Actually, an AP bridges from wired to wireless. The WET11 can take a pod of up to 30 wired machines (or rather, machines attached to a single Ethernet segment) and bridge all of their traffic to ANOTHER access point. The WET11 is a wired-to-wireless bridge that hooks into a wireless-to-wired bridge.
Put 30 machines in a room on one side of the San Francisco Bay, plug in a WET11, point an antenna to an AP on the other side, and presto: you're running a large (very latent) Ethernet network.
We're talking two issues here: cheap and simple. I don't think the average consumer would spend $1,300 to get the wireless bridge + AP that the Cisco gear would cost, and I don't think the average consumer could figure out how to configure it!
The WET11 is pretty straightforward, just as hard as hooking up a client to an AP, and it's dirt cheap.
I'm sure this sounds confusing. If you're already running a wireless network or setting one up, you need at least one AP running infrastructure mode. If you wanted to add a WAP11-based bridge, you had to add TWO WAP11s--one for each side of the bridge. You cannot run a WAP11 as an AP and a bridge; it's essentially too separate modes. Many, many people wrote me asking why they couldn't associate to the WAP11 as an AP and have it bridge, and I said, ask the firmware makers.
So the WET11 reduces your equipment needs and also allows you to go generic: you don't need a Linksys on the other side.
First, which kind of wireless networks doesn't it belong on? It might be a bad addition to certain topologies that lack redundancy or are already crowded or rely on community cooperation. But it's a great addition to many kinds of networks in which you're looking to span or create a loose, fake mesh (it doesn't have mesh routing protocols, but i wouldn't be surprised to see hacks when mesh routing becomes an open-source reality).
Second, security is definitely an issue because you're beaming a ton of network traffic over the link. But because it's a client association session, someone can't just tune into your WET11 and monitor traffic; they have to get access to the AP that it's connecting to.
It's very rare to see a Slashdot post about an event far away without hearing from an on-site participant. The press release went out about this new service down in Kiwi-ville, and some of the specs they describe seem, well, a little difficult to swallow as they exceed some of the physics and technology that major manufacturers are employing.
Any Kiwis read Slashdot and can confirm coverage? Or is this Slashdot-by-press-release?
3G is low speed and it doesn't exist everywhere, nor will it be ubiquitous. There are a few popular places where companies are trying to roll it out, but the overall cost is somewhere north of $100 billion to get major cities hooked up.
3G is microcell based, meaning that you have relatively high power transmission compared with Wi-Fi, but large enough cells that you have a lot of people sharing a very few available channels. Thus when more than a handful of people are using 3G data services, the 100 Kbps or 300 Kbps or whatever they claim today as a maximum is split down into 3K chunks.
With Wi-Fi, because it's picocell, tiny itty bitty cells, you can typically increase density (and the equipment's cheaper and requires fewer towers or other spots to make work) and keep overall bandwidth closer to the 4 Mbps that most devices throughput.
They do have that motivation, but they also make many many hundreds of millions of dollars off Mac users from sales of Office, from mice sales (their latest mice have Mac drivers), and on and on.
Yesterday, they announced they're writing a ground-up version of the MSN 8 client for Mac OS X. They can make a lot of money as an ISP, too.
If Apple still had the market share it used to, the conspiracy theory might have made sense. But because Microsoft can literally make as much or more money from a Mac user's "seat" as they can from a Windows user's, it's a moot point.
The new OS X Remote Desktop Client lets them sell more Windows 2000 Server licenses to companies with Mac users, who can now "login" as Windows terminal session. Lots of Mac users in mixed environments have had to buy Virtual PC before this (which pays a fee for every copy sold to MSFT as well!).
There's a good money for MS, and they follow the money.
Actually, because of the flaws in the other Mac OS X browsers that I've tried to date, including horrible crashes and occasional disk corruption with earlier versions of Netscape 6, the fact that rendering is slow is a small price to pay for standards-compliant CSS, JavaScript, DOM.
I may be silly, and I hate the slow rendering, but IE 5.2.1 for OS X typically gets it right when other browsers seem to mess up on the details.
Seattle's county, King County, handles solid waste disposal, and launched a project two years ago that turned into a pretty serious change in how computers are disposed of at dumps. Yes, people can still try to slip in electronics, but you can no longer drop off monitors, and other CRTs will follow. The county works with local businesses, and has found safe and well-documented U.S. sources to send the products to.
For instance, monitors are disassembled, and the tubes sent to Pennsylvania, where the glass is smelted, and the lead separate for reuse. (The poster who mentioned that LCDs change this equation are right: no new smelters for recycling are being built because CRTs will no longer exist outside specialized uses, so existing smelters will handle the tens of millions of discards.)
Likewise, circuits and other components are sent to companies that often offer job retraining and are nonprofits to safely, under OSHA rules, extract useful materials. One outfit in the SF Bay Area can even get usable epoxy out of circuit boards which can be reused.
The real problem with computing as with white goods (appliances) and other products like cars is that the manufacturers are only required to use safe techniques in building them. Disposal is not part of the price tag. This is changing gradually in Europe, and it's clear to all concerned that if there were a federal mandate, we'd all see savings over the lifecycle of the product: we wouldn't have surprise billion-dollar cleanup funds, and would stop poisoning the rest of the world.
HP and other companies have taken some great steps with toner cartridges and some other limited products that they build in such a way that they can be easily disassembled and much of the parts reused or refashioned.
Existing Mac OS X 10.1 owners, my friend, not all Mac OS owners.
Many 10.1 owners have purchased machines in the last six months or bought full retail copies of the software.
The dispute is: why not offer a $79 or $59 price (something like that) for 10.1 registered owners who purchased a computer or the retail version since, say, March.
But I don't care: they don't serialize the operating system, so I'm happy to purchase it.
Starbucks is smarter than that: MobileStar and now T-Mobile are spending money on Starbucks behalf in order to have the customer base. Starbucks, at last report, has spent $0.00 to put wireless access in. The whole thing sounds like a "Starbucks" issue; it's actually their wireless service provider partner plus Starbucks.
I strongly disagree with your claim that there's a "hump" past which it's easy to write useful code. It takes years to become skilled enough to contribute meaningfully to Linux, or to XFree, or to KDE, or to GCC.
Sorry, I didn't mean useful to everyone, but rather code that does something. Crafts are hard (and fontmaking is an a craft and an art), and programming is also hard. But while you can't create a good work of art -- or most people can't -- that's meaningful in its own right just after learning the rudiments of tools, you can write small prorgams that work and carry out tasks pretty quickly.
But the jump from programs that work to contributions -- I didn't mean to imply that that could happen fast!
This is a very interesting notion: is coding better to learn over years than design/art? I would argue that for someone committed to it, absolutely the same difference.
What I've worried about in this thread has been an idea that you could sit down and in a matter of weeks, start making fonts that other people would feel good about using.
The difference between coding and designing fonts is surely that you can start coding useful stuff right away, typically, after you cross an early hump on understanding programming at all. Hello world! But with font design, you might spend literally months and months of intense work before you could make a font that someone else would find passable.
So my objection here hasn't been that people can't learn, but rather that a lot of folks would be spending a lot of town on a futile activity if they didn't pursue it for quite a while rather than perhaps focusing energy in areas which would be challenging but doable over different spans of effort.
It's surely one of those, did you major in computer science or art questions. I majored in art, but I spend a good part of each week writing perl and PHP.
I can't argue with such lovely optimism and good feelings. Despite my disagreement with some of the points you make, it's very hard to not have a big smile on my face after reading your post!
Hundreds of people learning to make fonts won't result in good fonts, just hundreds of people frustrated at the amount of time they wasted in making fonts nobody uses.
Designing fonts from scratch takes years to learn; even copying fonts takes quite a while. I've worked with type designers and have, in fact, created my own fonts, one of which is a rendition of an older font (from the 30s) called Albertus.
It's a tedious process even with good tools. It's mostly about drawing and then matching those drawings to PostScript-possible splines.
Unlike kernel development or software collaboration, in which hundreds of people can each contribute something that winds up in the final results (or even tens of thousands), font design is a lonely profession with lots of abandoned work.
Weirdly, none of us have thought about a blog or journal on the case. I wonder what our lawyers will think? EFF has a truly terrific, hip bunch of people behind this (not just saying that because they read Slashdot), and I wouldn't be surprised if we could pull something off like that. Thanks for the suggestion!
Yeah, when Larry Lessig said at OSCon, what are you doing? I thought -- Hey, I'm actually doing something! I hope to attend the actual trial.
I'm one of the defendants -- why doesn't anyone ever say the suit includes regular Slashdot reader Glenn Fleishman? cuz Craig is arguably much cooler than I. One large part of my involvement in the suit is that I don't believe that any company nor the government should be allowed to outlaw devices or uses or media formats before or after the fact because there simply might be some ways in which that technology could infringe on copyright.
Copyright is held in the public interest -- it's part of the public good as a means to ensure the creation and dissemination of knowledge. Fair use is a tool to allow individuals to have reasonable access and use of materials they license or buy from copyright holders. With the expansion of copyright law, there's no connection any more between the notion of copyright as a limited grant by the people of the United States (and other countries, too, of course) and the utility to which that copyright can be put to use.
I'm an author as well as a defendent in this case, and I support copyright as a method by which words, images, and motion can be protected for a limited time to allow the artists, writers, and other creators to make a living. If other modalities arise in which I would copyright nothing but still be able to pay the bills, I would certainly be interested in that as would most authors I know.
The point is this: I don't ask Xerox and Canon to stop selling copy machines because they might photocopy articles that appear in magazines. I don't ask ISPs to filter all content because my words might pass through without payment. I don't require my readers to peruse advertisements and read my articles in one sitting. (You can make the case that one useful item built into new color copiers is their ability to recognize when currency is being photocopied and prevent it -- that has compelling public and private interest all over it, even though it prevents certain kinds of art.)
This is why it's useful to save junk faxes you receive: if you report them, they migth get added to a large case, and then help shut down a junk faxer.
We get waves of junk fax, all crap -- nothing even marginally businesslike (get free vacations, buy toner cheap, etc.). They all look crappy, all have stupid calls to action, typically no easy way to reply even!
I started faxing the faxers back saying, hi, thanks for giving us all your contact information; here's the relevant US statute that proves what you just did is illegal and we know how to find you. Never send us anything again, or we'll pursue all civil options available to us.
Oddly, the faxes have stopped. Who knew that would work. Now if it only worked with spammers!
What, the fact that they publish interesting stories (some written by yours truly) and then have the temerity to link to books they sell...the whole article is thus an ad? Yeah, they sell a book on networking your car, using talkd, and driving to a perl conference...
It depends on whether you're bridging or routing. Obilix notes that the MAC address is stripped, but that's only if you're routing. Many DSL and cable networks use a pure bridge mode in which you're essentially a leg of a larger Ethernet network instead of on your own IP subnet. Because of this, it would be possible for the ISP to see what kind of traffic is happening, although technically, the Ethernet packet that would get routed out over the bridged network on their end would have a MAC address of the NAT gateway machine, not the individual machines. But that traffic could be monitored, pretty easily.
Also, NAT patterns, even if they're over routed connections, can be identified. A lot of NAT gateways aren't very clever because they don't need to be. If you're an ISP trying to "fight" address sharing, you could set up a cheap lab with typical consumer equipment, most of which shares the same firmware from Atmel (Linksys, SMC, Bufffalo, D-Link, etc.). You could use some very simple analysis software to watch traffic and identify gateways.
Fortunately, the makers of these inexpensive NAT hardware gateways have typically responded (or their firmware developers have that they license from) and added anti-anti-ballistic missile code. It wasn't very long after cable companies were restricting access to a single MAC address that Linksys and other devices added a 'clone MAC address' option.
Reverend Matt, you're speaking my religion: too many folks raise the bloody flag of the 1st amendment when this is clearly commercial speech and commercial contract. Great post!
The FCC was obliged in the late 1990s to reallocate the spectrum currently allotted to UHF 50 to 59 and 60 to 72 in multiple auctions after Digital TV (DTV) was in full swing. There's nothing new about this. The story at the top of this page makes it sound as though this is a sudden effort to steal UHF for wireless. It ain't.
The broader public interest issue was debated and buried and lost years ago, and the juggernaut of DTV has moved a few inches, not toppling the analog signals as were expected.
The UHF broadcasters, just like everyone else, have been assigned new DTV frequencies, but it's ridiculous to ask small broadcasters to foot the bill to turn over to DTV, especially with few views and little interest.
But it will happen. The former FCC head, Kennard, said he thought it was more like 2020 instead of 2007 when he spoke on the issue last year.
I was lucky enough to suck it off the Sundance channel via ReplayTV. It took me weeks to watch it. Parts of the film, like the LONG DRIVE BACK INTO THE CITY are hypnotic, mesmering, trying to show us an alien intelligence of alien coldness in our own environment.
Frankly, I'd love to see the less brainy alternative. Tarkovsky's is brilliant, barely accessible, odd. The flip side would be terrifying, fast paced, etc. The book is full of terrifying moments, which I think don't get captured through the intellectualization of Tarkovsky's film, although he captures the horrible, horrible isolation and alienation.
Referers aren't always sent. What would be smart is to use PHP or something similar to validate a referrer via a token. For 99.99% of all cases, this would allow NPR to put up a disclaimer page about the content.
The ombudsman seems to confuse linking with endorsement, partly because NPR makes all its streams available. If the streams were inside a PHP wrapper, a user trying to clickthrough to a stream would first see a page that said, "Hey, this is NPR. We don't have anything to do with where you came from. Click Listen to proceed." The Listen button would use a PHP session token to authenticate to the wrapper.
From within NPR, the site would always just pass these tokens.
The tokens don't have to be secure; most users wouldn't try to bypass them. Those that did for whatever reason would clearly know they were bypassing them. Solves the whole problem. Uses technology instead of lawyers.
Slashdot Mirror
Actually, an AP bridges from wired to wireless. The WET11 can take a pod of up to 30 wired machines (or rather, machines attached to a single Ethernet segment) and bridge all of their traffic to ANOTHER access point. The WET11 is a wired-to-wireless bridge that hooks into a wireless-to-wired bridge.
Put 30 machines in a room on one side of the San Francisco Bay, plug in a WET11, point an antenna to an AP on the other side, and presto: you're running a large (very latent) Ethernet network.
We're talking two issues here: cheap and simple. I don't think the average consumer would spend $1,300 to get the wireless bridge + AP that the Cisco gear would cost, and I don't think the average consumer could figure out how to configure it!
The WET11 is pretty straightforward, just as hard as hooking up a client to an AP, and it's dirt cheap.
I'm sure this sounds confusing. If you're already running a wireless network or setting one up, you need at least one AP running infrastructure mode. If you wanted to add a WAP11-based bridge, you had to add TWO WAP11s--one for each side of the bridge. You cannot run a WAP11 as an AP and a bridge; it's essentially too separate modes. Many, many people wrote me asking why they couldn't associate to the WAP11 as an AP and have it bridge, and I said, ask the firmware makers.
So the WET11 reduces your equipment needs and also allows you to go generic: you don't need a Linksys on the other side.
These are two great questions.
First, which kind of wireless networks doesn't it belong on? It might be a bad addition to certain topologies that lack redundancy or are already crowded or rely on community cooperation. But it's a great addition to many kinds of networks in which you're looking to span or create a loose, fake mesh (it doesn't have mesh routing protocols, but i wouldn't be surprised to see hacks when mesh routing becomes an open-source reality).
Second, security is definitely an issue because you're beaming a ton of network traffic over the link. But because it's a client association session, someone can't just tune into your WET11 and monitor traffic; they have to get access to the AP that it's connecting to.
Yeah, that's right, I can keep a useful history over weeks or months with 320 Gb hard drives littering the place...
Fer chrissakes: mirroring is NOT backing up. $300 to $400 for a drive is not the same as $35 per 60 Gb tape.
It's very rare to see a Slashdot post about an event far away without hearing from an on-site participant. The press release went out about this new service down in Kiwi-ville, and some of the specs they describe seem, well, a little difficult to swallow as they exceed some of the physics and technology that major manufacturers are employing.
Any Kiwis read Slashdot and can confirm coverage? Or is this Slashdot-by-press-release?
My dear Mr. L00zer,
3G is low speed and it doesn't exist everywhere, nor will it be ubiquitous. There are a few popular places where companies are trying to roll it out, but the overall cost is somewhere north of $100 billion to get major cities hooked up.
3G is microcell based, meaning that you have relatively high power transmission compared with Wi-Fi, but large enough cells that you have a lot of people sharing a very few available channels. Thus when more than a handful of people are using 3G data services, the 100 Kbps or 300 Kbps or whatever they claim today as a maximum is split down into 3K chunks.
With Wi-Fi, because it's picocell, tiny itty bitty cells, you can typically increase density (and the equipment's cheaper and requires fewer towers or other spots to make work) and keep overall bandwidth closer to the 4 Mbps that most devices throughput.
They do have that motivation, but they also make many many hundreds of millions of dollars off Mac users from sales of Office, from mice sales (their latest mice have Mac drivers), and on and on.
Yesterday, they announced they're writing a ground-up version of the MSN 8 client for Mac OS X. They can make a lot of money as an ISP, too.
If Apple still had the market share it used to, the conspiracy theory might have made sense. But because Microsoft can literally make as much or more money from a Mac user's "seat" as they can from a Windows user's, it's a moot point.
The new OS X Remote Desktop Client lets them sell more Windows 2000 Server licenses to companies with Mac users, who can now "login" as Windows terminal session. Lots of Mac users in mixed environments have had to buy Virtual PC before this (which pays a fee for every copy sold to MSFT as well!).
There's a good money for MS, and they follow the money.
Actually, because of the flaws in the other Mac OS X browsers that I've tried to date, including horrible crashes and occasional disk corruption with earlier versions of Netscape 6, the fact that rendering is slow is a small price to pay for standards-compliant CSS, JavaScript, DOM.
I may be silly, and I hate the slow rendering, but IE 5.2.1 for OS X typically gets it right when other browsers seem to mess up on the details.
Seattle's county, King County, handles solid waste disposal, and launched a project two years ago that turned into a pretty serious change in how computers are disposed of at dumps. Yes, people can still try to slip in electronics, but you can no longer drop off monitors, and other CRTs will follow. The county works with local businesses, and has found safe and well-documented U.S. sources to send the products to.
For instance, monitors are disassembled, and the tubes sent to Pennsylvania, where the glass is smelted, and the lead separate for reuse. (The poster who mentioned that LCDs change this equation are right: no new smelters for recycling are being built because CRTs will no longer exist outside specialized uses, so existing smelters will handle the tens of millions of discards.)
Likewise, circuits and other components are sent to companies that often offer job retraining and are nonprofits to safely, under OSHA rules, extract useful materials. One outfit in the SF Bay Area can even get usable epoxy out of circuit boards which can be reused.
The real problem with computing as with white goods (appliances) and other products like cars is that the manufacturers are only required to use safe techniques in building them. Disposal is not part of the price tag. This is changing gradually in Europe, and it's clear to all concerned that if there were a federal mandate, we'd all see savings over the lifecycle of the product: we wouldn't have surprise billion-dollar cleanup funds, and would stop poisoning the rest of the world.
HP and other companies have taken some great steps with toner cartridges and some other limited products that they build in such a way that they can be easily disassembled and much of the parts reused or refashioned.
Existing Mac OS X 10.1 owners, my friend, not all Mac OS owners.
Many 10.1 owners have purchased machines in the last six months or bought full retail copies of the software.
The dispute is: why not offer a $79 or $59 price (something like that) for 10.1 registered owners who purchased a computer or the retail version since, say, March.
But I don't care: they don't serialize the operating system, so I'm happy to purchase it.
Starbucks is smarter than that: MobileStar and now T-Mobile are spending money on Starbucks behalf in order to have the customer base. Starbucks, at last report, has spent $0.00 to put wireless access in. The whole thing sounds like a "Starbucks" issue; it's actually their wireless service provider partner plus Starbucks.
I strongly disagree with your claim that there's a "hump" past which it's easy to write useful code. It takes years to become skilled enough to contribute meaningfully to Linux, or to XFree, or to KDE, or to GCC.
Sorry, I didn't mean useful to everyone, but rather code that does something. Crafts are hard (and fontmaking is an a craft and an art), and programming is also hard. But while you can't create a good work of art -- or most people can't -- that's meaningful in its own right just after learning the rudiments of tools, you can write small prorgams that work and carry out tasks pretty quickly.
But the jump from programs that work to contributions -- I didn't mean to imply that that could happen fast!
This is a very interesting notion: is coding better to learn over years than design/art? I would argue that for someone committed to it, absolutely the same difference.
What I've worried about in this thread has been an idea that you could sit down and in a matter of weeks, start making fonts that other people would feel good about using.
The difference between coding and designing fonts is surely that you can start coding useful stuff right away, typically, after you cross an early hump on understanding programming at all. Hello world! But with font design, you might spend literally months and months of intense work before you could make a font that someone else would find passable.
So my objection here hasn't been that people can't learn, but rather that a lot of folks would be spending a lot of town on a futile activity if they didn't pursue it for quite a while rather than perhaps focusing energy in areas which would be challenging but doable over different spans of effort.
It's surely one of those, did you major in computer science or art questions. I majored in art, but I spend a good part of each week writing perl and PHP.
I can't argue with such lovely optimism and good feelings. Despite my disagreement with some of the points you make, it's very hard to not have a big smile on my face after reading your post!
Hundreds of people learning to make fonts won't result in good fonts, just hundreds of people frustrated at the amount of time they wasted in making fonts nobody uses.
Designing fonts from scratch takes years to learn; even copying fonts takes quite a while. I've worked with type designers and have, in fact, created my own fonts, one of which is a rendition of an older font (from the 30s) called Albertus.
It's a tedious process even with good tools. It's mostly about drawing and then matching those drawings to PostScript-possible splines.
Unlike kernel development or software collaboration, in which hundreds of people can each contribute something that winds up in the final results (or even tens of thousands), font design is a lonely profession with lots of abandoned work.
Weirdly, none of us have thought about a blog or journal on the case. I wonder what our lawyers will think? EFF has a truly terrific, hip bunch of people behind this (not just saying that because they read Slashdot), and I wouldn't be surprised if we could pull something off like that. Thanks for the suggestion!
Yeah, when Larry Lessig said at OSCon, what are you doing? I thought -- Hey, I'm actually doing something! I hope to attend the actual trial.
I'm one of the defendants -- why doesn't anyone ever say the suit includes regular Slashdot reader Glenn Fleishman? cuz Craig is arguably much cooler than I. One large part of my involvement in the suit is that I don't believe that any company nor the government should be allowed to outlaw devices or uses or media formats before or after the fact because there simply might be some ways in which that technology could infringe on copyright.
Copyright is held in the public interest -- it's part of the public good as a means to ensure the creation and dissemination of knowledge. Fair use is a tool to allow individuals to have reasonable access and use of materials they license or buy from copyright holders. With the expansion of copyright law, there's no connection any more between the notion of copyright as a limited grant by the people of the United States (and other countries, too, of course) and the utility to which that copyright can be put to use.
I'm an author as well as a defendent in this case, and I support copyright as a method by which words, images, and motion can be protected for a limited time to allow the artists, writers, and other creators to make a living. If other modalities arise in which I would copyright nothing but still be able to pay the bills, I would certainly be interested in that as would most authors I know.
The point is this: I don't ask Xerox and Canon to stop selling copy machines because they might photocopy articles that appear in magazines. I don't ask ISPs to filter all content because my words might pass through without payment. I don't require my readers to peruse advertisements and read my articles in one sitting. (You can make the case that one useful item built into new color copiers is their ability to recognize when currency is being photocopied and prevent it -- that has compelling public and private interest all over it, even though it prevents certain kinds of art.)
This is why it's useful to save junk faxes you receive: if you report them, they migth get added to a large case, and then help shut down a junk faxer.
We get waves of junk fax, all crap -- nothing even marginally businesslike (get free vacations, buy toner cheap, etc.). They all look crappy, all have stupid calls to action, typically no easy way to reply even!
I started faxing the faxers back saying, hi, thanks for giving us all your contact information; here's the relevant US statute that proves what you just did is illegal and we know how to find you. Never send us anything again, or we'll pursue all civil options available to us.
Oddly, the faxes have stopped. Who knew that would work. Now if it only worked with spammers!
What, the fact that they publish interesting stories (some written by yours truly) and then have the temerity to link to books they sell...the whole article is thus an ad? Yeah, they sell a book on networking your car, using talkd, and driving to a perl conference...
It depends on whether you're bridging or routing. Obilix notes that the MAC address is stripped, but that's only if you're routing. Many DSL and cable networks use a pure bridge mode in which you're essentially a leg of a larger Ethernet network instead of on your own IP subnet. Because of this, it would be possible for the ISP to see what kind of traffic is happening, although technically, the Ethernet packet that would get routed out over the bridged network on their end would have a MAC address of the NAT gateway machine, not the individual machines. But that traffic could be monitored, pretty easily.
Also, NAT patterns, even if they're over routed connections, can be identified. A lot of NAT gateways aren't very clever because they don't need to be. If you're an ISP trying to "fight" address sharing, you could set up a cheap lab with typical consumer equipment, most of which shares the same firmware from Atmel (Linksys, SMC, Bufffalo, D-Link, etc.). You could use some very simple analysis software to watch traffic and identify gateways.
Fortunately, the makers of these inexpensive NAT hardware gateways have typically responded (or their firmware developers have that they license from) and added anti-anti-ballistic missile code. It wasn't very long after cable companies were restricting access to a single MAC address that Linksys and other devices added a 'clone MAC address' option.
Reverend Matt, you're speaking my religion: too many folks raise the bloody flag of the 1st amendment when this is clearly commercial speech and commercial contract. Great post!
The FCC was obliged in the late 1990s to reallocate the spectrum currently allotted to UHF 50 to 59 and 60 to 72 in multiple auctions after Digital TV (DTV) was in full swing. There's nothing new about this. The story at the top of this page makes it sound as though this is a sudden effort to steal UHF for wireless. It ain't.
The broader public interest issue was debated and buried and lost years ago, and the juggernaut of DTV has moved a few inches, not toppling the analog signals as were expected.
The UHF broadcasters, just like everyone else, have been assigned new DTV frequencies, but it's ridiculous to ask small broadcasters to foot the bill to turn over to DTV, especially with few views and little interest.
But it will happen. The former FCC head, Kennard, said he thought it was more like 2020 instead of 2007 when he spoke on the issue last year.
I was lucky enough to suck it off the Sundance channel via ReplayTV. It took me weeks to watch it. Parts of the film, like the LONG DRIVE BACK INTO THE CITY are hypnotic, mesmering, trying to show us an alien intelligence of alien coldness in our own environment.
Frankly, I'd love to see the less brainy alternative. Tarkovsky's is brilliant, barely accessible, odd. The flip side would be terrifying, fast paced, etc. The book is full of terrifying moments, which I think don't get captured through the intellectualization of Tarkovsky's film, although he captures the horrible, horrible isolation and alienation.
Referers aren't always sent. What would be smart is to use PHP or something similar to validate a referrer via a token. For 99.99% of all cases, this would allow NPR to put up a disclaimer page about the content.
The ombudsman seems to confuse linking with endorsement, partly because NPR makes all its streams available. If the streams were inside a PHP wrapper, a user trying to clickthrough to a stream would first see a page that said, "Hey, this is NPR. We don't have anything to do with where you came from. Click Listen to proceed." The Listen button would use a PHP session token to authenticate to the wrapper.
From within NPR, the site would always just pass these tokens.
The tokens don't have to be secure; most users wouldn't try to bypass them. Those that did for whatever reason would clearly know they were bypassing them. Solves the whole problem. Uses technology instead of lawyers.