Mod parent up. That's a significant, and seldom mentioned, point. When communism was taken seriously as a competitor to capitalism, capitalism had to deliver a better standard of living for the average worker. That's no longer the case.
If it's the "fuse" in the "nose cone", it's probably the radar proximity fuze, used to detonate a nuclear weapon at a specific height above ground. This is essential only for ICBMs intended for use against hardened targets, where the detonation has to occur at just the right height to maximize the blast effect against something like a missile silo lid.
If you're delivering your bomb in a Ryder truck, this component is unnecessary.
The pace of change is slowing down.
Look at four 50 year periods in history.
1808 In 1808, life was pretty much like it had been for the previous thousand years. Land travel was on foot or by horse; most people never went fifty miles from their birthplace in their entire life. Heating was from burning wood; lighting from candles. Everything was made by hand.
But things were just starting to pick up steam, literally. The first locomotive was in 1804.
The very first passenger train ran in 1807. Iron was rare, and steel rarer still.
1858 Railroads connected the major cities in Europe, England, and the US east of the Mississippi. Gas lighting had appeared in cities. Some ships were steam powered. Western Union had telegraphs up and running. Factories were coal burning and steam powered. Textiles were being manufactured by power looms and were much cheaper. Iron was plentiful; steel was still rare. The first oil well was a year in the future.
1908 Major cities had electricity. Telephones were available. All commercial shipping was steam powered. The first cars were running, and the first aircraft had flown. Big hydroelectric plants at Niagara Falls were running. Steel was widely available and cheap. The first skyscrapers had been built.
An active oil industry was producing.
1958 Radio, TV, electronics, computers, and atomic power were all working. Transistor radios were available. Oil and natural gas were supplanting coal. Huge farm surpluses were a normal event in the US. The first satellites were in orbit. Large jet transports were flying. Good highway system pervasive. Vaccines for polio, tetanus, diphtheria, yellow fever. Antibiotics widely available. The problems of transportation, power, manufacturing, and agriculture had all been overcome, more than overcome, for the first time in history.
2008 Improvements over 1958, but few breakthroughs. No major new power sources.
Energy costs up during this period, for the first time in 200 years. No major new form of transportation. No major improvement in space launch technology. Some progress in biotech but no major life extension. Much progress in electronics and computers.
Ah, instant gratification. The system rates on demand, then caches for 30 days. So the first user to query a new site has to wait. On the other hand, any new site can be rated immediately; there's no wait for a crawler to get to it. Linksys is now rated, and it now moves up in the rankings. And Wikipedia is listed as non-commercial; it's in ".org", there's no conflicting info from Open Directory, and it doesn't have ads.
Besides, the "alternate operating system" thing for the WRT45G never worked very well. If you were reading over the wireless link from a fast local server, the Linux software would consistently garble packets. It only worked if the air link never bottlenecked.
Distribution, $0.90? $900 for a thousand CDs? No way, not for WalMart.
This is WalMart you're shipping to. You ship to them by the truckload, not one CD at a time.
Any in-store costs come under retail overhead, not distribution.
The promotion costs need to shrink. Maybe we'll see the labels begging for time on webcasts.
Label overhead is far too high. The labels don't really do much today except promote; they don't directly employ artists, they don't run recording studios, they don't manufacture CDs, and they don't do physical distribution and warehousing. That's all outsourced. But management overhead hasn't been cut accordingly.
As the WalMart VP says: "The labels price things based on what they believe they can get -- a pricing philosophy a lot of industries have. But we like to price things as cheaply as we possibly can, rather than charge as much as we can get. It's a big difference in philosophy, and we try to help other people see that."
We're back to the Yahoo! model because people have figured out how to game the system, namely Google, without adding content that's important to the searcher.
It's not hard to throw out most of the bottom-feeders.
We do it. The crowd at Search Engine Watch (which, despite the name, is all about advertising, not search quality) is writing me angry messages for doing that. Now that we've demonstrated that 36% of Google AdSense advertisers are bottom-feeders, they know they're being watched. Some feel they're being targeted.
Bear in mind that most search requests are really, really dumb. That's what Google has to answer. In fact, most Google search requests don't hit the search engine at all; there's a cache of common queries and answers in all the front end machines, and a sizable fraction of requests are answered from cache.
Wikia shows the problem with this approach. Coverage of Star [Wars|Trek|Gate|Craft] is extensive. Coverage of, say, bank regulation is nonexistent. If you want to find out how we got into the subprime mortgage mess or what to do about it, Wikia search is totally useless.
That's what you get from volunteer editors. Wikipedia does better, but most of the good contributions were made years ago.
Today, you pay the editors, or you get fancruft.
It's amusing that the author of the article feels overwhelmed by The Economist. That's a very well written magazine with good reporters; they had the only reporter in Lhasa when the Chinese clamped down, and they have a good analysis this week of the issues surrounding derivatives. If this guy can't handle The Economist, his organization's answers will probably be dumbed down to the level of, say, "People". That level of crap one can get for free, from many existing sources.
Remember Google Answers? Nobody really cared, and Google shut it down.
There's a whole industry of expensive, small-circulation specialist newsletters, but those are niche operations run by specialists in narrow fields.
The key feature seems to be "In other words, the Ringside platform allows business owners to gain insight into the social graph of users, relationships, groups, interactions, and sharing that is occurring on their Web site". Right. More targeted ads.
I have a browser extension that monitors advertiser (not user) behavior and reports it to a server. I mentioned this over on Search Engine Watch, where the Adwords crowd hangs out. Anger, threats, intimidation... The idea that someone is tracking advertisers, instead of users, just drives some of them nuts.
As the one who devised much of this congestion control strategy (see my RFC 896 and RFC 970, years before Van Jacobson), I suppose should say something.
The way this was supposed to work is that TCP needs to be well-behaved because it is to the advantage of the endpoint to be well-behaved. What makes this work is enforcement of fair queuing at the first router entering the network. Fair queuing balances load by IP address, not TCP connection, and "weighted fair queueing" allows quality of service controls to be imposed at the entry router.
The problem now is that the DOCSIS approach to cable modems, at least in its earlier versions, doesn't impose fair queuing at entry to the network from the subscriber side. So congestion occurs further upstream, near the cable headend, in the "middle" of the network.
By then, there are too many flows through the routers to do anything intelligent on a per-flow basis.
We still don't know how to handle congestion in the middle of an IP network. The best we have is "random early drop", but that's a hack. The whole Internet depends on stopping congestion near the entry point of the network. The cable guys didn't get this right in the upstream direction, and now they're hurting.
I'd argue for weighted fair queuing and QOS in the cable box. Try hard to push the congestion control out to the first router. DOCSIS 3 is a step in the right direction, if configured properly. But DOCSIS 3 is a huge collection of tuning parameters in search of a policy, and is likely to be grossly misconfigured.
The trick with quality of service is to offer either high-bandwidth or low latency service, but not both together. If you request low latency, your packets go into a per-IP queue with a high priority but a low queue length. Send too much and you lose packets. Send a little, and they get through fast. If you request high bandwidth, you get lower priority but a longer queue length, so you can fill up the pipe and wait for an ACK.
But I have no idea what to do about streaming video on demand, other than heavy buffering.
Multicast works for broadcast (non-on-demand) video, but other than for sports fans who want to watch in real time, it doesn't help much.
(I've previously suggested, sort of as a joke, that when a stream runs low on buffered content, the player should insert a pre-stored commercial while allowing the stream to catch up. Someone will probably try that.)
The trouble with studying engineering today is that it used to pay better. In 1970, the IEEE reported that electrical engineers and lawyers were making about the same salaries.
I had a quite good undergraduate engineering education. What sucked was going through Stanford CS for a Master's in the mid-1980s. I went through just as it was becoming clear that expert systems weren't going to lead to strong AI, but many on the faculty didn't want to admit it. Yet the expert systems people were still in charge. This was just as the "AI Winter" was starting, and the first-round AI startups were going bust. The whole experience was disappointing. I was fed way too much bullshit, and I knew it at the time. I have the Stanford diploma, but as an educational experience, it sucked.
Stanford finally had to transfer computer science from Arts and Sciences to Engineering and put in adult supervision. It's much improved now.
Ah, yet another class of ads to locate, rate, and filter. Now Adblock and CustomizeGoogle need to be updated.
We probably should look into rating the advertisers with AdRater. Outright ad blocking seems overkill for this class of ad, but rating doesn't interfere with user searches.
The revolt against excessive advertising is growing.
Sao Paulo, Brazil eliminated outdoor advertising last year. All of it.
Network Solutions is giving in to terrorists in censoring this.
But I'm sure it will be available everywhere real soon, probably on everything from BitTorrent to the Internet Archive.
Back in the 1970s, there were Islamic threats of bombings over "Mohammed, Messenger of God", and it was pulled from theaters. Yet today, nobody cares. That film is a straight biography of Mohammed; some of the less crazy Islamic countries use it as an educational film. You can buy it on DVD for $29.95 if you like. I've seen it; the battle scenes are impressive and the Arabian horses are very good ones, but the acting is disappointing.
On a threat scale of 10, with WWI and WWII as a 10, and the USSR as an 8, Islamic terrorism scores around 0.5. They'll probably make trouble again, and we'll deal with it when it happens. Meanwhile, ignore their threats.
Good points. If you're doing an embedded hard real time system, you might need the small size and timing repeatability of VXworks. If not, you don't need VXworks at all. It's much easier to write to a POSIX interface. There's not much software for VXworks you'd want to port to Linux. So a VXworks wrapper for Linux doesn't meet any widespread need. Maybe someone needed it for a specific porting job, so it was worth doing once.
I'd still like to see a real message passing system, like the one from QNX, in Linux. Even just writing web apps, Linux interprocess communication is still too lame. We have people using SOAP and JSON over sockets to do an interprocess subroutine call. I recently had to use Python pickling over pipes to a subprocess, and it's so primitive coming from QNX, where you have MsgSend/MsgRecv and a real client/server interprocess communication model.
I read through that and winced. It's one of those backwards compatible hacks that makes a language ugly. Current Javascript has a class implementation based on copying a base object, but no real class abstraction. This follows the model in Self. Most other object oriented languages (C++, Python, Java) have explicit class declarations. Javascript 2.0 adds class declarations without throwing out the old mechanism. This is a mess. I understand why they were forced to that decision, but it's still a mess.
The trend continues. They threw in most of the things Python 3K has and current Javascript doesn't: type annotations, generators, and packages, and namespaces. There's type checking, but it's optional for now. (This is like the transition from K&R C to ANSI C). Java-type interfaces where thrown in. At least they didn't add templates. It's a collection of features in search of a design.
In the end, we get something that's like a mixture of Java and Python.
Our school gets a bomb threat, and the teachers and administrators are freaked out. They move us all, I kid you not, to the football field where we are fenced in by chain link fence, about 1/3 of which is covered by barbed wire.
That kind of dumb response happens at higher levels. A few years back, there were three incidents where the U.S. Capitol was evacuated because a light aircraft had entered the Washington area without authorization. I was amazed at that response. The official response to an air raid is to send everyone outside? Not down to the basements?
(Especially since the U.S. Capitol has plenty of basement space, plus tunnels to most of the neighboring buildings.)
Worse, if it's a real attack with a light aircraft, it has to be chemical or biological.
A light aircraft just isn't big enough to cause substantial damage to a big stone building, even if it has explosives. Remember, there'd been an anthrax attack on the Capitol (we still don't know who was behind that), so that possibility was more than theoretical. But no, they send everyone outside to be sprayed.
Of course, it turned out that, each time, it was some inept pilot noodling around in a small plane, flying VFR and totally lost. One of the pilots responsible is even suing to get his pilots license back.
(NORAD actually did something intelligent about this. It's always VFR pilots who cluelessly wander into the Washington air defense identification zone, usually on clear days. IFR-qualified pilots know how to use their navaids, plan where they're going, file flight plans, have transponders identifying them on radar, and are in constant contact with ATC. So NORAD set up a laser system with several sites in the Washington area. Aircraft in the wrong place and not in contact with ATC get flashed with a bright red-red-green signal to tell them to get out of there, before they meet up with an F-16 in their face. This has been reasonably effective in getting clueless VFR pilots to turn around.)
This has got to stop. There's lots of software, from pre-fetching browsers to web crawlers to proxies to HTML validators that read other web sites.
I have a proxy-like tool available on line which occasionally gets abused. It's a lightweight HTML validator and debugging tool, intended to answer the question "why did the crawler misinterpret my site". You put in a URL, and the indicated page is fetched and parsed into a tree. Javascript and embedded objects are removed, and links are made absolute. Then the tree is converted back to properly indented HTML for display. This shows a web crawler's view of a web site; if the output doesn't look reasonably similar to the original page, your site probably won't be indexed properly by search engines.
This tool is regularly abused as a proxy server. At one point, somebody even built a call to it into the Debian build process because they needed to read HTTPS from some code that didn't know how to talk SSL. (That's been fixed.) There's still some bozo who has a script that reads pages from the Weather Channel through this tool. Every five minutes.
I finally to put a big banner at the top of each output page to discourage people from using it as a general purpose proxy server.
So now I have to worry about someone using it to read the FBI's p0rn collection.
Don't compare Apple with Dell. Compare it with Sony or Nintendo. Those companies are equally closed and secretive. Akio Morita (1921-1999) was Sony's founder and the equivalent of Steve Jobs. Sony hasn't been doing too well since Morita died.
How it works - some technical details
on
New BigDog Robot Video
·
· Score: 5, Informative
This is very nice work. It's good to see Raibert doing robotic locomotion again, and finally, with a big enough budget.
Back in the 1980s and early 1990s, Raibert headed the MIT Leg Lab, which produced the first legged robots with real balance control. Raibert started with one-legged hopping machines, to force the balance issue. His big insight was that balance is more important than gait.
In 1992, he left MIT and did a startup, Boston Dynamics, and went off into simulation. Most of the simulations weren't dynamic, just kinematic. Now he's back to robotics, and dynamics, again.
I've worked on control of robot running on rough terrain. So I understand the problems. Watching the Big Dog video, I have a reasonably good idea of how it works. This is quite impressive. DARPA got its $40 million worth.
First, it has slip control, like automotive ABS, for its feet. The first insight on the hard cases for locomotion is that balance is more important than gait. The second is that slip control is more important than balance. The key to slip control is keeping the transverse forces at foot-ground contact below the point where the feet break loose. ("Inside the static friction cone", for those familiar with the terminology.) Watch it move on ice. The feet do not slip at all unless there's real trouble, as when someone kicks the thing. The transverse forces are being held below the break-loose point. Given the load on the foot, the actuator forces (hydraulic cylinders on Big Dog) must be coordinated to keep the transverse force below the ground coefficient of friction times the longitudinal load.
Finding the ground coefficient of friction can be either trial and error (if it slips, reduce the value) or they may have actual slip sensing in the foot, like humans and animals. Humans, incidentally, tend to maintain a contact force about 20% above the break-loose point, as a safety margin.
Big Dog's reaction to a slip is to immediately raise the foot and go for a new foot placement. That's an emergency behavior, though; it's the prevention of slip that makes it work.
Watch the robot's reaction when it slips on ice, and, once you know what to look for,
you'll see how it does it. The first priority is to recover traction. As soon as a foot slips, it's lifted and placed in a new position. The second priority is to recover balance. As the robot starts to roll to the right, it executes a violent twist to the right and throws out the right front foot. It needs a foot position within the traction limits to provide the roll moment needed to recover balance, and it has a good enough planner to find one. Look at that sequence and ask yourself first "where does the foot need to be to get traction", then "where does the foot need to be to recover balance". Then you'll understand how it works.
Big Dog has, finally, true gaitless locomotion. Decades of locomotion research have focused on gait, foot sequence, "central patten generators", and similar mechanisms that deal with the easy cases. Wrong answer. The right answer is to think of legs as assets that can be deployed to maintain slip and stability criteria. It's very clear that Big Dog does this; it can use its feet (and knees!) as necessary. It's not constrained to a gait pattern at all.
There's a true dynamics predictor and planner in there. This is not just a reactive robot, like Brooks' little machines. Nor is it a straightforward ZMP ("zero moment point") stabilization system, like Asimo. (Think of ZMP as a generalization of center of gravity to include momentum.) There's a planner with a horizon of (I think) about two foot placements ahead, and it has "what if" internal simulation capability. That's why this robot moves so well. It can predict, at least approximately, what's going to happen for its next move, and plans on that basis. That's why its movement are so smooth. Without that, you'
I'd be interested in seeing the criteria, and sample data, for determining the quality of advertisers before I view your report as having any legitimacy.
But what does your cell phone talk to? That's a real issue with WiFi-enabled cell phones. I haven't seen a report of this yet, but clearly one could create an iPhone-based attack which looked for vulnerable WiFi networks and attacked them. Just carrying an iPhone with unapproved software to a workplace with WiFi creates a security risk. An external attacker might not be able to get inside the building, but with an iPhone as a Trojan horse...
We've been doing some tracking recently, but aimed at the advertiser side. We have a plug-in for Firefox which rates ads. A little icon is displayed next to each ad, showing what our system knows about the advertiser. As we tell users of the plug in, "AdRater 'phones home', but tells us as little as possible. AdRater sends the domain name associated with each advertisment you see to SiteTruth." SiteTruth then sends back advertiser information, in XML, which the plug-in turns into icons.
We use this to find out what the advertisers are doing. Individuals are entitled to privacy; advertisers are not. We're building up a picture of the on-line advertising market. We now have, for example, a list of Google's AdSense advertisers.
Soon we'll be issuing reports on advertiser quality. (Ads on Bloomberg: mostly legit. Ads on LinkedIn: quality varies, mostly OK. Ads on MySpace: mostly bottom-feeders.) More on this in coming weeks.
It's not just advertisers tracking users any more. Sometimes it's the other way round.
with separate electric motors for each wheel then you must have a way to synchronize all the motors to run at the same speed...
i would be willing to bet the electronic that would go in to a system would be plenty complicated, not impossible - but not a brainless task either...
Code for that is in most of the better vehicle traction and stability control systems right now. Ordinary cars now have two axes of rate gyro, steering wheel sensors, wheel encoders, and computer-controlled individual wheel braking.
I was thinking that it was from Klaus Fuchs, but no, it's the British Penny report. "This report was completed on 1 July, was entitled Plutonium Weapon - General Description (UK Public Record Office File AVIA 65/1163, "Implosion") and gave the British atomic weapons program a preliminary design description roughly equivalent in terms of detail to the description provided the Soviets by Klaus Fuchs."
Slashdot Mirror
Mod parent up. That's a significant, and seldom mentioned, point. When communism was taken seriously as a competitor to capitalism, capitalism had to deliver a better standard of living for the average worker. That's no longer the case.
Then again, look at China's growth rate.
If it's the "fuse" in the "nose cone", it's probably the radar proximity fuze, used to detonate a nuclear weapon at a specific height above ground. This is essential only for ICBMs intended for use against hardened targets, where the detonation has to occur at just the right height to maximize the blast effect against something like a missile silo lid.
If you're delivering your bomb in a Ryder truck, this component is unnecessary.
The pace of change is slowing down. Look at four 50 year periods in history.
Progress is flatlining.
Ah, instant gratification. The system rates on demand, then caches for 30 days. So the first user to query a new site has to wait. On the other hand, any new site can be rated immediately; there's no wait for a crawler to get to it. Linksys is now rated, and it now moves up in the rankings. And Wikipedia is listed as non-commercial; it's in ".org", there's no conflicting info from Open Directory, and it doesn't have ads.
Besides, the "alternate operating system" thing for the WRT45G never worked very well. If you were reading over the wireless link from a fast local server, the Linux software would consistently garble packets. It only worked if the air link never bottlenecked.
Distribution, $0.90? $900 for a thousand CDs? No way, not for WalMart.
This is WalMart you're shipping to. You ship to them by the truckload, not one CD at a time. Any in-store costs come under retail overhead, not distribution.
The promotion costs need to shrink. Maybe we'll see the labels begging for time on webcasts. Label overhead is far too high. The labels don't really do much today except promote; they don't directly employ artists, they don't run recording studios, they don't manufacture CDs, and they don't do physical distribution and warehousing. That's all outsourced. But management overhead hasn't been cut accordingly.
As the WalMart VP says: "The labels price things based on what they believe they can get -- a pricing philosophy a lot of industries have. But we like to price things as cheaply as we possibly can, rather than charge as much as we can get. It's a big difference in philosophy, and we try to help other people see that."
We're back to the Yahoo! model because people have figured out how to game the system, namely Google, without adding content that's important to the searcher.
It's not hard to throw out most of the bottom-feeders. We do it. The crowd at Search Engine Watch (which, despite the name, is all about advertising, not search quality) is writing me angry messages for doing that. Now that we've demonstrated that 36% of Google AdSense advertisers are bottom-feeders, they know they're being watched. Some feel they're being targeted.
Bear in mind that most search requests are really, really dumb. That's what Google has to answer. In fact, most Google search requests don't hit the search engine at all; there's a cache of common queries and answers in all the front end machines, and a sizable fraction of requests are answered from cache.
Wikia shows the problem with this approach. Coverage of Star [Wars|Trek|Gate|Craft] is extensive. Coverage of, say, bank regulation is nonexistent. If you want to find out how we got into the subprime mortgage mess or what to do about it, Wikia search is totally useless. That's what you get from volunteer editors. Wikipedia does better, but most of the good contributions were made years ago.
Today, you pay the editors, or you get fancruft.
It's amusing that the author of the article feels overwhelmed by The Economist. That's a very well written magazine with good reporters; they had the only reporter in Lhasa when the Chinese clamped down, and they have a good analysis this week of the issues surrounding derivatives. If this guy can't handle The Economist, his organization's answers will probably be dumbed down to the level of, say, "People". That level of crap one can get for free, from many existing sources.
Remember Google Answers? Nobody really cared, and Google shut it down.
There's a whole industry of expensive, small-circulation specialist newsletters, but those are niche operations run by specialists in narrow fields.
The key feature seems to be "In other words, the Ringside platform allows business owners to gain insight into the social graph of users, relationships, groups, interactions, and sharing that is occurring on their Web site". Right. More targeted ads.
I have a browser extension that monitors advertiser (not user) behavior and reports it to a server. I mentioned this over on Search Engine Watch, where the Adwords crowd hangs out. Anger, threats, intimidation... The idea that someone is tracking advertisers, instead of users, just drives some of them nuts.
As the one who devised much of this congestion control strategy (see my RFC 896 and RFC 970, years before Van Jacobson), I suppose should say something.
The way this was supposed to work is that TCP needs to be well-behaved because it is to the advantage of the endpoint to be well-behaved. What makes this work is enforcement of fair queuing at the first router entering the network. Fair queuing balances load by IP address, not TCP connection, and "weighted fair queueing" allows quality of service controls to be imposed at the entry router.
The problem now is that the DOCSIS approach to cable modems, at least in its earlier versions, doesn't impose fair queuing at entry to the network from the subscriber side. So congestion occurs further upstream, near the cable headend, in the "middle" of the network. By then, there are too many flows through the routers to do anything intelligent on a per-flow basis.
We still don't know how to handle congestion in the middle of an IP network. The best we have is "random early drop", but that's a hack. The whole Internet depends on stopping congestion near the entry point of the network. The cable guys didn't get this right in the upstream direction, and now they're hurting.
I'd argue for weighted fair queuing and QOS in the cable box. Try hard to push the congestion control out to the first router. DOCSIS 3 is a step in the right direction, if configured properly. But DOCSIS 3 is a huge collection of tuning parameters in search of a policy, and is likely to be grossly misconfigured.
The trick with quality of service is to offer either high-bandwidth or low latency service, but not both together. If you request low latency, your packets go into a per-IP queue with a high priority but a low queue length. Send too much and you lose packets. Send a little, and they get through fast. If you request high bandwidth, you get lower priority but a longer queue length, so you can fill up the pipe and wait for an ACK.
But I have no idea what to do about streaming video on demand, other than heavy buffering. Multicast works for broadcast (non-on-demand) video, but other than for sports fans who want to watch in real time, it doesn't help much. (I've previously suggested, sort of as a joke, that when a stream runs low on buffered content, the player should insert a pre-stored commercial while allowing the stream to catch up. Someone will probably try that.)
John Nagle
The trouble with studying engineering today is that it used to pay better. In 1970, the IEEE reported that electrical engineers and lawyers were making about the same salaries.
I had a quite good undergraduate engineering education. What sucked was going through Stanford CS for a Master's in the mid-1980s. I went through just as it was becoming clear that expert systems weren't going to lead to strong AI, but many on the faculty didn't want to admit it. Yet the expert systems people were still in charge. This was just as the "AI Winter" was starting, and the first-round AI startups were going bust. The whole experience was disappointing. I was fed way too much bullshit, and I knew it at the time. I have the Stanford diploma, but as an educational experience, it sucked.
Stanford finally had to transfer computer science from Arts and Sciences to Engineering and put in adult supervision. It's much improved now.
Ah, yet another class of ads to locate, rate, and filter. Now Adblock and CustomizeGoogle need to be updated.
We probably should look into rating the advertisers with AdRater. Outright ad blocking seems overkill for this class of ad, but rating doesn't interfere with user searches.
The revolt against excessive advertising is growing. Sao Paulo, Brazil eliminated outdoor advertising last year. All of it.
Network Solutions is giving in to terrorists in censoring this.
But I'm sure it will be available everywhere real soon, probably on everything from BitTorrent to the Internet Archive.
Back in the 1970s, there were Islamic threats of bombings over "Mohammed, Messenger of God", and it was pulled from theaters. Yet today, nobody cares. That film is a straight biography of Mohammed; some of the less crazy Islamic countries use it as an educational film. You can buy it on DVD for $29.95 if you like. I've seen it; the battle scenes are impressive and the Arabian horses are very good ones, but the acting is disappointing.
On a threat scale of 10, with WWI and WWII as a 10, and the USSR as an 8, Islamic terrorism scores around 0.5. They'll probably make trouble again, and we'll deal with it when it happens. Meanwhile, ignore their threats.
Good points. If you're doing an embedded hard real time system, you might need the small size and timing repeatability of VXworks. If not, you don't need VXworks at all. It's much easier to write to a POSIX interface. There's not much software for VXworks you'd want to port to Linux. So a VXworks wrapper for Linux doesn't meet any widespread need. Maybe someone needed it for a specific porting job, so it was worth doing once.
I'd still like to see a real message passing system, like the one from QNX, in Linux. Even just writing web apps, Linux interprocess communication is still too lame. We have people using SOAP and JSON over sockets to do an interprocess subroutine call. I recently had to use Python pickling over pipes to a subprocess, and it's so primitive coming from QNX, where you have MsgSend/MsgRecv and a real client/server interprocess communication model.
And QuickTime tries to install iTunes. Does this mean that installing QuickTime now forces you through a Safari install?
Does it make Safari the default browser, disabling Internet Explorer?
First, ignore the ad for the blog and go directly to the language specification.
I read through that and winced. It's one of those backwards compatible hacks that makes a language ugly. Current Javascript has a class implementation based on copying a base object, but no real class abstraction. This follows the model in Self. Most other object oriented languages (C++, Python, Java) have explicit class declarations. Javascript 2.0 adds class declarations without throwing out the old mechanism. This is a mess. I understand why they were forced to that decision, but it's still a mess.
The trend continues. They threw in most of the things Python 3K has and current Javascript doesn't: type annotations, generators, and packages, and namespaces. There's type checking, but it's optional for now. (This is like the transition from K&R C to ANSI C). Java-type interfaces where thrown in. At least they didn't add templates. It's a collection of features in search of a design.
In the end, we get something that's like a mixture of Java and Python.
The RIAA and the MPAA will love this. At last, content can be licensed to the individual, not the device. "Pay per viewer", at last.
And you can't cover the camera; if it can't see you to identify your biometrics, your licenses won't validate.
Our school gets a bomb threat, and the teachers and administrators are freaked out. They move us all, I kid you not, to the football field where we are fenced in by chain link fence, about 1/3 of which is covered by barbed wire.
That kind of dumb response happens at higher levels. A few years back, there were three incidents where the U.S. Capitol was evacuated because a light aircraft had entered the Washington area without authorization. I was amazed at that response. The official response to an air raid is to send everyone outside? Not down to the basements? (Especially since the U.S. Capitol has plenty of basement space, plus tunnels to most of the neighboring buildings.)
Worse, if it's a real attack with a light aircraft, it has to be chemical or biological. A light aircraft just isn't big enough to cause substantial damage to a big stone building, even if it has explosives. Remember, there'd been an anthrax attack on the Capitol (we still don't know who was behind that), so that possibility was more than theoretical. But no, they send everyone outside to be sprayed.
Of course, it turned out that, each time, it was some inept pilot noodling around in a small plane, flying VFR and totally lost. One of the pilots responsible is even suing to get his pilots license back.
(NORAD actually did something intelligent about this. It's always VFR pilots who cluelessly wander into the Washington air defense identification zone, usually on clear days. IFR-qualified pilots know how to use their navaids, plan where they're going, file flight plans, have transponders identifying them on radar, and are in constant contact with ATC. So NORAD set up a laser system with several sites in the Washington area. Aircraft in the wrong place and not in contact with ATC get flashed with a bright red-red-green signal to tell them to get out of there, before they meet up with an F-16 in their face. This has been reasonably effective in getting clueless VFR pilots to turn around.)
This has got to stop. There's lots of software, from pre-fetching browsers to web crawlers to proxies to HTML validators that read other web sites.
I have a proxy-like tool available on line which occasionally gets abused. It's a lightweight HTML validator and debugging tool, intended to answer the question "why did the crawler misinterpret my site". You put in a URL, and the indicated page is fetched and parsed into a tree. Javascript and embedded objects are removed, and links are made absolute. Then the tree is converted back to properly indented HTML for display. This shows a web crawler's view of a web site; if the output doesn't look reasonably similar to the original page, your site probably won't be indexed properly by search engines.
This tool is regularly abused as a proxy server. At one point, somebody even built a call to it into the Debian build process because they needed to read HTTPS from some code that didn't know how to talk SSL. (That's been fixed.) There's still some bozo who has a script that reads pages from the Weather Channel through this tool. Every five minutes. I finally to put a big banner at the top of each output page to discourage people from using it as a general purpose proxy server.
So now I have to worry about someone using it to read the FBI's p0rn collection.
Don't compare Apple with Dell. Compare it with Sony or Nintendo. Those companies are equally closed and secretive. Akio Morita (1921-1999) was Sony's founder and the equivalent of Steve Jobs. Sony hasn't been doing too well since Morita died.
This is very nice work. It's good to see Raibert doing robotic locomotion again, and finally, with a big enough budget.
Back in the 1980s and early 1990s, Raibert headed the MIT Leg Lab, which produced the first legged robots with real balance control. Raibert started with one-legged hopping machines, to force the balance issue. His big insight was that balance is more important than gait. In 1992, he left MIT and did a startup, Boston Dynamics, and went off into simulation. Most of the simulations weren't dynamic, just kinematic. Now he's back to robotics, and dynamics, again.
I've worked on control of robot running on rough terrain. So I understand the problems. Watching the Big Dog video, I have a reasonably good idea of how it works. This is quite impressive. DARPA got its $40 million worth.
First, it has slip control, like automotive ABS, for its feet. The first insight on the hard cases for locomotion is that balance is more important than gait. The second is that slip control is more important than balance. The key to slip control is keeping the transverse forces at foot-ground contact below the point where the feet break loose. ("Inside the static friction cone", for those familiar with the terminology.) Watch it move on ice. The feet do not slip at all unless there's real trouble, as when someone kicks the thing. The transverse forces are being held below the break-loose point. Given the load on the foot, the actuator forces (hydraulic cylinders on Big Dog) must be coordinated to keep the transverse force below the ground coefficient of friction times the longitudinal load. Finding the ground coefficient of friction can be either trial and error (if it slips, reduce the value) or they may have actual slip sensing in the foot, like humans and animals. Humans, incidentally, tend to maintain a contact force about 20% above the break-loose point, as a safety margin.
Big Dog's reaction to a slip is to immediately raise the foot and go for a new foot placement. That's an emergency behavior, though; it's the prevention of slip that makes it work. Watch the robot's reaction when it slips on ice, and, once you know what to look for, you'll see how it does it. The first priority is to recover traction. As soon as a foot slips, it's lifted and placed in a new position. The second priority is to recover balance. As the robot starts to roll to the right, it executes a violent twist to the right and throws out the right front foot. It needs a foot position within the traction limits to provide the roll moment needed to recover balance, and it has a good enough planner to find one. Look at that sequence and ask yourself first "where does the foot need to be to get traction", then "where does the foot need to be to recover balance". Then you'll understand how it works.
Big Dog has, finally, true gaitless locomotion. Decades of locomotion research have focused on gait, foot sequence, "central patten generators", and similar mechanisms that deal with the easy cases. Wrong answer. The right answer is to think of legs as assets that can be deployed to maintain slip and stability criteria. It's very clear that Big Dog does this; it can use its feet (and knees!) as necessary. It's not constrained to a gait pattern at all.
There's a true dynamics predictor and planner in there. This is not just a reactive robot, like Brooks' little machines. Nor is it a straightforward ZMP ("zero moment point") stabilization system, like Asimo. (Think of ZMP as a generalization of center of gravity to include momentum.) There's a planner with a horizon of (I think) about two foot placements ahead, and it has "what if" internal simulation capability. That's why this robot moves so well. It can predict, at least approximately, what's going to happen for its next move, and plans on that basis. That's why its movement are so smooth. Without that, you'
I'd be interested in seeing the criteria, and sample data, for determining the quality of advertisers before I view your report as having any legitimacy.
Sure. See these documents.
Guys, look, even I take my cell phone to work.
But what does your cell phone talk to? That's a real issue with WiFi-enabled cell phones. I haven't seen a report of this yet, but clearly one could create an iPhone-based attack which looked for vulnerable WiFi networks and attacked them. Just carrying an iPhone with unapproved software to a workplace with WiFi creates a security risk. An external attacker might not be able to get inside the building, but with an iPhone as a Trojan horse...
We've been doing some tracking recently, but aimed at the advertiser side. We have a plug-in for Firefox which rates ads. A little icon is displayed next to each ad, showing what our system knows about the advertiser. As we tell users of the plug in, "AdRater 'phones home', but tells us as little as possible. AdRater sends the domain name associated with each advertisment you see to SiteTruth." SiteTruth then sends back advertiser information, in XML, which the plug-in turns into icons.
We use this to find out what the advertisers are doing. Individuals are entitled to privacy; advertisers are not. We're building up a picture of the on-line advertising market. We now have, for example, a list of Google's AdSense advertisers.
Soon we'll be issuing reports on advertiser quality. (Ads on Bloomberg: mostly legit. Ads on LinkedIn: quality varies, mostly OK. Ads on MySpace: mostly bottom-feeders.) More on this in coming weeks.
It's not just advertisers tracking users any more. Sometimes it's the other way round.
with separate electric motors for each wheel then you must have a way to synchronize all the motors to run at the same speed... i would be willing to bet the electronic that would go in to a system would be plenty complicated, not impossible - but not a brainless task either...
Code for that is in most of the better vehicle traction and stability control systems right now. Ordinary cars now have two axes of rate gyro, steering wheel sensors, wheel encoders, and computer-controlled individual wheel braking.
I was thinking that it was from Klaus Fuchs, but no, it's the British Penny report. "This report was completed on 1 July, was entitled Plutonium Weapon - General Description (UK Public Record Office File AVIA 65/1163, "Implosion") and gave the British atomic weapons program a preliminary design description roughly equivalent in terms of detail to the description provided the Soviets by Klaus Fuchs."