giving over the private keys would have completely destroyed the Lavabit business, which was an unreasonable burden to take in assisting law enforcement.
You do when they have a warrant.
Just saying "You do when they have a warrant" is no longer sufficient. There's ample evidence that judicial oversight has been compromised by the FISA court et al., and this is a particularly strong case of government overreach.
According to news reports, there are around 1000 analysts at NSA engaged in surveillance. Let's assume half of them are looking at foreign traffic and half at domestic traffic. That's 500 analysts for 350 million population, or 1 analyst for every 700,000 people. What makes you think you are special enough to deserve their attention?
Okay, let's look at those statistics more closely.
500 analysts for 350 million people continuously is 500 analysts for roughly 1 million people per day each year, or roughly 1 analyst is spending an entire day looking at 2,000 people. Each year. So there's a 1-in-2,000 chance that sometime this year, an analyst will be pawing through your online behaviour.
(Of course, if you assume that the analyst spends 1 hour on each person, it drops to 1-in-250 chance that sometime during the year you will be "analyzed" by an NSA agent.)
Now consider the power of computers. Is it reasonable to think that 1 computer could collect and analyze the E-mail and online speech of 2,000 people in a single day of compute time? Assuming you put certain keywords in your online text ("I'm going to kill some time this afternoon by watching the presidential debate"), how likely do you think it will be that you win the 1-in-250 chance?
Let's add in ambiguous laws. The recent trend is not to charge people with doing harm, but conspiracy for doing harm. One recent news report told of a couple of people charged with "conspiracy to join Al-Qaeda". Note that these two people didn't do a terorrist act, they didn't contribute to a terrorist group, and they weren't even a member of a terrorist group. They were talking about joining a terrorist group. People are commonly charged with "conspiracy to grow marijuana" (google has many links).
We've reached the point where you can be arrested when no overt crime has been committed.
There's a recent news story where, for the first time, the DOJ is informing a defendant that they used NSA/warrant-less surveillance to gather evidence. They used mass surveillance to get enough probable cause to apply for a real warrant which resulted in evidence of a crime.
The important bit of the previous is that the DOJ was conflicted about revealing this information. The prosecutor felt that it was only a "procedural decision", since no evidence from the mass-surveillance warrant would be introduced at trial. (A couple of lawyers in the DOJ argued for disclosure.)
All evidence indicates that they analyze everyone's online presence all the time, and use that information to pick-and-choose people for prosecution when no overt crime has been committed.
This thing about TrueCrypt has turned into a test-case for software security practices.
Loosely, the security audit has three sections: 1) Look at the sources, 2) Look at the license agreement, 3) Verify that the binaries match the sources.
Item #3 above seems to be a generic problem: assuming that the source is correct, how can people verify that binaries were compiled from the source? Any minor change to the compiler can generate different instructions for the same code, and many compilers insert time-stamps and signatures (login of user running the compiler) into the final executable. This makes comparing binaries a problem.
Is there a better way to solve this? For example, could the GNU compiler digitally sign an executable, so that the executable could be verified by a web app on the GNU foundation page? (Generate a signature from source+binary and include it in the executable, for example.)
Can someone come up with a good solution for executable integrity?
Look, nobody likes taxes, licensing restrictions, having to clean your car, or requiring you don't just hang out at the airport where people will pay tons of money.
The reasons we have those is that unlicensed cabs were a big problem.
Unlicensed cabs were a big problem because cabs and customers were not regulated.
The government stepped in and cleaned up the cabs, enforcing a standard of quality control of the cabbies but not the customers. It's the "regulation" model, and it was appropriate for its time, but it only addressed half the issue: a customer could jump out and run away without paying, could slit the seat, could vomit in the seat, or do other unsavory things.
Over time the regulation became less enforced, watered down, corrupt, and fewer people cared. This has resulted in the situation we have now, where many cabs are filthy and disgusting, the cabbie will screw you out of money in various ways (jimming the meter, taking the long route, &c), and it's not particularly safe.
In game theory terms, it's two kids dividing a cake: mom tells one kid to divide the cake equally, then leaves.
With the rise of ubiquitous communication we can now go to a newer model: both cabbies and customers can be vetted by the system. The cabbies are reviewed by the feedback of customers, and the customers are reviewed by the cabbies. Anyone who slits a seat or vomits will get a bad review and won't have access to the drivers in the future. Anyone who drives a filthy car will get a bad review and not have access to passengers in the future.
The game-theory model is different. Instead of one side promising to obey regulation, it's two sides regulating each other. It's the "one child divides the cake, the other child chooses which piece to eat" model.
This is an example of bad regulation which stifles innovation. Cab regulation ensured quality and was done with the best of intentions, but it's been subverted and there's now a better way.
There's a lot of push towards I/O systems that are more convenient for the manufacturers, but less so for the end consumer.
The first keyboards were heavy and had tactile feedback. If you fumbled a key typing in your password, you knew whether it entered because you could feel the "click". Nowadays the keyboard is lighter than a paperback and there's no feedback - accidentally brush a key with your finger and you have to look (for non-password text entry) or start over.
Twist knobs are highly intuitive, especially when coupled with feedback. Twist a knob and see the hands of the clock move, or see the numbers change. Control the speed natively, and if you go too far it's obvious how to back up. Nowadays we have buttons to tap, incrementing the count by 1 each time. Tap 50 times to set the minute display, and if you go too far you have to go all the way around again. This was done largely because buttons are easy to fabricate (using PCB contacts), not because they are inherently better.
Modern typing is done on the display (phone, surface), so not only don't you have tactile feedback you can't feel the boundaries of the keys, and your fingers mask the key display. And it's really tiny - in order to access all the keys you have to type extra keys that switch between keyboards (upper/lower/symbol). Again, it was done for ease of manufacturing, not ease of use.
Is the ribbon any easier than, for example, cascading menus? The problem with the Windows original menu system was that every application put their commands in the top-level Start->Programs folder, leading to start menus containing hundreds of links. (I take the time to move StartMenu command links into subfolders by type, which makes it much easier, but on my dad's computer it's impossible to find anything.)
Ever since minority report people have been touting the wonders of air-gesture input, and that it is the next "big thing", but is it better? (Actually, I remember it from Johnny Mnemonic, 7 years earlier.) Seems like this is just something that's easier to manufacture, but not easier to use. Sure, the customer will be able to do everything they could do with a mouse/keyboard, but more slowly, less conveniently, and with lots of frustration. That's an externality to the manufacturers, but it's better for them because they don't have to build in a touch interface. Probably [electrically] more reliable, too.
Is this really progress? I wonder what Edward Tufte would say about modern interfaces.
I nodded this in the firehose because it looked interesting.
There's not much information in the linked article. Can anyone give us more info? Anyone who reads Finnish care to comment on the source - is it reliable, are the researchers legitimate?
Self-awareness is one of those terms that doesn't have a constructive definition, so it's not a good metric.
In a literal sense the meaning is "has sensory input relating to self", in which case my laptop is self-aware: it knows the amount of charge in it's battery and will beg me (by "beeping" plaintively) for more energy. (Usually to no avail - I'm a right bastard.)
Most of the literature takes the term to mean "conscious", which also doesn't have a good definition. The best I can come up with is that a machine is "conscious" when it has an internal model of its environment, with itself as a separate entity. Thus, humans have mirror neurons which allow them to learn by seeing someone else's mistake, can recognize themselves in a mirror, &c. Great apes recognize themselves in a mirror, while monkeys do not.
It's not clear whether humans have free will at all. At some level, an intelligence has to perform the risk/reward calculation and take the best outcome for action. Most incidents of bad outcome are due to random chance, improperly calculating the odds, not having enough knowledge, or insufficient perspective. (A soldier diving on a grenade is working to benefit his people, which is intelligence at the genetic level - the genes in his society are more likely to survive even when the individual is sacrificed.)
We do know that our universe does not preclude free-will. Randomness, which is an aspect of every sensory input we have, is necessary for free will, but no one knows whether it is sufficient. If there were no randomness, then an outcome could be completely predicted - including (in theory) human decisions.
No one really knows if there are other requirements for free will, or what the list of satisfiable conditions should be.
Perhaps we should start by completely defining what "free will" even means.
Some people have been studying the phenomenon of "upheval" in it's generic form. It's spawned a lot of studies/papers and even popular books, viz: The Tipping Point.
The overall summary is that you can't just point out how bad something is, you have to give people an action they can take to help fix the problem.
There is widespread distrust, anger, and annoyance at the NSA due to the revelations. There's no public outrage because there's really nothing anyone can do. "Joe Sixpack" has no actions to take: voting doesn't help, writing congresscritters doesn't help, even public mass demonstrations don't seem to help. What you are seeing is Learned Helplessness: an animal doesn't take actions to help themselves, because they're convinced that the actions will have no effect.
Consider the recent history of cell phones or music distribution: people were complaining that cell phones were a walled ecology with no innovation and poor functionality. You had to get carrier approval to run a program on a cell phone, and they would only allow the simplest, meager functionality. You were lucky if your carrier allowed you to have tetris.
People complained that if you wanted music, you had to purchase a physical CD, for an ensemble collection and for an exorbitant fee. Usually you had to purchase an entire CD for a single song you liked.
As soon as an option was given, people flocked to the new systems in droves, uptake was very fast.
Make secure E-mail easy to use with trivial installation and the situation will change overnight. There will be a flood of new users.
Everyone hates the situation, but for most people there's nothing they can do about it.
What no one seems to get is that no one in federal government(*) cares what's right for society, for the people, or even for their own survival.
The purpose of government is to siphon funds away from individuals and give it to corporations. That's the length and breadth of it, there are no other considerations.
The purpose of airport security is to give money to scanner companies. (Oh, these scanners don't work? We'll throw them out and purchase your newer model.) The purpose of Obamacare is to give money to insurance agencies. The purpose of Obamaphone is to give money to the phone companies, the purpose of military spending is to give money to military contractors, and the purpose of the war on drugs is to give money to private prisons.
Every time one of these "this is the right move, but the government is doing the opposite" articles come up (one or two a day, it seems) it's framed in terms of an isolated, poor choice within a sea of government actions that are generally benevolent to the population and make our life better.
It's not an isolated incident, everything the federal government does has one purpose and makes sense within that framework. Occasionally it also benefits the people, but that's more happenstance than plan. It's the "random guess is occasionally correct" principle.
Federal government is a runaway train that's going to crash and burn, taking the country down with it. We can let this happen, or we can curtail it beforehand. Either way works, but fixing it beforehand would seem a better plan.
*Note: I'm making a distinction between federal and state government. Most of the federal government could disappear without negative impact on the people. And yes, I said "most".
Everyone knows that the Tea party is a bunch of comic, laughable clowns with no grounding in reality. I mean, why else would they be so thoroughly lampooned using derogatory terms and snarky, dismissive comments.
Even Obama himself (praise be his name) mocks them.
Personally, I've found more fugly code turds in various closed source projects than I've touched than in the open source world.
Is code the only aspect of note in an open source project?
How is the project named? Is it something reminiscent of the function (like PaintShop Pro, Photoshop, Internet Explorer) or something entirely random, forcing more cognitive load on an uninformed user (Gimp, Firefox, Juice)? Does it have a newish, edgy name to give it that extra sizzle (pantyshot, upskirt).
How is the project configured? Is is a list of poorly-written technobabble? Does the installation instructions begin with the history of the project (of which I am not interested), require other packages which I have to research and choose, does it require cryptic installation actions and complex setup that has to be done by hand?
How does the project look? Are the panels laid out with ease-of-use in mind, or they just show everything and "let the user arrange them as they like"? Is the text font and color scheme appropriate, or is it default, the user can choose the one they like?
Are there lots of icons for every little action, no matter how small (the "kitchen sink" philosophy), or is there a well-chosen subset that balances functionality with ease-of-use? Do the icon shapes bring the function to mind, or are they more-or-less random shapes that rely on popups to tell the user what they do?
Is the documentation well-written by people who are good at explaining things, or is it just a wiki editable by anyone, maintained by the users, with no real structure?
Has the code been tested by someone who is not the lead coder (and not the users)? Does the project use regression tests?
Yeah, nice code you've got there. If that's all I wanted in a product, yours would be a slam dunk.
Apparently Mr. Moalin once missed a telephone call from "Aden Hashi Ayrow, the senior al Shabaab leader," which makes it likely that a little more was going on than merely the donation of "a small sum of money."
Is this important?
He's claiming not that the evidence is wrong, he's claiming that it was collected illegally.
It's often been said that the defense of freedom is the defense of scoundrels (H. L. Mencken). We believe that a kiddie porn merchant has the right to a fair trial, the KKK has the right to assemble, and Rosa Parks has the right to sit in the front of the bus.
Should we base the legitimacy of rights and freedoms on the character of the accused party?
Once again, possibility is not the same as probability.
Yes, it's possible that the elevated radiation levels will cause problems. Now, what's the risk, and what's the tradeoff between mitigating *that* risk versus mitigating some *other* risk?
Security is a tradeoff, always. The value of something is not the face value, but the face value times the probability of occurrence.
So if the probability of damage (say, the number of people getting cancer from going to the event) times the value of damage (taken informally as $1 million per human life lost, but depends on estimates and philosophy) is higher than other foreseeable risks, then we should address the problem.
Risks shouldn't be ignored, just compared to other risks. If the utility losses for other risks are higher, then we should spend our finite resources on the other risks first.
How much risk utility is embodied in this problem compared to, say dying from accidentally swallowing (and choking on) a bee?
...but journalism must sell news. I suppose someone swallowing something wouldn't be very interesting.
... much better that you experience problems when you anticipate them than when everything else is going wrong, too.
So what you're saying is: if you make a 1 million dollar mistake, your response should be "Phew! At least I didn't lose 5 million!".
An outside observer might suggest that losing the $1 million is bad on its face. Mitigating the outrage by making false comparisons is the sort of thing politicians do, as a dodge for responsibility.
Should we be sanguine about these sorts of problems because they're not the worst possible scenario? Is that an acceptable excuse?
This is no surprise really. Who would you trust to program a computer in charge of your life?
You trust a true nerd: Someone who is obsessive about correctness, some distance down the Asperger's spectrum, and who's convinced that the consequences of having a bug are their fault. Hygiene and dress-code are secondary.
I used to code aircraft avionics software (microcontroller stuff for altimeters, airspeed, cabin pressurization, &c). Some of my avionics-related courses asked "are you willing to be the first passenger in an aircraft running this software? raise your hand", and typically mine was the only hand showing.
There's a mindset for making safety-certified software, and not everyone has it. Most people rationalize doing a poor job by denying responsibility: the boss told them to do it, they have to feed their family, everyone else does it, and so on and so on. It's the mindset that allows the NSA get away with rights violations: no one takes responsibility at any level.
A true nerd is a little like Rain Man, and will feel responsible for accidents that happen because of his mistakes. In my mind it feels like walking a tightrope over a canyon with no net - I'm always scared of screwing up and I have this mental image of screaming people plunging to their doom. I'm not making this up, the image sometimes pops into my mind while I'm on a project.
I don't trust my coding skills, of course: there has to be a QA department with testers going over the code, proper paper trails and procedures, independent customer testing, and management that cares about quality. With all this, it still takes courage for me to work on an aircraft project.
I've met people who do and others who do not have this mindset. One FAA engineer (DER - Designated Engineer Responsible) asked about whether using a 1-byte code checksum (at startup, to verify code integrity) was sufficient and maybe 2-bytes would be safer, and *nothing else* about the project. A 2nd FAA engineer tested the system through literally all the specifications, verifying that the product did what it was supposed to do. As uncomfortable as the 2nd DER was making management, I'd much rather work with him: he understands what's at stake.
I don't think it's a case of trusting Google over Ford, or even an application company versus a car company. It's the mindset of the people making the product, and the level to which they feel responsible for the final product. It's only a little bit the mindset of management.
tl;dr: It's not the type of company, it's the type of individuals who make the product.
A friend of mine is researching power surges in the local town.
Most building codes under-specify the gauge of the neutral/return wire. For illustration, if you have three phases each rated for N amps, there is one shared neutral/return wire rated at N amps going out. At the end of the runs all phases are connected to the shared neutral line.
This is due to the nature of 3-phase electricity: the phases will tend to cancel out, so in a perfect setup you would need no neutral/return at all. Of course, the load on each phase won't exactly balance, and the load can vary as people connect/disconnect appliances, so you still need the neutral line in practice.
(Not true for house wiring, which has one or two phases coming in. Each phase has a return with the same gauge as the supply.)
This was fine when appliances were (generally) resistive loads, but nowadays switch-mode power supplies are common. When you do some math, it turns out that this type of load appears equivalent to 120 Hz power coming together at the neutral/return junction. Since 120 Hz [equivalent] power does not cancel out, the power in the return wire can be 3x as large as the building codes allow.
I've got a book explaining all this. Typically the neutral line will heat up and catch fire, breaking the circuit. Once that happens the various phases are connected without a neutral, playing hob with whatever is on those circuits and making occasional high-power ground loops and other unexpected behaviours.
Instead of a gatekeeper, I'd rather have a layer of software that automatically lies about myself (such as always giving my name as "John Doe" or my GPS location as being somewhere in the open desert near Timbuktu or something), so that not only the data hoarders don't get my personal information, but their data pool gets polluted. Bad data is much more of a problem to them than no data at all.
I've been doing that for some years.
In early September, my bank implemented a new type of authentication process. Before I could log in, it asked me a series of questions culled from the public records of my name - it said as much when it started.
The questions were multiple choice, five answers, and went like this:
In what town is 35 Granite Ave located? . Greenville . Lexington . Berwick . Nashua . Holliston
Needless to say, I've never been to 35 Granite Ave (that I can remember), never lived there, and don't have the first clue what they were on about. My "polluted public records" came back to bite me.
The bank representative couldn't help because they don't make the web page, the web page techs can't help because they outsource to a service, &c &c. It took extreme measures from one very helpful bank rep to allow me to log in, on a system which had been giving me no problems for many yeas. I'd be screwed if it were the cable, ISP, or phone company.
I'm still in favour of polluting records. If the person asking doesn't have any business knowing whatever it is they're asking, I will lie.
It looks like I'll have to start keeping track of the lies.
"Its willingness to add Windows as a second operating system underscores the lengths to which Microsoft will go to get manufacturers to carry its software.
Now that they have underscored lengths, will that... er... um... what?
willingness to add... underscores... to which... will go... to get... to carry
For the advanced student: Parse this into the canonical <subj><verb><predicate> form. 10 points.
HTC, the first company to make both Windows and Android phones, hasn’t unveiled a new Windows-based handset since June and has no current plans to release any more, said one person.
Which person is that?
Microsoft, with 3.7 percent of the market, is finding it necessary to make concessions after agreeingto acquire Nokia Oyj’s handset unit, which competes with other smartphone makers.
Apropos of nothing, which of these is the main verb?
Myerson was planning to visit Asia this month and meet with senior executives at Taoyuan, Taiwan-based HTC to discuss his proposal, one of the people said."
This is Bloomberg, right? Are they supposed to be good at writing?
Slashdot Mirror
I envision a lot of use for geolocation services.
A site could disallow access from problematic countries. For example, allowing East Asian countries read access but not post (to forums) access might cut down on sock puppet and spam replies.
Of more interest is the NSA angle. Suppose your website disallows visitors from within 50 miles of Washington, DC. Or better yet, shows sanitized links to visitors known to be associated with the government.
Any IT person will know that this is trivial to circumvent, but look at it from their point of view: Nothing they use locally will see the links they need, anyone outside the radius can't send a link into the circle for review, and setting up a tunnel (VPN &c) to a location outside the radius is a pain, and all the effort could be invalidated by the website adding the tunnel exit to the block list.
It wouldn't be hard to keep a global list similar to the SPAM blocklist sites that have lists of IPs used by government. You could download a blacklist that includes the local police station, state police, and FBI building. People could "report" access from government agencies like they currently report spam activity. It would be much easier to hold that demonstration without the police knowing your plans in advance.
Again these are not difficult to circumvent, but it makes it harder for the criminals to get in, and economy of scale is on your side: one blocklist would have to be circumvented by each agency addressed. One action on your part needs actions from multiple parties to compensate.
If there were a simple implementation of this - say, an Apache plugin that periodically grabs the blacklist - it would be a big headache for the overlords.
The US can kill an American and his teenage son, yet no one can challenge the action because they were not directly affected. If all the relatives are taken out in one action, then the US is free and clear.
We can't just protest to have unconstitutional laws removed, we have to prove they were used on us. Simply keep quiet about parallel construction and you're good to go. If the defendant says "yes they did" and the US says "no we didn't", then the constitutionality of the law makes no difference, the US is free and clear.
This thing about not challenging a law because it doesn't affect you is bullshit.
If a law is unconstitutional, then it should be possible to challenge the law on its face.
In the case of Lavabit, the government demanded, and was given, a warrant for the HTTPS private key to monitor the online actions of a couple of defendants. This would allow the FBI to monitor not only the specific defendants, but all Lavabit customers.
And I want to be totally clear about this: The government asked to install a pen trap device *and* have the private keys which would have allowed it to monitor all Lavabit customers.
(Unlike phone companies, E-mail providers are under no legal obligation to make surveillance easy, or even possible, by the government.)
Third parties have a duty to assist law enforcement, but that duty does not extend "regardless of the burden involved". The ACLU argument is that giving over the private keys would have completely destroyed the Lavabit business, which was an unreasonable burden to take in assisting law enforcement.
You do when they have a warrant.
Just saying "You do when they have a warrant" is no longer sufficient. There's ample evidence that judicial oversight has been compromised by the FISA court et al., and this is a particularly strong case of government overreach.
You can't take warrants at face value any more.
According to news reports, there are around 1000 analysts at NSA engaged in surveillance. Let's assume half of them are looking at foreign traffic and half at domestic traffic. That's 500 analysts for 350 million population, or 1 analyst for every 700,000 people. What makes you think you are special enough to deserve their attention?
Okay, let's look at those statistics more closely.
500 analysts for 350 million people continuously is 500 analysts for roughly 1 million people per day each year, or roughly 1 analyst is spending an entire day looking at 2,000 people. Each year. So there's a 1-in-2,000 chance that sometime this year, an analyst will be pawing through your online behaviour.
(Of course, if you assume that the analyst spends 1 hour on each person, it drops to 1-in-250 chance that sometime during the year you will be "analyzed" by an NSA agent.)
Now consider the power of computers. Is it reasonable to think that 1 computer could collect and analyze the E-mail and online speech of 2,000 people in a single day of compute time? Assuming you put certain keywords in your online text ("I'm going to kill some time this afternoon by watching the presidential debate"), how likely do you think it will be that you win the 1-in-250 chance?
Let's add in ambiguous laws. The recent trend is not to charge people with doing harm, but conspiracy for doing harm. One recent news report told of a couple of people charged with "conspiracy to join Al-Qaeda". Note that these two people didn't do a terorrist act, they didn't contribute to a terrorist group, and they weren't even a member of a terrorist group. They were talking about joining a terrorist group. People are commonly charged with "conspiracy to grow marijuana" (google has many links).
We've reached the point where you can be arrested when no overt crime has been committed.
There's a recent news story where, for the first time, the DOJ is informing a defendant that they used NSA/warrant-less surveillance to gather evidence. They used mass surveillance to get enough probable cause to apply for a real warrant which resulted in evidence of a crime.
The important bit of the previous is that the DOJ was conflicted about revealing this information. The prosecutor felt that it was only a "procedural decision", since no evidence from the mass-surveillance warrant would be introduced at trial. (A couple of lawyers in the DOJ argued for disclosure.)
All evidence indicates that they analyze everyone's online presence all the time, and use that information to pick-and-choose people for prosecution when no overt crime has been committed.
Sock puppet, begone!
How to Build a Happier Brain:
http://www.youtube.com/watch?v=y0V4TZAyd8I
That's older technology. You can view the new technology here:http://www.youtube.com/watch?v=sBCy-aY26zs#t=17.
Same stuff, different interpretation. Enjoy!
This thing about TrueCrypt has turned into a test-case for software security practices.
Loosely, the security audit has three sections: 1) Look at the sources, 2) Look at the license agreement, 3) Verify that the binaries match the sources.
Item #3 above seems to be a generic problem: assuming that the source is correct, how can people verify that binaries were compiled from the source? Any minor change to the compiler can generate different instructions for the same code, and many compilers insert time-stamps and signatures (login of user running the compiler) into the final executable. This makes comparing binaries a problem.
Is there a better way to solve this? For example, could the GNU compiler digitally sign an executable, so that the executable could be verified by a web app on the GNU foundation page? (Generate a signature from source+binary and include it in the executable, for example.)
Can someone come up with a good solution for executable integrity?
Look, nobody likes taxes, licensing restrictions, having to clean your car, or requiring you don't just hang out at the airport where people will pay tons of money.
The reasons we have those is that unlicensed cabs were a big problem.
Unlicensed cabs were a big problem because cabs and customers were not regulated.
The government stepped in and cleaned up the cabs, enforcing a standard of quality control of the cabbies but not the customers. It's the "regulation" model, and it was appropriate for its time, but it only addressed half the issue: a customer could jump out and run away without paying, could slit the seat, could vomit in the seat, or do other unsavory things.
Over time the regulation became less enforced, watered down, corrupt, and fewer people cared. This has resulted in the situation we have now, where many cabs are filthy and disgusting, the cabbie will screw you out of money in various ways (jimming the meter, taking the long route, &c), and it's not particularly safe.
In game theory terms, it's two kids dividing a cake: mom tells one kid to divide the cake equally, then leaves.
With the rise of ubiquitous communication we can now go to a newer model: both cabbies and customers can be vetted by the system. The cabbies are reviewed by the feedback of customers, and the customers are reviewed by the cabbies. Anyone who slits a seat or vomits will get a bad review and won't have access to the drivers in the future. Anyone who drives a filthy car will get a bad review and not have access to passengers in the future.
The game-theory model is different. Instead of one side promising to obey regulation, it's two sides regulating each other. It's the "one child divides the cake, the other child chooses which piece to eat" model.
This is an example of bad regulation which stifles innovation. Cab regulation ensured quality and was done with the best of intentions, but it's been subverted and there's now a better way.
We should embrace the better way.
There's a lot of push towards I/O systems that are more convenient for the manufacturers, but less so for the end consumer.
The first keyboards were heavy and had tactile feedback. If you fumbled a key typing in your password, you knew whether it entered because you could feel the "click". Nowadays the keyboard is lighter than a paperback and there's no feedback - accidentally brush a key with your finger and you have to look (for non-password text entry) or start over.
Twist knobs are highly intuitive, especially when coupled with feedback. Twist a knob and see the hands of the clock move, or see the numbers change. Control the speed natively, and if you go too far it's obvious how to back up. Nowadays we have buttons to tap, incrementing the count by 1 each time. Tap 50 times to set the minute display, and if you go too far you have to go all the way around again. This was done largely because buttons are easy to fabricate (using PCB contacts), not because they are inherently better.
Modern typing is done on the display (phone, surface), so not only don't you have tactile feedback you can't feel the boundaries of the keys, and your fingers mask the key display. And it's really tiny - in order to access all the keys you have to type extra keys that switch between keyboards (upper/lower/symbol). Again, it was done for ease of manufacturing, not ease of use.
Is the ribbon any easier than, for example, cascading menus? The problem with the Windows original menu system was that every application put their commands in the top-level Start->Programs folder, leading to start menus containing hundreds of links. (I take the time to move StartMenu command links into subfolders by type, which makes it much easier, but on my dad's computer it's impossible to find anything.)
Ever since minority report people have been touting the wonders of air-gesture input, and that it is the next "big thing", but is it better? (Actually, I remember it from Johnny Mnemonic, 7 years earlier.) Seems like this is just something that's easier to manufacture, but not easier to use. Sure, the customer will be able to do everything they could do with a mouse/keyboard, but more slowly, less conveniently, and with lots of frustration. That's an externality to the manufacturers, but it's better for them because they don't have to build in a touch interface. Probably [electrically] more reliable, too.
Is this really progress? I wonder what Edward Tufte would say about modern interfaces.
I nodded this in the firehose because it looked interesting.
There's not much information in the linked article. Can anyone give us more info? Anyone who reads Finnish care to comment on the source - is it reliable, are the researchers legitimate?
Self-awareness is one of those terms that doesn't have a constructive definition, so it's not a good metric.
In a literal sense the meaning is "has sensory input relating to self", in which case my laptop is self-aware: it knows the amount of charge in it's battery and will beg me (by "beeping" plaintively) for more energy. (Usually to no avail - I'm a right bastard.)
Most of the literature takes the term to mean "conscious", which also doesn't have a good definition. The best I can come up with is that a machine is "conscious" when it has an internal model of its environment, with itself as a separate entity. Thus, humans have mirror neurons which allow them to learn by seeing someone else's mistake, can recognize themselves in a mirror, &c. Great apes recognize themselves in a mirror, while monkeys do not.
It's not clear whether humans have free will at all. At some level, an intelligence has to perform the risk/reward calculation and take the best outcome for action. Most incidents of bad outcome are due to random chance, improperly calculating the odds, not having enough knowledge, or insufficient perspective. (A soldier diving on a grenade is working to benefit his people, which is intelligence at the genetic level - the genes in his society are more likely to survive even when the individual is sacrificed.)
We do know that our universe does not preclude free-will. Randomness, which is an aspect of every sensory input we have, is necessary for free will, but no one knows whether it is sufficient. If there were no randomness, then an outcome could be completely predicted - including (in theory) human decisions.
No one really knows if there are other requirements for free will, or what the list of satisfiable conditions should be.
Perhaps we should start by completely defining what "free will" even means.
I know this is satire, but fuck you
Dude, really?
Check out John Cleese's lecture on creativity.
Then tell me if you're one of the people who believe in absolute solemnity for certain subjects, that they cannot be joked about in any way.
By way of illustration, here's a parody of torture.
Joe Sixpack and Lisa Liberal don't seem to care.
Some people have been studying the phenomenon of "upheval" in it's generic form. It's spawned a lot of studies/papers and even popular books, viz: The Tipping Point.
The overall summary is that you can't just point out how bad something is, you have to give people an action they can take to help fix the problem.
There is widespread distrust, anger, and annoyance at the NSA due to the revelations. There's no public outrage because there's really nothing anyone can do. "Joe Sixpack" has no actions to take: voting doesn't help, writing congresscritters doesn't help, even public mass demonstrations don't seem to help. What you are seeing is Learned Helplessness: an animal doesn't take actions to help themselves, because they're convinced that the actions will have no effect.
Consider the recent history of cell phones or music distribution: people were complaining that cell phones were a walled ecology with no innovation and poor functionality. You had to get carrier approval to run a program on a cell phone, and they would only allow the simplest, meager functionality. You were lucky if your carrier allowed you to have tetris.
People complained that if you wanted music, you had to purchase a physical CD, for an ensemble collection and for an exorbitant fee. Usually you had to purchase an entire CD for a single song you liked.
As soon as an option was given, people flocked to the new systems in droves, uptake was very fast.
Make secure E-mail easy to use with trivial installation and the situation will change overnight. There will be a flood of new users.
Everyone hates the situation, but for most people there's nothing they can do about it.
What no one seems to get is that no one in federal government(*) cares what's right for society, for the people, or even for their own survival.
The purpose of government is to siphon funds away from individuals and give it to corporations. That's the length and breadth of it, there are no other considerations.
The purpose of airport security is to give money to scanner companies. (Oh, these scanners don't work? We'll throw them out and purchase your newer model.) The purpose of Obamacare is to give money to insurance agencies. The purpose of Obamaphone is to give money to the phone companies, the purpose of military spending is to give money to military contractors, and the purpose of the war on drugs is to give money to private prisons.
Every time one of these "this is the right move, but the government is doing the opposite" articles come up (one or two a day, it seems) it's framed in terms of an isolated, poor choice within a sea of government actions that are generally benevolent to the population and make our life better.
It's not an isolated incident, everything the federal government does has one purpose and makes sense within that framework. Occasionally it also benefits the people, but that's more happenstance than plan. It's the "random guess is occasionally correct" principle.
Federal government is a runaway train that's going to crash and burn, taking the country down with it. We can let this happen, or we can curtail it beforehand. Either way works, but fixing it beforehand would seem a better plan.
*Note: I'm making a distinction between federal and state government. Most of the federal government could disappear without negative impact on the people. And yes, I said "most".
That can't be!
Everyone knows that the Tea party is a bunch of comic, laughable clowns with no grounding in reality. I mean, why else would they be so thoroughly lampooned using derogatory terms and snarky, dismissive comments.
Even Obama himself (praise be his name) mocks them.
It can't be true... can it?
"PNAS authors must provide the National Academy of Sciences with an exclusive license to publish their work"
Why would anyone do this?
Personally, I've found more fugly code turds in various closed source projects than I've touched than in the open source world.
Is code the only aspect of note in an open source project?
How is the project named? Is it something reminiscent of the function (like PaintShop Pro, Photoshop, Internet Explorer) or something entirely random, forcing more cognitive load on an uninformed user (Gimp, Firefox, Juice)? Does it have a newish, edgy name to give it that extra sizzle (pantyshot, upskirt).
How is the project configured? Is is a list of poorly-written technobabble? Does the installation instructions begin with the history of the project (of which I am not interested), require other packages which I have to research and choose, does it require cryptic installation actions and complex setup that has to be done by hand?
How does the project look? Are the panels laid out with ease-of-use in mind, or they just show everything and "let the user arrange them as they like"? Is the text font and color scheme appropriate, or is it default, the user can choose the one they like?
Are there lots of icons for every little action, no matter how small (the "kitchen sink" philosophy), or is there a well-chosen subset that balances functionality with ease-of-use? Do the icon shapes bring the function to mind, or are they more-or-less random shapes that rely on popups to tell the user what they do?
Is the documentation well-written by people who are good at explaining things, or is it just a wiki editable by anyone, maintained by the users, with no real structure?
Has the code been tested by someone who is not the lead coder (and not the users)? Does the project use regression tests?
Yeah, nice code you've got there. If that's all I wanted in a product, yours would be a slam dunk.
Apparently Mr. Moalin once missed a telephone call from "Aden Hashi Ayrow, the senior al Shabaab leader," which makes it likely that a little more was going on than merely the donation of "a small sum of money."
Is this important?
He's claiming not that the evidence is wrong, he's claiming that it was collected illegally.
It's often been said that the defense of freedom is the defense of scoundrels (H. L. Mencken). We believe that a kiddie porn merchant has the right to a fair trial, the KKK has the right to assemble, and Rosa Parks has the right to sit in the front of the bus.
Should we base the legitimacy of rights and freedoms on the character of the accused party?
Once again, possibility is not the same as probability.
Yes, it's possible that the elevated radiation levels will cause problems. Now, what's the risk, and what's the tradeoff between mitigating *that* risk versus mitigating some *other* risk?
Security is a tradeoff, always. The value of something is not the face value, but the face value times the probability of occurrence.
So if the probability of damage (say, the number of people getting cancer from going to the event) times the value of damage (taken informally as $1 million per human life lost, but depends on estimates and philosophy) is higher than other foreseeable risks, then we should address the problem.
Risks shouldn't be ignored, just compared to other risks. If the utility losses for other risks are higher, then we should spend our finite resources on the other risks first.
How much risk utility is embodied in this problem compared to, say dying from accidentally swallowing (and choking on) a bee?
...but journalism must sell news. I suppose someone swallowing something wouldn't be very interesting.
... much better that you experience problems when you anticipate them than when everything else is going wrong, too.
So what you're saying is: if you make a 1 million dollar mistake, your response should be "Phew! At least I didn't lose 5 million!".
An outside observer might suggest that losing the $1 million is bad on its face. Mitigating the outrage by making false comparisons is the sort of thing politicians do, as a dodge for responsibility.
Should we be sanguine about these sorts of problems because they're not the worst possible scenario? Is that an acceptable excuse?
The gene pool is half empty!
No, the gene pool is twice as big as it needs to be.
This is no surprise really. Who would you trust to program a computer in charge of your life?
You trust a true nerd: Someone who is obsessive about correctness, some distance down the Asperger's spectrum, and who's convinced that the consequences of having a bug are their fault. Hygiene and dress-code are secondary.
I used to code aircraft avionics software (microcontroller stuff for altimeters, airspeed, cabin pressurization, &c). Some of my avionics-related courses asked "are you willing to be the first passenger in an aircraft running this software? raise your hand", and typically mine was the only hand showing.
There's a mindset for making safety-certified software, and not everyone has it. Most people rationalize doing a poor job by denying responsibility: the boss told them to do it, they have to feed their family, everyone else does it, and so on and so on. It's the mindset that allows the NSA get away with rights violations: no one takes responsibility at any level.
A true nerd is a little like Rain Man, and will feel responsible for accidents that happen because of his mistakes. In my mind it feels like walking a tightrope over a canyon with no net - I'm always scared of screwing up and I have this mental image of screaming people plunging to their doom. I'm not making this up, the image sometimes pops into my mind while I'm on a project.
I don't trust my coding skills, of course: there has to be a QA department with testers going over the code, proper paper trails and procedures, independent customer testing, and management that cares about quality. With all this, it still takes courage for me to work on an aircraft project.
I've met people who do and others who do not have this mindset. One FAA engineer (DER - Designated Engineer Responsible) asked about whether using a 1-byte code checksum (at startup, to verify code integrity) was sufficient and maybe 2-bytes would be safer, and *nothing else* about the project. A 2nd FAA engineer tested the system through literally all the specifications, verifying that the product did what it was supposed to do. As uncomfortable as the 2nd DER was making management, I'd much rather work with him: he understands what's at stake.
I don't think it's a case of trusting Google over Ford, or even an application company versus a car company. It's the mindset of the people making the product, and the level to which they feel responsible for the final product. It's only a little bit the mindset of management.
tl;dr: It's not the type of company, it's the type of individuals who make the product.
At the end of the runs all phases are connected to the shared neutral line.
Poor choice of words: I don't mean to say that power and return are wired together. I meant to say that all three phases share one neutral line.
Post before coffee, regret at leisure.
A friend of mine is researching power surges in the local town.
Most building codes under-specify the gauge of the neutral/return wire. For illustration, if you have three phases each rated for N amps, there is one shared neutral/return wire rated at N amps going out. At the end of the runs all phases are connected to the shared neutral line.
This is due to the nature of 3-phase electricity: the phases will tend to cancel out, so in a perfect setup you would need no neutral/return at all. Of course, the load on each phase won't exactly balance, and the load can vary as people connect/disconnect appliances, so you still need the neutral line in practice.
(Not true for house wiring, which has one or two phases coming in. Each phase has a return with the same gauge as the supply.)
This was fine when appliances were (generally) resistive loads, but nowadays switch-mode power supplies are common. When you do some math, it turns out that this type of load appears equivalent to 120 Hz power coming together at the neutral/return junction. Since 120 Hz [equivalent] power does not cancel out, the power in the return wire can be 3x as large as the building codes allow.
I've got a book explaining all this. Typically the neutral line will heat up and catch fire, breaking the circuit. Once that happens the various phases are connected without a neutral, playing hob with whatever is on those circuits and making occasional high-power ground loops and other unexpected behaviours.
Instead of a gatekeeper, I'd rather have a layer of software that automatically lies about myself (such as always giving my name as "John Doe" or my GPS location as being somewhere in the open desert near Timbuktu or something), so that not only the data hoarders don't get my personal information, but their data pool gets polluted. Bad data is much more of a problem to them than no data at all.
I've been doing that for some years.
In early September, my bank implemented a new type of authentication process. Before I could log in, it asked me a series of questions culled from the public records of my name - it said as much when it started.
The questions were multiple choice, five answers, and went like this:
In what town is 35 Granite Ave located?
. Greenville
. Lexington
. Berwick
. Nashua
. Holliston
Needless to say, I've never been to 35 Granite Ave (that I can remember), never lived there, and don't have the first clue what they were on about. My "polluted public records" came back to bite me.
The bank representative couldn't help because they don't make the web page, the web page techs can't help because they outsource to a service, &c &c. It took extreme measures from one very helpful bank rep to allow me to log in, on a system which had been giving me no problems for many yeas. I'd be screwed if it were the cable, ISP, or phone company.
I'm still in favour of polluting records. If the person asking doesn't have any business knowing whatever it is they're asking, I will lie.
It looks like I'll have to start keeping track of the lies.
"Its willingness to add Windows as a second operating system underscores the lengths to which Microsoft will go to get manufacturers to carry its software.
Now that they have underscored lengths, will that... er... um... what?
willingness to add... underscores... to which ... will go... to get ... to carry
For the advanced student: Parse this into the canonical <subj><verb><predicate> form. 10 points.
HTC, the first company to make both Windows and Android phones, hasn’t unveiled a new Windows-based handset since June and has no current plans to release any more, said one person.
Which person is that?
Microsoft, with 3.7 percent of the market, is finding it necessary to make concessions after agreeing to acquire Nokia Oyj’s handset unit, which competes with other smartphone makers.
Apropos of nothing, which of these is the main verb?
Myerson was planning to visit Asia this month and meet with senior executives at Taoyuan, Taiwan-based HTC to discuss his proposal, one of the people said."
This is Bloomberg, right? Are they supposed to be good at writing?