I've definitely got to put in another vote for SessionSaver. Though support is dicey (is someone maintaining this code?), it's a necessary feature for those of us who heavily use Firefox's tabs. Even when Firefox is stable, SessionSaver saves you from restoring your tabs when the OS crashes/must be rebooted/etc.
"in the united states, it's legal to sell armour-piercing ammuniction -- bullets whose sole purpose of design is to go through bullet proof vests; obviously a device designed to kill or maim human beings. the manufacturers to do not even make the pretense of proposing other uses for said ammunition. this activity is all fine and legal."
Good point, but isn't a plain-old bullet's sole purpose of design to go through human flesh? It's definitely designed to kill or maim human beings. I guess that killing or maiming human beings isn't illegal in and of itself (I guess it could be self-defense), but there's some pretty obvious "infringing uses." I'd say fingerprint-proof gun handles are a better example... since they're only of use to people who are doing things with their guns they shouldn't be.
Does this mean it's time for an International Obfuscated Python Code Contest?
(only slightly obfuscated reference to the International Obfuscated C Code Contest )
Unlike the author of the article, I *am* a techie. I'm familiar with various methods of recognizing and filtering spam, detecting spoofed return addresses, blacklisting, whitelisting, challenge-response, signing, etc. But the best solution available to every recipient of spam right now is to click those links (apologies to the author of an article arguing the same point, if someone could link it up I'd appreciate it).
Consider this: spammers will do anything and everything they can to make their emails seem like they're not spam. Until we achieve new milestones in AI (and perhaps not even then), your PC will not know whether an incoming email is a recommendation from an old friend who you ran into on the street, or an unsolicited "recommendation" from a paid spammer. We need a system for punishing email that lets the recipient decide which emails are spam.
Second, the only part of a spam that cannot be forged, spoofed, or hidden is the URL (or other contact point) of the web site advertised. Thus the only dependable way to punish the one responsible for the spam is to create cost for that site. This is as simple as clicking the link (preferably many, many times - an automated script would come in useful). This will drive up the hosting costs of the site, and drive down the percentage of site visitors who buy anything. While you're at it, you can throw in fake names, addresses, credit cards, etc. to further tax their payment processing system (anything that makes the site request an authentication from the credit card company would be a good way to push up their costs).
Right now, spammers enjoy a very low, but positive, response rate to their spams. Imagine if, for every idiot shortsighted enough to provide his credit card and contact information to a complete stranger with dodgy business tactics, there were several users who received this spam and loaded the site without any intent to purchase. Quickly, spam's economic equation would turn unprofitable. While a little bit of simple technology (a script that mimics a browser hitting the site a few hundred times) would help, you could also just repeatedly hit those links and ignore the content (might make for some interesting stories if an IT department sees proxy logs, but if everyone started doing it I'm sure they'd catch on, especially if you use a script that doesn't actually display the likely-objectionable content).
It's no panacea, but having 1% of recipients punish the advertised web site will be more effective than having 99% of the recipients filter out the spam entirely... and there's no reason that filtering and punishing couldn't peacefully coexist as anti-spammer tactics.
So stop hitting delete, and start clicking those links!
So who's going to be the first to study the underwater propagation of 802.11 signals? Unfortunately, I don't imagine that a Pringles can would last for too long underwater... then there's the added problem of keeping the fish from moving in and calling it home....and if Verizon was managing the hotspots, would they put buoys out to see with "You're Hot" signs, like the phone booths in NYC?
Re:Fitness Measurement for the GA? Here it is...
on
Learning Robots
·
· Score: 1
...and how does the snake know how far it moved? The fun thing about GAs is that they're very creative when it comes to solving the problem posed to them - so creative that they often find ways to "cheat." For example, an early GA experiment that tried to evolve "creatures" (composed of rectilinear solids with simple joints) that would move in their simulated environment. First, individuals evolved that took advantage of a conservation of momentum bug in the environment's physics package: they hit themselves to create momentum. When this bug was fixed, creatures evolved to start the simulation in a very tall, thin orientation, and simply fell over when the simulation began. Since the fitness equation just calculated movement of the creature's center of mass, this was an efficient strategy.
I have two points here: 1) GAs meant to control robots should be run onboard the robot, including the selection steps. Virtual environments are only crude models of the real world. 2) Be careful what you wish for. If the GA fitness evaluation procedure is not very carefully defined, and definitively measureable, the GA will likely come up with solutions that do not produce the desired behavior.
Fitness Measurement for the GA?
on
Learning Robots
·
· Score: 2, Interesting
This article is unclear on the nature of the fitness measurement used by the GA. The following quote leads one to believe that the GA is running onboard the snake's processors (rather than in simulation):
"Once the robot was mobile, the team disabled some of its segments to see if it could adapt to injury. Initially it was immobilised, says Bentley, but as the GA continued to try to improve the locomotion, it gradually worked out how to move again, albeit more awkwardly and with an ungainly, dragging gait - but it was still good enough to get the robot to its destination."
If the GA is running onboard the robot, then how does it know whether it's successfully moving (let alone in the right direction)? Is there a sensor that's providing information on distance to a target?
Slashdot Mirror
I've definitely got to put in another vote for SessionSaver. Though support is dicey (is someone maintaining this code?), it's a necessary feature for those of us who heavily use Firefox's tabs. Even when Firefox is stable, SessionSaver saves you from restoring your tabs when the OS crashes/must be rebooted/etc.
So mysterious that no one can think of any posts?
Does this mean it's time for an International Obfuscated Python Code Contest? (only slightly obfuscated reference to the International Obfuscated C Code Contest )
My friend just got his this week, so it's legit. Can't help with the iPaq or IR/headphone jack adapter though.
Quite the contrary, electronic grounding reduces potential to zero.
Your job could get reallocated across the grid based on its needs at any given moment. ... we already have this. They call it offshoring.
Consider this: spammers will do anything and everything they can to make their emails seem like they're not spam. Until we achieve new milestones in AI (and perhaps not even then), your PC will not know whether an incoming email is a recommendation from an old friend who you ran into on the street, or an unsolicited "recommendation" from a paid spammer. We need a system for punishing email that lets the recipient decide which emails are spam.
Second, the only part of a spam that cannot be forged, spoofed, or hidden is the URL (or other contact point) of the web site advertised. Thus the only dependable way to punish the one responsible for the spam is to create cost for that site. This is as simple as clicking the link (preferably many, many times - an automated script would come in useful). This will drive up the hosting costs of the site, and drive down the percentage of site visitors who buy anything. While you're at it, you can throw in fake names, addresses, credit cards, etc. to further tax their payment processing system (anything that makes the site request an authentication from the credit card company would be a good way to push up their costs).
Right now, spammers enjoy a very low, but positive, response rate to their spams. Imagine if, for every idiot shortsighted enough to provide his credit card and contact information to a complete stranger with dodgy business tactics, there were several users who received this spam and loaded the site without any intent to purchase. Quickly, spam's economic equation would turn unprofitable. While a little bit of simple technology (a script that mimics a browser hitting the site a few hundred times) would help, you could also just repeatedly hit those links and ignore the content (might make for some interesting stories if an IT department sees proxy logs, but if everyone started doing it I'm sure they'd catch on, especially if you use a script that doesn't actually display the likely-objectionable content).
It's no panacea, but having 1% of recipients punish the advertised web site will be more effective than having 99% of the recipients filter out the spam entirely... and there's no reason that filtering and punishing couldn't peacefully coexist as anti-spammer tactics.
So stop hitting delete, and start clicking those links!
So who's going to be the first to study the underwater propagation of 802.11 signals? Unfortunately, I don't imagine that a Pringles can would last for too long underwater... then there's the added problem of keeping the fish from moving in and calling it home. ...and if Verizon was managing the hotspots, would they put buoys out to see with "You're Hot" signs, like the phone booths in NYC?
I have two points here: 1) GAs meant to control robots should be run onboard the robot, including the selection steps. Virtual environments are only crude models of the real world. 2) Be careful what you wish for. If the GA fitness evaluation procedure is not very carefully defined, and definitively measureable, the GA will likely come up with solutions that do not produce the desired behavior.
"Once the robot was mobile, the team disabled some of its segments to see if it could adapt to injury. Initially it was immobilised, says Bentley, but as the GA continued to try to improve the locomotion, it gradually worked out how to move again, albeit more awkwardly and with an ungainly, dragging gait - but it was still good enough to get the robot to its destination."
If the GA is running onboard the robot, then how does it know whether it's successfully moving (let alone in the right direction)? Is there a sensor that's providing information on distance to a target?