Mod parent up. PayPal needs to become a regulated bank. Until then, take your business elsewhere, to sites that accept credit cards. If someone can't qualify for a merchant account, you probably don't want to deal with them anyway.
RSS supports an "etag", which is supposed to indicate if the feed has something new. When the client polls an RSS server, and gets XML data, the data includes an etag value. When you poll an RSS feed, you supply the etag value from last time. If the feed hasn't changed, the client is supposed to get a 304 status.
Some RSS feed servers implement this, and some don't. Twitter doesn't. So polling Twitter via RSS results in far more network traffic than it should, and extra client work throwing out duplicates.
Google Voice, by the way, has the same problem with their API.
The way out of this is noise cancellation. That's available, and not all that expensive, but not standard. If players were required to have hardware support for noise-canceling earbuds, then the temptation to raise the output level would decline. You can get noise-canceling earbuds now, but they need an external unit with another battery and electronics to do the noise cancellation. If the cancellation electronics is moved to the player, where it should be, the overall cost will decline. Also, you get rid of the need for a second battery.
They need to come up with something that lets you read each item exactly once - no duplicates, and no missing items.
Here's a Twitter feed in XML. That's updated every 60 seconds, and if you miss something, it's gone.
Twitter has an RSS feed capability, but it doesn't properly support the "already read this ID" feature; every poll gets you a dump of the relevant inbox. (Some server side implementations of RSS get this right. Some have problems because they're front-ended by caching servers which lack cache coherency, and you get a different message ID depending on which server you happen to reach.
One of the fundamental problems in computer science is arranging things so that B keeps up with changes at A. There are known good solutions, (see "two-phase commit") and we're not seeing them enough. Even email doesn't do it right.
I looked into microcontrollers like six or seven years ago and was pretty much scared away.
The Arduno's ecosystem has helped in that area. There was a previous generation of microcontrollers with hobbyist support,
the PIC and the Basic Stamp. Those devices were getting rather dated; the Basic Stamp is descended from a 1970s National Semiconductor part. Moving to Atmel's ATmega128 was a step up, with 32-bit registers and a hardware multiplier. The industrial world made that step up a decade ago, but the hobbyist world was still struggling along with limited hardware. This is one of the reasons that entry-level hobbyist robots hadn't gotten much smarter for over a decade.
Although Atmel offered a complete set of free development tools for the ATmega line, they were never presented in a hobbyist-friendly format. Atmel has a huge range of products, and this is just one of many. It's not at all obvious what to order and download.
The Arduno cult is about branding, not technology. The CPU is an ATMega 128, a good little microcontroller. Boards for that CPU have been available for years. I was using this one years before the cult. It's Atmel that made this all possible, by building a microcontroller that requires very few external components to program and debug.
The Arduno people have their own language and terminology, talking about "shields" (daughterboards) and such. Too cult-like.
The launch history of the Bulava is discussed here. It's worked a few times, but they've been having failures in minor components like explosive bolts. That indicates quality control problems in the supply chain, not design problems.
It's hard to restart an entire high-tech supply chain when there hasn't been any demand for years. The US lost the ability to build nuclear weapons for over a decade.
The Russians have admitted it was an ICBM test. They were trying a launch from a nuclear submarine from a submerged position in the White Sea. The third stage of the ICBM failed.
The link in the article is an excellent example of how to make your web page load slowly by making it wait for slow ad servers. I gave up after 15 seconds.
And people wonder why the print media can't succeed on line.
There wasn't a known Proton-M launch this week. There was one on November 25, 2009. The article cited by the "anonymous poster" is from a 2006 launch.
Proton-M rockets are huge, and are launched from Baikonur in Kazakhstan. When one of those things is launched, it gets noticed.
Somebody local may have launched something, but it probably wasn't launched 1000 km away from Kazakhstan. There are pictures from multiple locations, and many show ground details. So get those pictures geo-located and aligned, and figure out where
it came from.
What would Google think if someone released their customer list?
We have it. A sample of Google AdWords advertisers:
saarc.autodesk.com
safeguarddd.com
safestepproducts.com
safetyawarenessposters.com
safetyproductsllc.com
safetyrailsource.com
sagemas.com
sagepayservices.com
sagonet.com
saideigama.com
There are about 22,000 Google AdWords customers known to us. Every time Google puts up an AdWords ad, it exposes the identity of the advertiser. Our AdRater browser plug-in rates on-line advertisers as their ads are presented to users. Unlike most plug-ins, we don't monitor user behavior. Instead, we monitor advertiser behavior, which is in some ways more interesting.
This doesn't violate Google's terms of service. Every request made of Google was made by a user, not us, during ordinary browsing. We're just watching the ads go by. It's like clipping ads from newspapers to see what your competitors are doing.
As we point out occasionally, about 35% of Google's advertisers are "bottom feeders". Google needs to raise the bar on who can run ads with them. Search Google for "Craigslist auto posting tool" and look at the paid ads. You can buy "Easy Ad Poster Deluxe", a program for spamming Craigslist, through Google Checkout, so Google isn't just advertising it, they're taking a cut of the revenue as well. That's embarrassing for Google, or should be.
I've seen that, too. Recently, Stanford University came up
on our short list of major sites being exploited by phishers. I was surprised, because Stanford is usually good about stopping that. It was a weird subdomain under "stanford.edu", and at first I thought someone had compromised Stanford's DNS to get their site under the "stanford.edu" domain. But no, it was just some minor machine that had had a break-in.
The directory with the phishing page was readable as a web page and contained the log of captured passwords, so I sent those to Stanford security and Bank of America security. Haven't heard back from either. After the end of the weekend, the site was taken down, and that took Stanford off the blacklist.
We've been reasonably successful at cleaning up that list. We're trying to popularize the idea that one verified phishing URL blacklists the whole domain until the problem is fixed. (The idea behind SiteTruth is to take a hard-line approach and measure the collateral damage so it can be minimized.) The oldest sites on that list are ones which won't respond to complaints by e-mail or phone. In some cases we've sent faxes.
The worst offenders are Piczo and FortuneCity. Piczo is some kind of social network/hosting service for teenage girls, and it's full of phishing pages, mostly for Habbo logins. PhishTank counts 15, and there are probably more. The phony pages are often not in English, and the Piczo abuse department may not recognize a French Habbo phishing page. This may be the next trend in phishing - put your page on a site run by someone unlikely to understand the page. I've seen a phishing page in Greek on an Indian site.
It's getting harder to run a phishing site. Since the end of "domain tasting", the business of high-volume bogus domain registration has tapered off. We haven't seen an "open redirector" on a major site in a while; eBay, Yahoo, and Microsoft Live all used to have at least one. The "url shorteners" are getting very aggressive about killing links to phishing sites. This might be winnable.
This is more like a radio station promotion. It would have worked if one of those blowhards on AM talk radio had announced a similar hunt with a call-in number. It didn't need the Internet.
There's still a role for air superiority fighters. Even if they're not used much in that role, if you don't have them, the other side has air superiority, which is Not Fun. The USAF likes to say that American troops have not had to fight under a hostile sky since WWII, and this did not happen by accident. They have a point.
Recon and close air support, though, is going to go UAV. Using an F-16 to take out a truck is not only overkill, you don't have enough fighters to do it very often. The big advantage of UAVs is that the US can afford lots of them, and can keep them airborne so that there's one in the neighborhood when needed. The US has plenty of heavy weapon systems that can take out an visible enemy concentration. As a result, the surviving enemies of the US hide and don't bunch up. Air support is more of a retail operation now than a wholesale one.
Here's the
map of existing coverage. The continental US, Europe, and Japan, have full coastal coverage. The port coasts of China and Australia are covered. Beyond that, not so much.
This isn't a safety system. It's for traffic and port management. Vessels show up in the system around the time when ports need to start thinking about where to put them.
They'll go to Weird Stuff Warehouse, the surplus place near Yahoo HQ. Want a few hundred servers cheap? They have them stacked up.
Dismantling a failed company is routine in Silicon Valley. Big assets are auctioned off by DoveBid. Miscellaneous computers go to Weird Stuff. Furniture and partitions go to Consolidated Office Outfitters. In less than a month, the building will be empty and ready for rental.
Makes sense. A stealthed recon aircraft should be small. Recon is mostly flying preprogrammed flight paths, so the pilot doesn't make many decisions. Hence a moderate-sized UAV.
The Air Force guys hate it, but UAVs are getting the job done. The Army is going for more automation; they use autoland on their Predators, and have far fewer crashes than the USAF stick jocks who land the things manually.
This would have been great for open source if Firefox 3 didn't suck. (There are no less that 19 threads on the Firefox forums containing the phrase "Firefox 3 sucks".) Firefox 2 was small, fast, and reliable. Firefox 3, even now, is less reliable than Firefox 2, slower, and a memory hog. Open source had its chance and blew it.
Salon is still around? I remember them from 2000 or so, but one day I hit a paywall and never went back. They also bought The Well, and managed to screw that up, too.
X would cover fixed costs such as the cost of billing.
Billing is a very large cost. In the telco world, the cost of billing passed the cost of transmission about two decades ago. That's even more true for Internet transmission at the retail level.
With proposed "economic" solutions, you have to factor in the costs of billing. Those costs apply both to the provider and the customer. If customers have to meter to control their expenses, that's a cost to the customer in attention, and drives business to competitors that require less attention.
Handling one phone call about a billing complaint eats up months of profits from selling the service. Complex billing means big call centers.
Slashdot Mirror
Mod parent up. PayPal needs to become a regulated bank. Until then, take your business elsewhere, to sites that accept credit cards. If someone can't qualify for a merchant account, you probably don't want to deal with them anyway.
RSS supports an "etag", which is supposed to indicate if the feed has something new. When the client polls an RSS server, and gets XML data, the data includes an etag value. When you poll an RSS feed, you supply the etag value from last time. If the feed hasn't changed, the client is supposed to get a 304 status.
Some RSS feed servers implement this, and some don't. Twitter doesn't. So polling Twitter via RSS results in far more network traffic than it should, and extra client work throwing out duplicates.
Google Voice, by the way, has the same problem with their API.
The way out of this is noise cancellation. That's available, and not all that expensive, but not standard. If players were required to have hardware support for noise-canceling earbuds, then the temptation to raise the output level would decline. You can get noise-canceling earbuds now, but they need an external unit with another battery and electronics to do the noise cancellation. If the cancellation electronics is moved to the player, where it should be, the overall cost will decline. Also, you get rid of the need for a second battery.
They need to come up with something that lets you read each item exactly once - no duplicates, and no missing items.
Here's a Twitter feed in XML. That's updated every 60 seconds, and if you miss something, it's gone.
Twitter has an RSS feed capability, but it doesn't properly support the "already read this ID" feature; every poll gets you a dump of the relevant inbox. (Some server side implementations of RSS get this right. Some have problems because they're front-ended by caching servers which lack cache coherency, and you get a different message ID depending on which server you happen to reach.
One of the fundamental problems in computer science is arranging things so that B keeps up with changes at A. There are known good solutions, (see "two-phase commit") and we're not seeing them enough. Even email doesn't do it right.
Clark Gable drove a custom built 1936 Duesenberg Speedster [howstuffworks.com] that likely cost him nothing more than the price of a fill-up.
That's nothing. Arnold Schwarzenegger wanted a personal HUMMV, and the entire civilian Hummer business was created for him.
I looked into microcontrollers like six or seven years ago and was pretty much scared away.
The Arduno's ecosystem has helped in that area. There was a previous generation of microcontrollers with hobbyist support, the PIC and the Basic Stamp. Those devices were getting rather dated; the Basic Stamp is descended from a 1970s National Semiconductor part. Moving to Atmel's ATmega128 was a step up, with 32-bit registers and a hardware multiplier. The industrial world made that step up a decade ago, but the hobbyist world was still struggling along with limited hardware. This is one of the reasons that entry-level hobbyist robots hadn't gotten much smarter for over a decade.
Although Atmel offered a complete set of free development tools for the ATmega line, they were never presented in a hobbyist-friendly format. Atmel has a huge range of products, and this is just one of many. It's not at all obvious what to order and download.
The Arduno cult is about branding, not technology. The CPU is an ATMega 128, a good little microcontroller. Boards for that CPU have been available for years. I was using this one years before the cult. It's Atmel that made this all possible, by building a microcontroller that requires very few external components to program and debug.
The Arduno people have their own language and terminology, talking about "shields" (daughterboards) and such. Too cult-like.
This seems to involve some feud over clones of renamed vaporware products. I think.
Whatever.
The launch history of the Bulava is discussed here. It's worked a few times, but they've been having failures in minor components like explosive bolts. That indicates quality control problems in the supply chain, not design problems.
It's hard to restart an entire high-tech supply chain when there hasn't been any demand for years. The US lost the ability to build nuclear weapons for over a decade.
as far as I know being able to launch a missile while the sub is submerged would be a huge leap forward in the nuclear arms race.
It was, when the US and USSR both achieved it in 1960.
The Russians have admitted it was an ICBM test. They were trying a launch from a nuclear submarine from a submerged position in the White Sea. The third stage of the ICBM failed.
The link in the article is an excellent example of how to make your web page load slowly by making it wait for slow ad servers. I gave up after 15 seconds.
And people wonder why the print media can't succeed on line.
There wasn't a known Proton-M launch this week. There was one on November 25, 2009. The article cited by the "anonymous poster" is from a 2006 launch.
Proton-M rockets are huge, and are launched from Baikonur in Kazakhstan. When one of those things is launched, it gets noticed.
Somebody local may have launched something, but it probably wasn't launched 1000 km away from Kazakhstan. There are pictures from multiple locations, and many show ground details. So get those pictures geo-located and aligned, and figure out where it came from.
What would Google think if someone released their customer list?
We have it. A sample of Google AdWords advertisers:
There are about 22,000 Google AdWords customers known to us. Every time Google puts up an AdWords ad, it exposes the identity of the advertiser. Our AdRater browser plug-in rates on-line advertisers as their ads are presented to users. Unlike most plug-ins, we don't monitor user behavior. Instead, we monitor advertiser behavior, which is in some ways more interesting. This doesn't violate Google's terms of service. Every request made of Google was made by a user, not us, during ordinary browsing. We're just watching the ads go by. It's like clipping ads from newspapers to see what your competitors are doing.
As we point out occasionally, about 35% of Google's advertisers are "bottom feeders". Google needs to raise the bar on who can run ads with them. Search Google for "Craigslist auto posting tool" and look at the paid ads. You can buy "Easy Ad Poster Deluxe", a program for spamming Craigslist, through Google Checkout, so Google isn't just advertising it, they're taking a cut of the revenue as well. That's embarrassing for Google, or should be.
I've seen that, too. Recently, Stanford University came up on our short list of major sites being exploited by phishers. I was surprised, because Stanford is usually good about stopping that. It was a weird subdomain under "stanford.edu", and at first I thought someone had compromised Stanford's DNS to get their site under the "stanford.edu" domain. But no, it was just some minor machine that had had a break-in.
The directory with the phishing page was readable as a web page and contained the log of captured passwords, so I sent those to Stanford security and Bank of America security. Haven't heard back from either. After the end of the weekend, the site was taken down, and that took Stanford off the blacklist.
We've been reasonably successful at cleaning up that list. We're trying to popularize the idea that one verified phishing URL blacklists the whole domain until the problem is fixed. (The idea behind SiteTruth is to take a hard-line approach and measure the collateral damage so it can be minimized.) The oldest sites on that list are ones which won't respond to complaints by e-mail or phone. In some cases we've sent faxes.
The worst offenders are Piczo and FortuneCity. Piczo is some kind of social network/hosting service for teenage girls, and it's full of phishing pages, mostly for Habbo logins. PhishTank counts 15, and there are probably more. The phony pages are often not in English, and the Piczo abuse department may not recognize a French Habbo phishing page. This may be the next trend in phishing - put your page on a site run by someone unlikely to understand the page. I've seen a phishing page in Greek on an Indian site.
It's getting harder to run a phishing site. Since the end of "domain tasting", the business of high-volume bogus domain registration has tapered off. We haven't seen an "open redirector" on a major site in a while; eBay, Yahoo, and Microsoft Live all used to have at least one. The "url shorteners" are getting very aggressive about killing links to phishing sites. This might be winnable.
This is more like a radio station promotion. It would have worked if one of those blowhards on AM talk radio had announced a similar hunt with a call-in number. It didn't need the Internet.
I'd suggest Scroogle (https://ssl.scroogle.org/ -- Google sans the crap), but it seems down at the moment.
Scroogle has limits on how much you can use it. If you use it too much, your IP address will be blocked and the site won't answer you at all.
There are now four or five e-book readers,
And since I wrote that a few days ago, there's another one, with its very own incompatible-by-design ecosystem.
There's still a role for air superiority fighters. Even if they're not used much in that role, if you don't have them, the other side has air superiority, which is Not Fun. The USAF likes to say that American troops have not had to fight under a hostile sky since WWII, and this did not happen by accident. They have a point.
Recon and close air support, though, is going to go UAV. Using an F-16 to take out a truck is not only overkill, you don't have enough fighters to do it very often. The big advantage of UAVs is that the US can afford lots of them, and can keep them airborne so that there's one in the neighborhood when needed. The US has plenty of heavy weapon systems that can take out an visible enemy concentration. As a result, the surviving enemies of the US hide and don't bunch up. Air support is more of a retail operation now than a wholesale one.
Here's the map of existing coverage. The continental US, Europe, and Japan, have full coastal coverage. The port coasts of China and Australia are covered. Beyond that, not so much.
This isn't a safety system. It's for traffic and port management. Vessels show up in the system around the time when ports need to start thinking about where to put them.
They'll go to Weird Stuff Warehouse, the surplus place near Yahoo HQ. Want a few hundred servers cheap? They have them stacked up.
Dismantling a failed company is routine in Silicon Valley. Big assets are auctioned off by DoveBid. Miscellaneous computers go to Weird Stuff. Furniture and partitions go to Consolidated Office Outfitters. In less than a month, the building will be empty and ready for rental.
Makes sense. A stealthed recon aircraft should be small. Recon is mostly flying preprogrammed flight paths, so the pilot doesn't make many decisions. Hence a moderate-sized UAV.
The Air Force guys hate it, but UAVs are getting the job done. The Army is going for more automation; they use autoland on their Predators, and have far fewer crashes than the USAF stick jocks who land the things manually.
This would have been great for open source if Firefox 3 didn't suck. (There are no less that 19 threads on the Firefox forums containing the phrase "Firefox 3 sucks".) Firefox 2 was small, fast, and reliable. Firefox 3, even now, is less reliable than Firefox 2, slower, and a memory hog. Open source had its chance and blew it.
Salon is still around? I remember them from 2000 or so, but one day I hit a paywall and never went back. They also bought The Well, and managed to screw that up, too.
X would cover fixed costs such as the cost of billing.
Billing is a very large cost. In the telco world, the cost of billing passed the cost of transmission about two decades ago. That's even more true for Internet transmission at the retail level. With proposed "economic" solutions, you have to factor in the costs of billing. Those costs apply both to the provider and the customer. If customers have to meter to control their expenses, that's a cost to the customer in attention, and drives business to competitors that require less attention.
Handling one phone call about a billing complaint eats up months of profits from selling the service. Complex billing means big call centers.