America's Army still has the best solution.
Their in-game implementation of the United States Army Disciplinary Barracks at Fort Leavenworth. They just put griefers in a barred cell from which there is no escape, and keep them there for a while. There's nothing to do in the cell, except peer out the little barred window and watch the sun go down.
What we need is an FTC rule that advertising any service quality or quantity with the words "up to" or substantially similar language is, by law, considered deceptive. Advertising should have to specify a guaranteed level of service. That would put cable and DSL on the same measurement scale, discourage underprovisioning, and make cellular data transfer rates in ads something you could rely on.
There's precedent for this. At various times in the past, the FTC had to tighten up the definition of "horsepower" for cars and "watts" for audio gear.
When I saw a menu at the top of Google's search page, I thought "They've lost it. Google has jumped the shark". They want to be a "portal", five years after that was a good idea.
Notice that the most prominent link to click on is "Advertising programs".
Interestingly, Sprint is going in the other direction. Some of their new phones do not have cameras, but meet MIL-STD 810F for ruggedization. Sprint used to offer ruggedized phones only on the Nextel network, but ruggedized phones are now appearing on the Sprint side too. Ruggedized models from both Motorola and Sanyo are now available.
Computers are so good at chess now that it's embarrassing. Unless you've been on the cover of Chess Life, any of the good PC chess programs can trounce you. Fritz at €119.90, runs on single or multiprocessor PCs, is rated at FIDE 2808 or so, and wins against Kasparov about half the time. If you're not a rated player, the chess programs for cell phones can beat you.
One of the experts in computer chess explained what's happened. Study of human grandmaster games indicates that about one move in ten is suboptimal, even at that level. That's enough to give computers that don't make mistakes a significant edge.
Computers are now so far ahead that there's a serious problem with cheating using a computer in chess competition, Several cheaters were caught at the 2006 World Open. "Two players are under suspicion of having received help from computers at the World Open in Philadelphia. One locked himself in a bathroom stall, the other, who was leading the event before the last round and stood to win $18,000, was caught wearing a "hearing aid" which turned out to be a wireless receiver used for surreptitious communications. The New York Times reports."
Chess players at major tournaments are now being searched.
I could see in-game ads for games in contemporary settings, like GTA, but in World of Warcraft, they'd just look silly.
This is an issue for Hollywood - advertisers can't do product placement in historicals, which makes them unpopular. (Although if you look very closely at Marie Antoinette's closet, you'll see a pair of Converse All-Stars.)
I don't have decent cellular coverage in my house, and I live one mile from downtown Palo Alto in Silicon Valley. Five cellular stores (not counting the Apple store) within walking distance, and I have to go to a window to get more than one bar on the phone. Gigahertz RF doesn't go through trees, you know.
That thing has an incredibly complex cycle, with losses all along the chain. There's ammonia, water, steam, air, and hot oil involved, with heat exchangers all over the place. The paper attached to it doesn't describe the basic thermodynamics in any real detail. It's sort of like a solar-powered Rankin cycle system. But much more complex, and without solid justification for the extra complexity.
This might be credible if they had a working prototype, even a little one. A prototype in the 1 KW range would be about right. That's a backyard project. A 1KW plant would need about 10 square meters of collector mirror, which isn't too hard. Then they'd have something. All they have now is hype.
It is hard to know whether a mouse-stroke "took" or not, and sometimes it resizes wrong. It just feels "unnatural".
We ran into that problem with our search/rating box. When you click on the search button, nothing visible happened immediately, confusing the user while the request was going out to the server and back. So we put "Searching..." and "Rating..." into the result area immediately when a request is made, for immediate feedback. Even though that text is often replaced with real results so fast you sometimes don't even see it, it's worth doing if there's any possibility of a delay of more than a second.
Do this on your web commerce sites, if they have an interface that doesn't load a new page. It beats those stupid "Don't push the BUY button more than once or you may be billed twice" messages.
There are proper NEMA color codes for indicators, and you'll see them on industrial equipment. Unfortunately, we went through a long period during which red LEDs were the only cheap color, and far too many red LEDs went onto equipment. Since LEDs are now available in all colors, it's time to go back to the traditional NEMA rules:
GREEN - equipment normal, no action required.
AMBER - abnormal condition, action may be required, but not immediate action.
RED - trouble condition, action is required. No red light should be illuminated during normal operation. If you see a red light, something needs to be done about it immediately.
BLUE - status indication, no specific meaning.
WHITE - status indication, no specific meaning.
Anything that goes in a factory or a rack should obey those simple rules.
Yes, the Shuttle boxes have way too much lamp power. You can read by the thing.
But the most annoying device I have right now is a Panasonic electric razor with four LEDs. One of which blinks red while charging. At night, the ceiling of my bedroom glows with red flashes.
That's right. A "?" is our usual rating for a business that's been heard of, but with no street address on the site. Most major blogs and such get a "?" rating.
These ratings are most meaningful for shopping sites, ones that you're buying from online, where there's a legal obligation to clearly identify the business.
That's right. "slashdot.com", which is a redirect, gets a low rating. Check "slashdot.org", the real site, which does better.
We're still having trouble with aliasing issues. Simply because A redirects to B doesn't mean that the people who run B also run A. Whether a redirect is entitled to the same rating as the actual site is a tough question. If yes, there are ways to exploit redirects through hostile pages. (Remember the problem with Google AdWords from last week.) If no, there are problems with sites that expose multiple names. We're working on ways to check validity of redirects. Thanks for the note.
Tribe.net redesigned their home page to use "Web 2.0" around the beginning of 2007. Now users could drag the various boxes around, rearrange the home page, and choose which elements they wanted. (Except for the ads, of course, which were immovable.)
The main effect was that "Tribe.net bug reports" became one of the most active groups. Tribe's traffic ratings in Alexa continued to slide.
There are uses for the asynchrony of XMLHttpRequest, though. Try our search and rating box. We have a site rating engine which rates sites on demand, and it takes takes about 8 to 30 seconds for sites it hasn't looked at yet. We needed a way to present this to the user without stalling the user's browsing.
So we needed a truly asynchronous web page, and we have one.
When you enter text into the box and click the big "Search" button, the site gets all the results it can get from the databases immediately, and updates the page. The sites for which ratings aren't yet available show as rotating "busy" icons, which are replaced over the next few seconds as the server reads the target web site, rates it, and sends the ratings back to the browser.
If we did this with stock HTML, the whole thing would feel so sluggish as to be useless. But with a dynamic page, the user gets useful results immediately, which improve over the next 8 to 30 seconds. The user's browsing isn't stalled. In fact, if you enter something new into the search box while updates are in progress, outstanding XMLHttpRequest requests are aborted, and you can do a new search without waiting for old ratings to complete.
Few "Web 2.0" sites seem to support as much asynchrony. Google Maps is probably the best known site that really is asynchronous in a useful way.
Yes, Scientology is nutty, but that's about normal for a religion. Could be worse. They don't have a big pedophile problem, suicide bombers, or televangelists, like some of their competitors.
Guy Kawasaki has been doing something like this for years, with
Garage.com. Their biggest success was Claria, which distributed adware. Not a great track record there.
I ran into repeated use of the term "don the purple" when describing the accession of Roman emperors. Yet I NEVER found a description of what "the purple" really meant.
It meant a toga with purple trim.
Before synthetic dyes were developed, some dye colors were very expensive to achieve, and purple was one of them. The source of the dye was sea snails, and the extraction process was a secret. In the later days of the Roman Empire, wearing purple was restricted to emperors. See "sumptuary laws".
"Software as a service" is mostly about putting ads into applications. Or at least the current version of "service" is. The previous try at this, "application service providers" (remember those) was pay per view, which didn't fly.
The only reason the "browser as a platform" idea is popular is because the process of installing software has become so awful. The original MacOS had approximately the right idea; put an application in any folder, anywhere, and the Finder will find it. What we have now is "Let us run this giant installer program with root/administrator privileges which will change settings all over the system". That's awful, and it's no better on Linux than it was on Windows.
Technically, this can be fixed, but politically, it means changing the way applications are detected and what they can do. That's infeasible in the Open Source world.
It's not all that hard to improve search. The problem is improving search when you're really in the business of selling ads.
With Yahoo, this is painfully obvious. Yahoo has a good search engine, but their home page and search result pages are so ad-heavy that they're annoying to use. Google has so far resisted the temptation to run picture ads, but there's heavy pressure for them to do so, from both investors and advertisers. The smaller search entrants tend to have more ads; they need the revenue.
As a technology demo, we have SiteTruth, a consumer-oriented search and rating system, in alpha test. We rate commercial web sites based on "business legitimacy". That starts with finding the business name and address of the organization behind the web site. That bypasses most "affiliates", "doorway pages", and similar junk sites, and gets you to the actual site selling something. As we take this further, we'll be validating incorporation status, business licenses, business credit ratings, and the other things one checks when checking out a business. You know, all that stuff you're supposed to check, but nobody ever does.
Then we push the sites with bad ratings to the bottom of the search results and off into obscurity, where they usually belong.
Site rating has been tried before, but usually based on user recommendations. User recommendations are too easy to fake; most of the people who write them are interested parties. (Our neighborhood video store is offering a free rental if you say good stuff about them on Yelp.) And coverage is usually narrow; there are more sites than people rating them.
Any news on when the rest of the world can finally buy the XO Laptops? Since, well, a 200dpi display for less then $200 sounds like a damn cool device for ebook reading and I really want one.
If you haven't actually seen an XO machine, just pictures, you may not realize that it's much smaller than a standard laptop. The keyboard is sized for kid-sized hands.
Dumb Wired writers, expecting instant gratification. Wired used to have reporters who actually went out and covered real stuff. Then they laid off most of the reporters and kept the "editors". Now they're just wannabe pundits. Saves on travel expenses.
That Tired writer isn't coming across as someone who spent long days digging something out of library stacks or public records. Or travelling around asking people questions to find out what really happened, like a real reporter. This is a lightweight.
If you want a children's encyclopedia, you can still get World Book.
Wikipedia has many problems, of course. Most of the good articles were in the first 500,000 created. What's coming in now is mostly junk - "State Route 92", "Star Wars Furry Adventure #6659", and similar crap. Wikia offers some hope for an amusing reason. Wikia took over Wookiepedia, the repository of Star Wars fancruft, which generates most of Wikia's traffic. They're monetizing the fan base. Over time, maybe all the popular culture stuff can be moved to Wikia. That would be a win.
What they're doing is straightforward, and it's much like what many virus scanners do. First, they look at web pages to see if there's anything suspicious that requires further analysis. If there is, they load the page into Internet Explorer (of course) in a virtual machine, and see if it changes its environment. The better virus scanners have been doing something like that for a few years now, running possible viruses in some kind of sandbox. Although they usually don't go all the way and run Internet Explorer in a virtual machine. (Are you allowed to do that under Microsoft's current EULA for IE 7?)
The main problem with Google's approach here is that it's after the fact. They won't notice a bad page until the next time they crawl it. Bad pages come and go so fast today that they'll always be behind. As the paper says, "Since many of the malicious URLs
are too short-lived to provide statistically meaningful data,
we analyzed only the URLs whose presence on the Internet
lasted longer than one week."
If Google implements this, the main effect will be to push attackers into changing site names for attack sites even faster.
It's all so backward. What we need is to run most of Internet Explorer in a tightly sandboxed environment on the user's machine, so that when you close the window, any browser damage goes away. That would actually work.
Slashdot Mirror
America's Army still has the best solution. Their in-game implementation of the United States Army Disciplinary Barracks at Fort Leavenworth. They just put griefers in a barred cell from which there is no escape, and keep them there for a while. There's nothing to do in the cell, except peer out the little barred window and watch the sun go down.
What domains do Acquantive ads come from? I need to update certain tables. Thanks.
What we need is an FTC rule that advertising any service quality or quantity with the words "up to" or substantially similar language is, by law, considered deceptive. Advertising should have to specify a guaranteed level of service. That would put cable and DSL on the same measurement scale, discourage underprovisioning, and make cellular data transfer rates in ads something you could rely on.
There's precedent for this. At various times in the past, the FTC had to tighten up the definition of "horsepower" for cars and "watts" for audio gear.
Owners of first-generation Intel Macs that used (32-bit only) Core Duo CPUs...
That was a wierd decision on Apple's part, since they had a 64-bit PowerPC system when they forced their user base to go back to 32 bits.
When I saw a menu at the top of Google's search page, I thought "They've lost it. Google has jumped the shark". They want to be a "portal", five years after that was a good idea.
Notice that the most prominent link to click on is "Advertising programs".
Interestingly, Sprint is going in the other direction. Some of their new phones do not have cameras, but meet MIL-STD 810F for ruggedization. Sprint used to offer ruggedized phones only on the Nextel network, but ruggedized phones are now appearing on the Sprint side too. Ruggedized models from both Motorola and Sanyo are now available.
"At Dell, customer service means no service at all." - Attorney General of the State of New York.
Now that's bad press. Visualize that on a billboard, or in a competitor's commercial.
Computers are so good at chess now that it's embarrassing. Unless you've been on the cover of Chess Life, any of the good PC chess programs can trounce you. Fritz at €119.90, runs on single or multiprocessor PCs, is rated at FIDE 2808 or so, and wins against Kasparov about half the time. If you're not a rated player, the chess programs for cell phones can beat you.
One of the experts in computer chess explained what's happened. Study of human grandmaster games indicates that about one move in ten is suboptimal, even at that level. That's enough to give computers that don't make mistakes a significant edge.
Computers are now so far ahead that there's a serious problem with cheating using a computer in chess competition, Several cheaters were caught at the 2006 World Open. "Two players are under suspicion of having received help from computers at the World Open in Philadelphia. One locked himself in a bathroom stall, the other, who was leading the event before the last round and stood to win $18,000, was caught wearing a "hearing aid" which turned out to be a wireless receiver used for surreptitious communications. The New York Times reports."
Chess players at major tournaments are now being searched.
I could see in-game ads for games in contemporary settings, like GTA, but in World of Warcraft, they'd just look silly.
This is an issue for Hollywood - advertisers can't do product placement in historicals, which makes them unpopular. (Although if you look very closely at Marie Antoinette's closet, you'll see a pair of Converse All-Stars.)
I don't have decent cellular coverage in my house, and I live one mile from downtown Palo Alto in Silicon Valley. Five cellular stores (not counting the Apple store) within walking distance, and I have to go to a window to get more than one bar on the phone. Gigahertz RF doesn't go through trees, you know.
That thing has an incredibly complex cycle, with losses all along the chain. There's ammonia, water, steam, air, and hot oil involved, with heat exchangers all over the place. The paper attached to it doesn't describe the basic thermodynamics in any real detail. It's sort of like a solar-powered Rankin cycle system. But much more complex, and without solid justification for the extra complexity.
This might be credible if they had a working prototype, even a little one. A prototype in the 1 KW range would be about right. That's a backyard project. A 1KW plant would need about 10 square meters of collector mirror, which isn't too hard. Then they'd have something. All they have now is hype.
It is hard to know whether a mouse-stroke "took" or not, and sometimes it resizes wrong. It just feels "unnatural".
We ran into that problem with our search/rating box. When you click on the search button, nothing visible happened immediately, confusing the user while the request was going out to the server and back. So we put "Searching..." and "Rating..." into the result area immediately when a request is made, for immediate feedback. Even though that text is often replaced with real results so fast you sometimes don't even see it, it's worth doing if there's any possibility of a delay of more than a second.
Do this on your web commerce sites, if they have an interface that doesn't load a new page. It beats those stupid "Don't push the BUY button more than once or you may be billed twice" messages.
No convention seems to be emerging for this.
There are proper NEMA color codes for indicators, and you'll see them on industrial equipment. Unfortunately, we went through a long period during which red LEDs were the only cheap color, and far too many red LEDs went onto equipment. Since LEDs are now available in all colors, it's time to go back to the traditional NEMA rules:
Anything that goes in a factory or a rack should obey those simple rules.
Yes, the Shuttle boxes have way too much lamp power. You can read by the thing.
But the most annoying device I have right now is a Panasonic electric razor with four LEDs. One of which blinks red while charging. At night, the ceiling of my bedroom glows with red flashes.
That's right. A "?" is our usual rating for a business that's been heard of, but with no street address on the site. Most major blogs and such get a "?" rating.
These ratings are most meaningful for shopping sites, ones that you're buying from online, where there's a legal obligation to clearly identify the business.
That's right. "slashdot.com", which is a redirect, gets a low rating. Check "slashdot.org", the real site, which does better.
We're still having trouble with aliasing issues. Simply because A redirects to B doesn't mean that the people who run B also run A. Whether a redirect is entitled to the same rating as the actual site is a tough question. If yes, there are ways to exploit redirects through hostile pages. (Remember the problem with Google AdWords from last week.) If no, there are problems with sites that expose multiple names. We're working on ways to check validity of redirects. Thanks for the note.
Tribe.net redesigned their home page to use "Web 2.0" around the beginning of 2007. Now users could drag the various boxes around, rearrange the home page, and choose which elements they wanted. (Except for the ads, of course, which were immovable.) The main effect was that "Tribe.net bug reports" became one of the most active groups. Tribe's traffic ratings in Alexa continued to slide.
There are uses for the asynchrony of XMLHttpRequest, though. Try our search and rating box. We have a site rating engine which rates sites on demand, and it takes takes about 8 to 30 seconds for sites it hasn't looked at yet. We needed a way to present this to the user without stalling the user's browsing.
So we needed a truly asynchronous web page, and we have one. When you enter text into the box and click the big "Search" button, the site gets all the results it can get from the databases immediately, and updates the page. The sites for which ratings aren't yet available show as rotating "busy" icons, which are replaced over the next few seconds as the server reads the target web site, rates it, and sends the ratings back to the browser.
If we did this with stock HTML, the whole thing would feel so sluggish as to be useless. But with a dynamic page, the user gets useful results immediately, which improve over the next 8 to 30 seconds. The user's browsing isn't stalled. In fact, if you enter something new into the search box while updates are in progress, outstanding XMLHttpRequest requests are aborted, and you can do a new search without waiting for old ratings to complete.
Few "Web 2.0" sites seem to support as much asynchrony. Google Maps is probably the best known site that really is asynchronous in a useful way.
Yes, Scientology is nutty, but that's about normal for a religion. Could be worse. They don't have a big pedophile problem, suicide bombers, or televangelists, like some of their competitors.
Guy Kawasaki has been doing something like this for years, with Garage.com. Their biggest success was Claria, which distributed adware. Not a great track record there.
I ran into repeated use of the term "don the purple" when describing the accession of Roman emperors. Yet I NEVER found a description of what "the purple" really meant.
It meant a toga with purple trim.
Before synthetic dyes were developed, some dye colors were very expensive to achieve, and purple was one of them. The source of the dye was sea snails, and the extraction process was a secret. In the later days of the Roman Empire, wearing purple was restricted to emperors. See "sumptuary laws".
"Software as a service" is mostly about putting ads into applications. Or at least the current version of "service" is. The previous try at this, "application service providers" (remember those) was pay per view, which didn't fly.
The only reason the "browser as a platform" idea is popular is because the process of installing software has become so awful. The original MacOS had approximately the right idea; put an application in any folder, anywhere, and the Finder will find it. What we have now is "Let us run this giant installer program with root/administrator privileges which will change settings all over the system". That's awful, and it's no better on Linux than it was on Windows.
Technically, this can be fixed, but politically, it means changing the way applications are detected and what they can do. That's infeasible in the Open Source world.
It's not all that hard to improve search. The problem is improving search when you're really in the business of selling ads.
With Yahoo, this is painfully obvious. Yahoo has a good search engine, but their home page and search result pages are so ad-heavy that they're annoying to use. Google has so far resisted the temptation to run picture ads, but there's heavy pressure for them to do so, from both investors and advertisers. The smaller search entrants tend to have more ads; they need the revenue.
As a technology demo, we have SiteTruth, a consumer-oriented search and rating system, in alpha test. We rate commercial web sites based on "business legitimacy". That starts with finding the business name and address of the organization behind the web site. That bypasses most "affiliates", "doorway pages", and similar junk sites, and gets you to the actual site selling something. As we take this further, we'll be validating incorporation status, business licenses, business credit ratings, and the other things one checks when checking out a business. You know, all that stuff you're supposed to check, but nobody ever does.
Then we push the sites with bad ratings to the bottom of the search results and off into obscurity, where they usually belong.
Site rating has been tried before, but usually based on user recommendations. User recommendations are too easy to fake; most of the people who write them are interested parties. (Our neighborhood video store is offering a free rental if you say good stuff about them on Yelp.) And coverage is usually narrow; there are more sites than people rating them.
Any news on when the rest of the world can finally buy the XO Laptops? Since, well, a 200dpi display for less then $200 sounds like a damn cool device for ebook reading and I really want one.
If you haven't actually seen an XO machine, just pictures, you may not realize that it's much smaller than a standard laptop. The keyboard is sized for kid-sized hands.
Dumb Wired writers, expecting instant gratification. Wired used to have reporters who actually went out and covered real stuff. Then they laid off most of the reporters and kept the "editors". Now they're just wannabe pundits. Saves on travel expenses.
That Tired writer isn't coming across as someone who spent long days digging something out of library stacks or public records. Or travelling around asking people questions to find out what really happened, like a real reporter. This is a lightweight. If you want a children's encyclopedia, you can still get World Book.
Wikipedia has many problems, of course. Most of the good articles were in the first 500,000 created. What's coming in now is mostly junk - "State Route 92", "Star Wars Furry Adventure #6659", and similar crap. Wikia offers some hope for an amusing reason. Wikia took over Wookiepedia, the repository of Star Wars fancruft, which generates most of Wikia's traffic. They're monetizing the fan base. Over time, maybe all the popular culture stuff can be moved to Wikia. That would be a win.
Here's the actual paper. It's a Usenix paper.
What they're doing is straightforward, and it's much like what many virus scanners do. First, they look at web pages to see if there's anything suspicious that requires further analysis. If there is, they load the page into Internet Explorer (of course) in a virtual machine, and see if it changes its environment. The better virus scanners have been doing something like that for a few years now, running possible viruses in some kind of sandbox. Although they usually don't go all the way and run Internet Explorer in a virtual machine. (Are you allowed to do that under Microsoft's current EULA for IE 7?)
The main problem with Google's approach here is that it's after the fact. They won't notice a bad page until the next time they crawl it. Bad pages come and go so fast today that they'll always be behind. As the paper says, "Since many of the malicious URLs are too short-lived to provide statistically meaningful data, we analyzed only the URLs whose presence on the Internet lasted longer than one week."
If Google implements this, the main effect will be to push attackers into changing site names for attack sites even faster.
It's all so backward. What we need is to run most of Internet Explorer in a tightly sandboxed environment on the user's machine, so that when you close the window, any browser damage goes away. That would actually work.