Berkeley has a celebrated Center for Human-Compatible AI (CHAI) (https://humancompatible.ai/), so of course Stanford must have an "Institute for Human-Centered AI" (HAI). Jeez, people, innovate!
You sound very confident, which is really bad when you're wrong.
First, let's take apart this "work at cost" thing. The cost for the publisher is virtually 0. Authors work free or pay. Reviewers work free. There's some administration, which $0.10 per download would easily cover (and then some). Instead they charge $35.95 per download (UC presumably pays less).
You're right that uploading papers to a free, popular pre-print server like arXiv is a great thing to do, except top Elsevier journals will reject your paper if you do.
So how can Elsevier afford to be expensive and get exclusive rights? That's the rent it extracts from publishing prestigious journals. Deciding which journal is prestigious is a hard coordination game. Even with a lot of effort (which researchers at UC and worldwide are exerting in various ways), it takes time to change the status quo.
UC simply feels, likely correctly, that the tables have turned far enough that Elsevier can now be taken out of the game.
Ok, so I'm not allowed to set my browser to not download ads. I guess I'm also not allowed to prevent ads from being displayed on screen? Am I allowed not to look? Or must it also reach my eyes? Am I allowed not to pay attention? Or must I let it into my brain? My mind? My soul? My very existence? Why not just force me to buy the damn thing you're promoting and get this over with?
Google is waging war on our psyche. This here is intelligence gathering, to better identify weak spots in our defenses, in preparation for an escalation of their assault.
"The first official RoboCup games and conference was held in 1997 with great success. Over 40 teams participated (real and simulation combined), and over 5,000 spectators attended." [http://www.robocup.org/a_brief_history_of_robocup]
By "real and simulation" they mean that AI has been playing soccer for more than 20 years, both in simulation (as in TFA) and in real, physical robots. Welcome to the world of AI research, South Korean!
(To be fair, it's probably some reporter's snafu, rather than a researcher's.)
There's a CC camera at every street corner, and soon on every person. If you've resigned to being watched whenever you're online, you've resigned to being watched nearly always.
The analogy between the government and a drug cartel is false, because (astonishingly) the government has much, much more power.
The most important mitigation of the insider threat is the monitoring, logging and auditing of insider activity by other insiders. While a court order forces the hands of *all* insiders, it's that much less likely that a drug cartel will bribe or kidnap the families of multiple Guavabit employees.
The real answer is completely different. Emails should be court-order resistant, because we should be able to digitally extend our minds, and our minds should be court-order resistant.
We're cyberfighting cyberterrorists to cyberkill them before they cyberkill us.
More seriously:
Think of the spygames of the cold war, with the punch that you don't need to physically be in the location you're attacking.
The objectives are as diverse as they ever were: gather intel, sabotage, manipulate data and the public.
So we're targeting any device worth spying on (that is, all of them, prioritized), any infrastructure, any database, any public (foreign or otherwise).
And we (our devices, infrastructure, database and public) are being targeted by any half-assed hacker with a laptop, and several armies of fully-assed ones.
It ends when the cyberwarring governments sign a non-proliferation treaty, or at least a non-aggression pact, so that their interests shift from offense to defense.
Then you'll see companies suddenly becoming liable for their vulnerabilities, and soon after the net will be much more secure.
A warning, though: a side effect may well be a much less free net, in the same sense that The West is not as free as when it was Wild.
Slashdot Mirror
Berkeley has a celebrated Center for Human-Compatible AI (CHAI) (https://humancompatible.ai/), so of course Stanford must have an "Institute for Human-Centered AI" (HAI).
Jeez, people, innovate!
You sound very confident, which is really bad when you're wrong.
First, let's take apart this "work at cost" thing. The cost for the publisher is virtually 0. Authors work free or pay. Reviewers work free. There's some administration, which $0.10 per download would easily cover (and then some). Instead they charge $35.95 per download (UC presumably pays less).
You're right that uploading papers to a free, popular pre-print server like arXiv is a great thing to do, except top Elsevier journals will reject your paper if you do.
So how can Elsevier afford to be expensive and get exclusive rights? That's the rent it extracts from publishing prestigious journals. Deciding which journal is prestigious is a hard coordination game. Even with a lot of effort (which researchers at UC and worldwide are exerting in various ways), it takes time to change the status quo.
UC simply feels, likely correctly, that the tables have turned far enough that Elsevier can now be taken out of the game.
Ok, so I'm not allowed to set my browser to not download ads.
I guess I'm also not allowed to prevent ads from being displayed on screen?
Am I allowed not to look? Or must it also reach my eyes?
Am I allowed not to pay attention? Or must I let it into my brain? My mind? My soul? My very existence?
Why not just force me to buy the damn thing you're promoting and get this over with?
The missions aren't really that long, if they're all scheduled for a single day, March 5.
Google is waging war on our psyche. This here is intelligence gathering, to better identify weak spots in our defenses, in preparation for an escalation of their assault.
Will we ever fight back?
"The first official RoboCup games and conference was held in 1997 with great success. Over 40 teams participated (real and simulation combined), and over 5,000 spectators attended." [http://www.robocup.org/a_brief_history_of_robocup]
By "real and simulation" they mean that AI has been playing soccer for more than 20 years, both in simulation (as in TFA) and in real, physical robots.
Welcome to the world of AI research, South Korean!
(To be fair, it's probably some reporter's snafu, rather than a researcher's.)
There's a CC camera at every street corner, and soon on every person.
If you've resigned to being watched whenever you're online, you've resigned to being watched nearly always.
If the results are very surprising, "statistical significance" is not enough. They're probably still wrong.
That's Bayes 101.
The analogy between the government and a drug cartel is false, because (astonishingly) the government has much, much more power.
The most important mitigation of the insider threat is the monitoring, logging and auditing of insider activity by other insiders.
While a court order forces the hands of *all* insiders, it's that much less likely that a drug cartel will bribe or kidnap the families of multiple Guavabit employees.
The real answer is completely different.
Emails should be court-order resistant, because we should be able to digitally extend our minds, and our minds should be court-order resistant.
Half of all research papers are not worth the paper they will never be printed on.
How many peer-reviewed papers are free to read?
We're cyberfighting cyberterrorists to cyberkill them before they cyberkill us. More seriously: Think of the spygames of the cold war, with the punch that you don't need to physically be in the location you're attacking. The objectives are as diverse as they ever were: gather intel, sabotage, manipulate data and the public. So we're targeting any device worth spying on (that is, all of them, prioritized), any infrastructure, any database, any public (foreign or otherwise). And we (our devices, infrastructure, database and public) are being targeted by any half-assed hacker with a laptop, and several armies of fully-assed ones. It ends when the cyberwarring governments sign a non-proliferation treaty, or at least a non-aggression pact, so that their interests shift from offense to defense. Then you'll see companies suddenly becoming liable for their vulnerabilities, and soon after the net will be much more secure. A warning, though: a side effect may well be a much less free net, in the same sense that The West is not as free as when it was Wild.
Has anyone else reading this wondered who the "Foreign Secretary on Surveillance" is, and why he's so worried?