Yeah and those from Manchester, UK might ask - "What is that bright thing in the sky? It hurts my eyes!".
"It's the Sun, my dear" "But the Sun is the newspaper you don't let me read!" "Yes, and you're not supposed to look at either, because you might go blind".
That wikipedia article describes the _characteristics_ of that type of scattering, but it doesn't really explain WHY the light gets scattered that way.
If they get a few million votes and steadily increase share every election, even if they don't win the bigger parties might decide to adjust their policies a bit.
A lot of people say it's just a two horse race because of the "first past the post" system. Big fucking deal. That doesn't matter as long as you can influence the direction the horses are heading.
Fact is the big parties have changed over the years, so they can and do change.
If you keep voting for a party that you don't like, it's effectively saying to them "keep doing whatever you are doing". Why should they change if they keep winning most of the votes?
If they see their grip loosening, believe me, they'll do something.
Yep. The Golden rule applies- Do unto others as you would have them do unto you.
Which for the intelligent also includes wanting others to use their brains to understand that people have different tastes (not everyone is a masochist who likes being whipped).
Seriously though as per a previous slashdot story there are a fair number of players who try to "exploit" the game or play it a way the game designers probably don't expect most of their players to play. Some game designers put in some easter eggs or "special features" for such people - after all I find it interesting that while you can't run up or forward jump up the portland lighthouse island in GTA3, you can jump _backwards_ up to the lighthouse - perhaps they intentionally made that bit for players like me:).
I suppose one might classify the people who do stuff like "Quake Done Quick" as runners.
But how about the players who fist fight their way through DOOM?
The Courts could rule that he didn't really have permission. After all a reasonable person seeing those files being shared would realize that it is far more likely that someone has made a mistake than someone has intentionally given permission to the world to access those files.
Just because I leave my car unlocked with the keys in them doesn't mean you have been given permission to drive the car away. Now if someone naughty then puts a sign saying "free car" on it, and someone else drives it off, it should be a lesser charge (one should realize that to get the free car, some paperwork needs to done to transfer the ownership).
If I give you my credit card info it doesn't mean you can go around using it to do your online shopping.
I don't have the full details but another possibility is the "protected computer" is not necessarily the computer with limewire, it could be the _other_ computers (in banks etc) the guy accessed to commit fraud.
With windows you can stick to "limited users" and don't share accounts, and make it harder for someone else's limewire to suddenly share your files without your permission.
I think that's the better approach, since it makes it harder for the kid to accidentally delete/corrupt/read your files.
Not impossible of course - since they have physical access to the computer.
You might end up breeding faster and more agile yellow jackets.
We use those zappers a lot to kill mosquitoes in my house.
The trouble is, after some years of that, the current generation of mosquitoes now seem to be smaller and faster. They even bite faster - they can land, draw blood and fly away to escape in a short time.
Worse is, when I'm drifting off to sleep with the zapper nearby, the mosquitoes seem to stay away when I'm alert and waiting, but just when I am about to doze off - one or two start to attack. So I "wake" and try to swat them. I might get a few. But after that if I lie down, wait and stay alert (eyes closed) they still seem to not approach till I start to doze off again. I think future generations of mosquitoes in my area would be the ones that tend to attack when I'm asleep.
The "doze detection" might just be my imagination, but the "small and fast" is definitely true (in other places the mosquitoes are so slow - and "fluffy" that I can even slowly prod them in the air with my finger and they don't zoom off).
Even so it's bullshit. It's the level of innovation as "crack eggs" to make an omelette. Or "crack eggs" on a suitable container to stop stuff from spilling and creating a mess.
That's the problem with patents, they reward the non-innovative people who come up with innovation that's just enough to slip by some overworked/lazy/stupid patent examiner. In contrast the really innovative people are more likely to come up with stuff that people don't quite understand till 20 years later. Go google "Mother of all Demos" - watch the demo and see how much was done more than 40 years ago. What was done blazed the way for later stuff in PARC, Apple etc.
Worse - the companies that actually make stuff are likely to infringe on something - so they cross license with each other (which is either a "broken window" or a "entry barrier" in economic terms). But the Patent Trolls don't create anything except lawsuits, so they don't infringe.
You have to be careful that your O/S or hardware does not leak out that the encrypted container file is being written/read to in areas where officially there's no data. Otherwise they say: "Well where's your other password - we know you're accessing the other parts".
For example, if disk errors are logged to a logfile, and there's a "reallocated sector" in an "unofficial" area in recent times, you might be in trouble...
Also SSDs try to avoid overwriting existing data in many cases - they write the updated data in a "clean block" and leave the old data where it is since that's faster than erasing and rewriting. So that could leak out usage info as well.
This could be masked if programs like truecrypt and rubberhose rewrite different blocks of the entire container (without destroying data) at unpredictable times. but some hardware could treat overwriting with the same data differently from writing different data.
The author's suggestion (which isn't new[1]) is also vulnerable to such problems - since "Aunt May" is unlikely to be using the encrypted container.
BTW it seems to me TrueCrypt's hidden partition system is inferior to Rubberhose's - since with truecrypt, using the "official" partition could cause you problems (even if you enter the hidden partition's passphrase there could be clashes).
[1] I suggested something similar to deal with what the author is talking about nearly 2 years ago: https://bugs.launchpad.net/ubuntu/+bug/148440 I'm probably not the only one. But yes that's vulnerable, and something like the "rewrite" thing might be required.
Another source of leakage could be if you (or someone/something) copied/backed up the container file, and the cops get access to the copy or copies - then they can compare which parts were changed, and things go downhill for you.
Yet another source of leakage could be you posting on slashdot or somewhere else that you do use hidden partitions/aspects.
Lastly, the advantage of creating encrypted partitions for everyone (and making them easy to use) is more people might use them. And more people might forget the passwords to them, including judges, police officers, juries.
It's easier to claim "I was messing about with it to see what it did, and forgot the password" and be believed, if more people have done that before.
I believe if you use the nonhidden partition regularly you might screw up the hidden partition.
So if they find that you regularly use truecrypt, but the nonhidden partition appears to be not so regularly used, they might start putting a bit more pressure on you.
You have to be careful that your O/S or hardware does not leak out that the encrypted container is being written/read to in areas where it officially shouldn't be.
For example, if disk errors are logged, and you are unlucky and there's a "reallocated sector" in the "hidden partition" area in recent times, you might be screwed.
I believe SSDs try to avoid overwriting existing data in many cases - they write the updated data in a "clean block" and leave the old data where it is since that's faster than erasing and rewriting. If they find that out, your hidden partition is not so hidden.
I'm also curious to learn whether the law was worth it to the UK people in this case.
Basically: how much worse things could have been for UK people if it weren't for that law, compared to how much worse things could be for the UK people because of this law.
Quote: Sir Christopher reported that all of the 15 section 49 notices served over the year - including the two that resulted in convictions - were in "counter terrorism, child indecency and domestic extremism" cases.
I thought "counter terrorism" was fighting terrorism? So what was "child indecency" and "domestic extremism"?
> Note that this is a separate problem from the one that was addressed in this article. That was about the trustworthiness of the results, not about fairness.
That's actually what I meant by fair (no cheating). Elections don't just have to be correct/trustworthy, they have to be seen to be correct (enough).
> the government represents the Party, not the people.
Thing is, in the 2008 US Presidential election, more than 98.5% of the voters who bothered to vote, voted for candidates of one of the Two Parties. In the 2004 election, that figure was 99%.
And if the other voters that could have voted but didn't, actually voted for some other particular candidate that candidate would have won, instead of either of the Two Party candidates.
So unless the US elections have been diebolded, I'd say the Two Parties are representing the voters as well as a "first past the post" system can (which is not that well, but you have to work with what you've got).
If the voters really don't like those candidates they should really be voting for someone else.
Especially the 37-40% who just stayed at home - if they really didn't like it, perhaps they should got out there and voted for someone, or even just write "None of the above". Even if they spread their votes over the other candidates and thus don't affect that particular election, when the voters and parties realize "None of the Two" adds up to something rather significant, the next election might be rather different. Or the Two Parties will start changing to try to maintain their 98-99% "share". As it is, those voters effectively don't count, and the Two Parties know that.
If you vote for someone you don't like just to try to keep someone else out, that often sends the wrong message to the other voters. Maybe voters should just do that sort of thing every other election. e.g. election #1 - voters show preferences without trying to play that game. election #2- voters play the game based on election #1. Otherwise it just degenerates to sheep voting to decide which of two wolves gets to eat them.
FWIW, I don't think a democratic election needs any fancy systems. Stick to paper ballots, keep the counting _open_ (and thus easily monitored by "everyone" within reason). There are plenty of ways to keep it simple and safe (except for postal votes - they're a bit of a problem). Simple is good because elections don't just have to be fair, they have to be seen as fair.
If you count votes behind closed doors like in the recent Iranian election, people get the impression that it's rigged.
That's why electronic voting is stupid - either the totals are calculated effectively behind closed doors, or it's the same thing as paper voting except just a lot more expensive.
Anyway, that's why I don't have such a glowing view of much AI research.
If we want nonhuman intelligence, we already have plenty around.
At least dogs have been bred to want to make us happy. So we should just use the AI related tech to augment dogs, humans and other animals.
Then once we've settled the many long term issues and implications of AI (laws, rights, what is "human" etc), we can go create "real AI". Otherwise it's like jumping into a situation we aren't ready for yet.
I daresay there aren't as many open source software that are really polished compared to commercial software.
Most OSS developers are happy enough to get things to the point of "mostly works" or more infamously: "WorksForMe".
Of course, the extra polishing or effort rarely goes to security, since real security rarely sells, you can get away with just _claiming_ bullshit like "Unbreakable" (like Oracle did).
But really, with commercial software, you're more likely (though still not common) to have some annoying noncoder that the programmer HAS to listen to, who's standing there and saying, "Nope that's not good enough for the users, it's got to be better than that". Yes, this is not that common, but it's still more likely than for OSS.
Because with most OSS the programmer doesn't have to listen to mere noncoders or anybody - they can just say - "Not good enough for you? Go fork yourself!". Heck lots of Slashdotters say that sort of thing when people complain about OSS - "Download the source, and fix it yourself".
Is there any software/system out there to help dogs "talk"?
Not the silly "bark" translation software. I mean serious stuff - e.g. a bunch of pedals/buttons which the dog can press to say stuff (nouns verbs etc).
Stephen Hawking uses a computer system to talk to others, why not help dogs do that too?
> Only thing we haven't been able to get her to do is make pseudo-talking sounds with any consistency until she gets really frustrated.
She may need a computer + speech synthesizer to help - since she understands words, show her how its done then she might get it. Then she might freak you out even more. Don't give her Stephen Hawking's voice though, that'll be even more freaky.;)
I can't wag my nonexistent tail and I don't expect dogs to talk "normally" either.
Crows or other corvids are very smart too (smarter than chimps in some ways). Anyway, given the sorts of stuff they eat, it's probably a good idea to not eat them;).
Octopuses are also quite smart. At least one seem to have rather poor memory though - forgets after a few days and has to relearn stuff.
Hmm, I wonder what it would be like to have a girlfriend that's intelligent, generally well behaved, etc... until you show her a ball, and then she goes to "must chase ball" mode.
Slashdot Mirror
Yeah and those from Manchester, UK might ask - "What is that bright thing in the sky? It hurts my eyes!".
"It's the Sun, my dear"
"But the Sun is the newspaper you don't let me read!"
"Yes, and you're not supposed to look at either, because you might go blind".
That wikipedia article describes the _characteristics_ of that type of scattering, but it doesn't really explain WHY the light gets scattered that way.
Not stopped. Scattered. If the blue light was stopped in the sky, it wouldn't reach our eyes and the sky would look less bluish.
They don't have to win the elections.
If they get a few million votes and steadily increase share every election, even if they don't win the bigger parties might decide to adjust their policies a bit.
A lot of people say it's just a two horse race because of the "first past the post" system. Big fucking deal. That doesn't matter as long as you can influence the direction the horses are heading.
Fact is the big parties have changed over the years, so they can and do change.
If you keep voting for a party that you don't like, it's effectively saying to them "keep doing whatever you are doing". Why should they change if they keep winning most of the votes?
If they see their grip loosening, believe me, they'll do something.
Yep. The Golden rule applies- Do unto others as you would have them do unto you.
Which for the intelligent also includes wanting others to use their brains to understand that people have different tastes (not everyone is a masochist who likes being whipped).
How about "Noobs" and "Incompetents"? ;)
:).
Seriously though as per a previous slashdot story there are a fair number of players who try to "exploit" the game or play it a way the game designers probably don't expect most of their players to play. Some game designers put in some easter eggs or "special features" for such people - after all I find it interesting that while you can't run up or forward jump up the portland lighthouse island in GTA3, you can jump _backwards_ up to the lighthouse - perhaps they intentionally made that bit for players like me
I suppose one might classify the people who do stuff like "Quake Done Quick" as runners.
But how about the players who fist fight their way through DOOM?
The Courts could rule that he didn't really have permission. After all a reasonable person seeing those files being shared would realize that it is far more likely that someone has made a mistake than someone has intentionally given permission to the world to access those files.
Just because I leave my car unlocked with the keys in them doesn't mean you have been given permission to drive the car away. Now if someone naughty then puts a sign saying "free car" on it, and someone else drives it off, it should be a lesser charge (one should realize that to get the free car, some paperwork needs to done to transfer the ownership).
If I give you my credit card info it doesn't mean you can go around using it to do your online shopping.
I don't have the full details but another possibility is the "protected computer" is not necessarily the computer with limewire, it could be the _other_ computers (in banks etc) the guy accessed to commit fraud.
With windows you can stick to "limited users" and don't share accounts, and make it harder for someone else's limewire to suddenly share your files without your permission.
I think that's the better approach, since it makes it harder for the kid to accidentally delete/corrupt/read your files.
Not impossible of course - since they have physical access to the computer.
You might end up breeding faster and more agile yellow jackets.
We use those zappers a lot to kill mosquitoes in my house.
The trouble is, after some years of that, the current generation of mosquitoes now seem to be smaller and faster. They even bite faster - they can land, draw blood and fly away to escape in a short time.
Worse is, when I'm drifting off to sleep with the zapper nearby, the mosquitoes seem to stay away when I'm alert and waiting, but just when I am about to doze off - one or two start to attack. So I "wake" and try to swat them. I might get a few. But after that if I lie down, wait and stay alert (eyes closed) they still seem to not approach till I start to doze off again. I think future generations of mosquitoes in my area would be the ones that tend to attack when I'm asleep.
The "doze detection" might just be my imagination, but the "small and fast" is definitely true (in other places the mosquitoes are so slow - and "fluffy" that I can even slowly prod them in the air with my finger and they don't zoom off).
Even so it's bullshit. It's the level of innovation as "crack eggs" to make an omelette. Or "crack eggs" on a suitable container to stop stuff from spilling and creating a mess.
That's the problem with patents, they reward the non-innovative people who come up with innovation that's just enough to slip by some overworked/lazy/stupid patent examiner. In contrast the really innovative people are more likely to come up with stuff that people don't quite understand till 20 years later. Go google "Mother of all Demos" - watch the demo and see how much was done more than 40 years ago. What was done blazed the way for later stuff in PARC, Apple etc.
Worse - the companies that actually make stuff are likely to infringe on something - so they cross license with each other (which is either a "broken window" or a "entry barrier" in economic terms). But the Patent Trolls don't create anything except lawsuits, so they don't infringe.
You have to be careful that your O/S or hardware does not leak out that the encrypted container file is being written/read to in areas where officially there's no data. Otherwise they say: "Well where's your other password - we know you're accessing the other parts".
For example, if disk errors are logged to a logfile, and there's a "reallocated sector" in an "unofficial" area in recent times, you might be in trouble...
Also SSDs try to avoid overwriting existing data in many cases - they write the updated data in a "clean block" and leave the old data where it is since that's faster than erasing and rewriting. So that could leak out usage info as well.
This could be masked if programs like truecrypt and rubberhose rewrite different blocks of the entire container (without destroying data) at unpredictable times. but some hardware could treat overwriting with the same data differently from writing different data.
The author's suggestion (which isn't new[1]) is also vulnerable to such problems - since "Aunt May" is unlikely to be using the encrypted container.
BTW it seems to me TrueCrypt's hidden partition system is inferior to Rubberhose's - since with truecrypt, using the "official" partition could cause you problems (even if you enter the hidden partition's passphrase there could be clashes).
[1] I suggested something similar to deal with what the author is talking about nearly 2 years ago: https://bugs.launchpad.net/ubuntu/+bug/148440
I'm probably not the only one. But yes that's vulnerable, and something like the "rewrite" thing might be required.
Another source of leakage could be if you (or someone/something) copied/backed up the container file, and the cops get access to the copy or copies - then they can compare which parts were changed, and things go downhill for you.
Yet another source of leakage could be you posting on slashdot or somewhere else that you do use hidden partitions/aspects.
Lastly, the advantage of creating encrypted partitions for everyone (and making them easy to use) is more people might use them. And more people might forget the passwords to them, including judges, police officers, juries.
It's easier to claim "I was messing about with it to see what it did, and forgot the password" and be believed, if more people have done that before.
I believe if you use the nonhidden partition regularly you might screw up the hidden partition.
So if they find that you regularly use truecrypt, but the nonhidden partition appears to be not so regularly used, they might start putting a bit more pressure on you.
You have to be careful that your O/S or hardware does not leak out that the encrypted container is being written/read to in areas where it officially shouldn't be.
For example, if disk errors are logged, and you are unlucky and there's a "reallocated sector" in the "hidden partition" area in recent times, you might be screwed.
I believe SSDs try to avoid overwriting existing data in many cases - they write the updated data in a "clean block" and leave the old data where it is since that's faster than erasing and rewriting. If they find that out, your hidden partition is not so hidden.
This is not about root passwords. This is an admin password to a blog system or CMS.
You're going to need admin accounts as long as you want to have different classes of users and have certain users manage some stuff.
Does that explain the "counter terrorism" charges too?
I'm also curious to learn whether the law was worth it to the UK people in this case.
Basically: how much worse things could have been for UK people if it weren't for that law, compared to how much worse things could be for the UK people because of this law.
Quote:
Sir Christopher reported that all of the 15 section 49 notices served over the year - including the two that resulted in convictions - were in "counter terrorism, child indecency and domestic extremism" cases.
I thought "counter terrorism" was fighting terrorism? So what was "child indecency" and "domestic extremism"?
Have a reasonably popular O/S give everyone crypto and an encrypted "container" whether they are going to use it or not.
However, it has to be done in a way where you can use that container without leaving evidence behind that you have used it.
Something like: https://bugs.launchpad.net/ubuntu/+bug/148440
Except better :).
Then that's what I call creating real plausible deniability.
> Note that this is a separate problem from the one that was addressed in this article. That was about the trustworthiness of the results, not about fairness.
That's actually what I meant by fair (no cheating). Elections don't just have to be correct/trustworthy, they have to be seen to be correct (enough).
Some sort of electronic voting can actually be technically decent: http://www.youtube.com/watch?v=ZDnShu5V99s
But paper ballots done right won't require as much educating of Joe Sixpack :).
> the government represents the Party, not the people.
Thing is, in the 2008 US Presidential election, more than 98.5% of the voters who bothered to vote, voted for candidates of one of the Two Parties. In the 2004 election, that figure was 99%.
And if the other voters that could have voted but didn't, actually voted for some other particular candidate that candidate would have won, instead of either of the Two Party candidates.
So unless the US elections have been diebolded, I'd say the Two Parties are representing the voters as well as a "first past the post" system can (which is not that well, but you have to work with what you've got).
If the voters really don't like those candidates they should really be voting for someone else.
Especially the 37-40% who just stayed at home - if they really didn't like it, perhaps they should got out there and voted for someone, or even just write "None of the above". Even if they spread their votes over the other candidates and thus don't affect that particular election, when the voters and parties realize "None of the Two" adds up to something rather significant, the next election might be rather different. Or the Two Parties will start changing to try to maintain their 98-99% "share". As it is, those voters effectively don't count, and the Two Parties know that.
If you vote for someone you don't like just to try to keep someone else out, that often sends the wrong message to the other voters. Maybe voters should just do that sort of thing every other election. e.g. election #1 - voters show preferences without trying to play that game. election #2- voters play the game based on election #1. Otherwise it just degenerates to sheep voting to decide which of two wolves gets to eat them.
FWIW, I don't think a democratic election needs any fancy systems. Stick to paper ballots, keep the counting _open_ (and thus easily monitored by "everyone" within reason). There are plenty of ways to keep it simple and safe (except for postal votes - they're a bit of a problem). Simple is good because elections don't just have to be fair, they have to be seen as fair.
If you count votes behind closed doors like in the recent Iranian election, people get the impression that it's rigged.
That's why electronic voting is stupid - either the totals are calculated effectively behind closed doors, or it's the same thing as paper voting except just a lot more expensive.
Anyway, that's why I don't have such a glowing view of much AI research.
If we want nonhuman intelligence, we already have plenty around.
At least dogs have been bred to want to make us happy. So we should just use the AI related tech to augment dogs, humans and other animals.
Then once we've settled the many long term issues and implications of AI (laws, rights, what is "human" etc), we can go create "real AI". Otherwise it's like jumping into a situation we aren't ready for yet.
I daresay there aren't as many open source software that are really polished compared to commercial software.
Most OSS developers are happy enough to get things to the point of "mostly works" or more infamously: "WorksForMe".
Of course, the extra polishing or effort rarely goes to security, since real security rarely sells, you can get away with just _claiming_ bullshit like "Unbreakable" (like Oracle did).
But really, with commercial software, you're more likely (though still not common) to have some annoying noncoder that the programmer HAS to listen to, who's standing there and saying, "Nope that's not good enough for the users, it's got to be better than that". Yes, this is not that common, but it's still more likely than for OSS.
Because with most OSS the programmer doesn't have to listen to mere noncoders or anybody - they can just say - "Not good enough for you? Go fork yourself!". Heck lots of Slashdotters say that sort of thing when people complain about OSS - "Download the source, and fix it yourself".
Hmm, strange. I'd have thought someone would have done it already. Even if it's just by some researcher.
After all, some people claim they can teach dogs to read:
http://www.amazon.com/Teach-Your-Bonnie-Bergin-Ed-D/dp/076792245X
If they can really read, "speaking" is not such a big problem nowadays.
Is there any software/system out there to help dogs "talk"?
Not the silly "bark" translation software. I mean serious stuff - e.g. a bunch of pedals/buttons which the dog can press to say stuff (nouns verbs etc).
Stephen Hawking uses a computer system to talk to others, why not help dogs do that too?
> Only thing we haven't been able to get her to do is make pseudo-talking sounds with any consistency until she gets really frustrated.
;)
She may need a computer + speech synthesizer to help - since she understands words, show her how its done then she might get it. Then she might freak you out even more. Don't give her Stephen Hawking's voice though, that'll be even more freaky.
I can't wag my nonexistent tail and I don't expect dogs to talk "normally" either.
Crows or other corvids are very smart too (smarter than chimps in some ways). Anyway, given the sorts of stuff they eat, it's probably a good idea to not eat them ;).
Octopuses are also quite smart. At least one seem to have rather poor memory though - forgets after a few days and has to relearn stuff.
http://www.pitara.com/discover/earth/online.asp?story=111
Hmm, I wonder what it would be like to have a girlfriend that's intelligent, generally well behaved, etc... until you show her a ball, and then she goes to "must chase ball" mode.