If your network is pushing traffic into someone else's network for them to deliver you pay for that, because that's how their network infrastructure gets paid for.
If two networks push traffic to each other in equal quantities, which is typically between "peer" consumer ISPs, because users email each other, etc., then it's cheaper if you run the link for free than if you meter and bill each other.
If you're running a content site, pushing content out to the internet to deliver, and accepting nothing back but the occasional HTTP GET, then you have to pay for the transit to get your data to consumers. All web sites pay for their bandwidth, either explicitly or as a part of a site hosting package. And that's now the infrastructure of the internet is paid for. There's no reason that Netflix would be an exception, other than their desire to get massive volumes of transit for free.
The reason that Cogent's bad behavior is relevant is that they sign peering agreements with consumer ISP's, with the condition that the actual traffic be balanced between Cogent and the consumer ISP. Then they use that "peering" arrangement to push traffic onto the ISPs network, and instead of also receiving traffic back to deliver to other ISPs, they pushed out far more than they receive, so they're far from balanced traffic between peers. A few months of that, and the consumer ISP cuts Cogent off, forcing them to pay for the transit that they're consuming, or to balance their traffic. They've been doing this for many years, and been cut off by Verizon, AT&T, Comcast, France Telecom,... pretty much every major ISP has had trouble with Cogent.
What happened here is that Cogent tried this trick with Comcast, and this time they tried to use their customer, Netflix, to force Comcast to let them get away with violating their peering agreement.
Safari uses WebKit, and the speed increase is enabled by default in WebKit for iOS and Mac OS X, so Safari should benefit from this speedup directly, presumably in the next Safari update.
I agree that Safari doesn't handle many tabs well. Though for me, neither does Chrome. Ever since they went to the "one process per tab" model (admittedly a good idea from the stability and security perspectives) both Chrome and Safari have kinda sucked with many tabs open, because each tab has much more overhead than they used to.
Peering agreements are between peers, with balanced traffic. Cogent doesn't run balanced links to the "peers", they push tons of traffic into the "peer" networks, and accept a trivial amount of traffic back (basically just requests). And Netflix' proposed "peering" with Comcast is even more one-sided.
That's transit, not peering, and they should be paying for it.
That's why Cogent keeps getting cut off: Level3, Spring, Comcast, Verizon. France Telecom,... there's a long history of Cogent making agreements and then pushing them aggressively until the other party cuts them off. And then issuing press releases blaming the other party for "violating net neutrality" or whatever. If you google for "cogent cut off for violating peering agreements" you'll get over 400,000 hits, because Congent's business model is to be cheaper than any other transit provider, and they can only do that by cheating and pushing their cost of business onto their "peers".
The problem with Cogent is that Cogent is a badly behaved bandwidth provider - they have very low prices because they don't like to pay for transit, they rely on peering arrangements which they then violate (by pushing data into the other network, when peering is supposed to be balanced traffic) and getting cut off. And many, many ISPs have cut off Cogent due to their consistently violating peering agreements, which is why Congent is such a cheap bandwidth provider - they have to charge low prices to get people to use their service that is constantly at risk of being shut off. This causes their customers to have bad service, or no service, which is why Netflix has having trouble delivering video through Cogent. The problem isn't Comcast, the problem is Netflix' use of a sleazy bandwidth provider.
But instead of switching to a better quality bandwidth provider, what Netflix was doing was running private data lines from their data centers into Comcast, bypassing Congent (which is fine), and demanding that Comcast give them the ports and transit for free (which is not fine). What Netflix should do is pay for transit from a bandwidth provider that hasn't abused all of the ISPS to the point where they get cut off. But Netflix doesn't want to pay for the bandwidth that they're consuming, so they're making threats and issuing press releases instead. Sleazy.
I think you're confused about the direction of transit. And you really shouldn't call me a liar just because you're confused about how networking is paid for. The direction of transit has nothing to do with who sends the request for data. The direction of transit is the direction of the flow of data volume, which (by a wide margin) flows from the content providers to consumers. Bandwidth providers, such as Netflix' provider (Cogent) receive data from content providers and delivers it out to consumer ISPs. Consumer ISPs, like Comcast, receive data from content providers and delivers it to consumers. The content providers pay their bandwidth providers, and consumers pay their ISPs, and that's how the internet gets paid for.
What Netflix tried to do was to bypass their bandwidth providers and run private links to Comcast, and then demanded that Comcast provide the transit for free.
Note that every web site on the internet pays for the bandwidth to deliver its content to anyone who browses the site. There's nothing magic about Netflix that they should be able to push their cost of doing business onto everyone else.
Actually, Netflix uses Cogent, a well known low-cost, unreliable bandwidth provider, not Level 3, a well-known, high-cost, high quality bandwidth provider. (As least, the routes I've seen) And Comcast didn't throttle Netflix. What actually happened was that Netflix decided that they didn't even want to pay Congent's low rates for bad bandwidth, they went to Comcast and proposed running direct lines, bypassing Congent, and they wanted Comcast to provide the ports and delivery for free.
Note that the bandwidth that Netflix was delivering through Cogent was being delivered just fine (as much as Congent is capable of delivering bandwidth). Comcast didn't interfere with Netflix at all, Netflix basically hosed their own deliveries in order to pressure Comcast to provide dedicated private links for free.
And no, this isn't the same as the caching servers that Netflix runs (and which Comcast has run for years).
No, by "free transit" I mean that Netflix didn't want to pay their ISP (Cogent) for the bandwidth, instead they wanted to bypass Congent and run data lines directly to Comcast and have Comcast deliver the data for free. Note that Comcast wasn't throttling Netflix, the "throttle" was that Netflix used a cheap, low quality ISP (Cogent) that delivered low quality bandwidth (i.e. Netflix caused the problems for their customers). Instead of paying for higher quality bandwidth, or Netflix tried to push Comcast to give them the delivery bandwidth for free, by calling it "peering".
Of course, Netflix delivering video to its customers isn't "peering".
Peering is for transit exchanges between consumer ISPs. That is, if (for example) Comcast and Verizon have customers communicating with each other in roughly balanced volumes, then Comcast and Verizon could agree to exchange the data for free instead of wasting time charging each other bills back and forth for that data. And that works because the traffic is balanced, so it gets the data delivered without wasting overhead of metering and billing each other.
But Netflix doesn't accept any traffic to deliver to consumers, they are a content provider that's pushing data (in huge quantities) onto consumer ISPs networks for them to deliver. So in their proposed "peering" relationship is entirely one-sided. And since they'd be the ones receiving all of the benefit of getting their data delivered, they pay for it. Like every other web site on the planet, Netflix has to pay for the bandwidth that they consume running their service.
You missed what was going on, then. What Netflix was trying to get for free was transit from Comcast. Specifically, Netflix wanted to bypass their ISP so they wouldn't have to pay for the transit, and instead run direct lines into Comcast's infrastructure, demanding that Comcast run the connections and transit the data for free. The transit that they were paying for was being delivered just fine the whole time. The only "fight" was that they demanded preferential treatment from Comcast. And it was particularly sleazy of Netflix' to call the transit "peering" because it clearly wasn't - peering is connecting between ISPs that have traffic between each other's customers, when balanced in both directions, so it's more efficient to deliver the traffic for free in both directions than to bill each other for the balanced traffic. But Netflix isn't an ISP, and their traffic isn't even remotely balanced - they push tons of traffic out into ISPs, and accept no traffic, so it's not balanced traffic between peers at all.
Perhaps you didn't know the details, so I'll forgive the snide comment. What Netflix tried to do with Comcast was to bypass Netflix' ISP and try to run direct lines. That'd be OK, but they tried to force Comcast to route all of Netflix content for free. That's a sleazy move by Netflix.
To elaborate a bit, reserved bandwidth is not the same as shared bandwidth.
The bandwidth that companies buy is reserved bandwidth, which is guaranteed capacity allocated to you and nobody else. That kind of bandwidth is expensive - let's say $10/Mbps (it's more in small quantities, less in large quantities, but let's make the math easy). So if you want 100 Mbps guaranteed, it'll cost you $1000/month. plus circuit fees, etc. In return for that money, you "own" the bandwidth, it's guaranteed available when you need it, and there's a Service Level Agreement with penalties if it's not there. And the cost is buy because to satisfy 100 customers, the ISP has to build 10,000 Mbps of capacity and keep it available, with redundancy.
The bandwidth that consumers buy is shared bandwidth, which is a capacity shared by everyone in the network. So if you get "up to 100 Mbps" you should expect to get 100 Mbps most of the time, but there are no guarantees. But you pay perhaps $100/month for that bandwidth (depending on your ISP, certainly less than $1,000). The way the ISP can sell shared bandwidth more cheaply is that they have a user behavioral model that tells them that if they have 100 customers that their total usage on average will peak at 1,000 Mbps, because not everyone will be home, online, and using max bandwidth. So they only have to build 1,000 Mbps of capacity, and because there are no SLAs, they don't need the redundancy that committed bandwidth would have.
So they're spending 1/10th as much to provision the "same bandwidth" for consumers as businesses, because it's not the same bandwidth. And that means that you can't expect committed bandwidth (SLA's, guaranteed capacity) if you didn't pay for it - if you pay cheap prices, you get what you pay for.
And that's not a bad thing. For most consumers, paying 1/10th as much for bandwidth is a great deal, and the bandwidth is almost always there when they want it. And if the internet connection is down for a few hours, it's an inconvenience, but they'll survive. But for businesses that rely on being reliably online (or they're out of businesses) paying 10x as much is worth it for the guarantees.
If Netflix pays for their bandwidth to their ISP to serve their content, and you pay your ISP for your bandwidth to get to the internet (including Netflix), that's the deal. If your ISP slows down Netflix' content to try to get Netflix to pay for improved performance, that's wrong. If Netflix tries to get bandwidth for free, as they were trying to do with Comcast, that's wrong, too.
It's making something like 3D printing accessible to people who aren't into CAD. I have had a 3Doodler for a few months, and it's been a huge hit with my daughters, who are very artistic but not CAD-oriented. Even though it's the same extruder and plastic, the experience of drawing with plastic is very different from 3D printing with a printer, and it appeals to different people.
I double checked their Kickstarter, and they say that it comes with a USB cable and a USB power supply, that it works with USB 3 and they're looking into a USB 2 solution. So they're not saying that the LIX will work on any USB port, they're saying that it'll work on their USB power supply, and with USB 3 (which has a much higher power budget, optionally). Given that running an extruder off of a battery is a dumb idea (it likely draws as much or more power than your laptop), running on wall power makes sense. Do you really need to do freehand ABS extrusion on an airplane?
Give that, it's pretty much the same functionality as the (awesome) 3Doodler, but in a sleeker package. It uses straight filament, has a motor and an extruder. The main difference is that it runs on 1.75mm filament instead of 3mm filament. 1.75mm filament shouldn't change the power consumption - basically you need to transfer the same heat into the plastic to melt the same volume of plastic to extrude, whether it's coming in a 1.75mm or a 3mm diameter filament.
Caveat: if the LIX extrudes a thinner line of plastic, then either it'll draw faster, or use less power. The LIX says it has an 0.6mm nozzle. I'm not sure what the 3Doodler's nozzle size is - about the same size, I think. And 3Doodler just announced a range of accessories, including smaller and larger nozzles. So it'll be interesting to compare them.
I will say that, after a fair amount of testing, running the 3Doodler on curved filament from a spool didn't work out too well - it tends to jam in the straight path through the pen. That's not so much of an issue with ABS, but PLA is pretty stiff, at least the 3mm filament that 3Doodler uses, so I'd recommend using the straight filament that the vendors sell, not filament from a spool.
The 3Doodler exists, and people were making devices like that from spare 3D printer extruders for many years before that, so there's no doubt that you can melt and extrude ABS.
The only question with the Lix is whether they can convert power into melted ABS efficiently enough to do so from a USB port. Technically USB 3 can provide 100 watts of power, which is far more than is needed to melt and extrude plastic. So if USB 3, with its power budget, is their target, it's doable.
Where it's iffy is the "any USB port", which means going back to 500 mA @ 5 volt USB 1 ports, which is only 2.5 amps. If they are extremely efficient at heating the ABS (most 3D printers' extruders radiate significant heat, since power efficiency isn't the priority) it might be possible to do it. Though it might be pretty slow. The 3Doodler requires some patience, so if the Lix is a lot slower, I think they'd get a lot of complaints. Though "it's slow on very old USB ports, faster on newer USB ports" isn't a bad story.
To the people saying that you can't extrude plastic vertically - you can. I've posted pictures at http://kickrev.blogspot.com/20... . It takes some practice, and a steady hand, but it's very doable.
Kickstarters have delivered thousands of products successfully, so they're clearly not all scams. There have been a few scams that made it onto the site, which were shut down as people dug into their claims - the "crowd" doesn't passively hand over money, they dig with impressive thoroughness when they think they're being taken advantage of, and Kickstart shuts down projects as a result of crowd investigations. Of course, Kickstarter also filters out many projects (presumably including most scams), so if you look at Kickstarter, the projects generally look plausible. Not all a good idea, of course, but that doesn't make them a scam. The result, for me, is that of the large number of Kickstarts that I've backed, only one was a scam (or massive incompetence) - the vast majority deliver, and the ones that don't are people who (as far as I can tell) honestly got in over their heads and couldn't pull it off, which is the risk that you have to accept when contributing to a startup. Kickstarter is not a store.
I can compare to the VC route. I've done two VC-backed startups (both ended in successful acquisitions, woot!), so I can make a comparison. If anything, because there's usually a lot less money in a Kickstarter than a VC-backed company, there's less incentive to scam, and greater transparency from Kickstarters than from people pitching VCs. And because Kickstarters are mainly shooting for modest goals, rather than VC's "shooting for the moon", the success rate for Kickstarts is a lot higher than VC-backed startups.
Even though they're both ways to fund things, Kickstarter and VCs are very different worlds. Kickstarter's average successful project raises $40K, and nobody gets equity. Most VCs aren't interesting in any deal that doesn't have a lot more zeroes in it, and of course they get tons of equity in return. Anywhere there's money on the table people will try scamming, but both Kickstarter and VCs have mechanisms to protect them from abuse, that work well enough that overall the systems work.
I'm talking about the difference between theory and reality. In theory, guns might make people safer. In reality, however, people with guns get shot, and shoot other people, far more often than people without guns. So while they might feel safer, the numbers don't support their feelings. You see, in real life people aren't always rational - sometimes they get angry, or feel suicidal, etc., and when those feelings hit and they have a gun, people are much more likely to die than if they don't have a gun.
I'm not trying to "punish" anyone. I'm trying to keep people from making bad decisions due to lack of information about the risks that they're putting themselves and their loved ones in by keeping a gun in their home.
It sounds like we agree about requiring people to know how to use a gun before they can buy it, and in punishing those who commit crimes with guns. As are 90% of Americans polled. Sadly, the NRA is strongly opposed to both measures, and they have enough money to buy politicians to make sure that they don't do what the voters want.
Exactly - the founders made quite clear that they were referring to The People as a whole having a right to bear arms, through their state militia. They quite clearly opposed the idea that everyone had a right to unlimited guns - they killed anti-American rebels (e.g. Whiskey Rebellion). Remember, they founded the United States, they didn't hate it!
Let's look at it another way - gun deaths in the US are 10x higher per capita than other countries. Are gun owners unwilling to accept even minimal requirements for gun safety, even at the cost of many people's lives?
No, the 2nd Amendment gives states the right to organize a citizen militia armed with guns. It does not give all individuals the unrestricted right to buy guns. Guns were regulated when the US was founded, and have always been regulated. There have always been laws about what kinds of guns you can buy, who can buy them, where you can carry them, etc. The only debate is what the regulations should be. For example, the NRA's position is that felons should be able to buy unlimited quantities of any guns they like, anonymously. Most people oppose those positions, because they think it's a good idea to keep felons from buying guns, and would like to be able to track gun sales back to dealers in order to prevent them from selling guns to felons.
As for the "if you want to restrict guns, move somewhere that people agree with you", we already live there - 90% of Americans disagree with the NRA's most extreme positions. But apparently huge piles of cash from gun manufacturers affects politicians behaviors more than the opinions of the voters.
Really? A gun that has a battery that you have to check ever year is worse than a gun that can be taken from you and used to kill you? Are you sure about that?
If he hadn't had a gun, he wouldn't have done something stupid like confront robbers face-to-face. Instead he would have called the police and waited a few minutes for them to arrest the unarmed robbers.
By having he gun, he made a foolish choice that resulted in his death.
No, you're assuming your own conclusion. Robbers rob empty buildings because it's safer and easier than doing so when the building is occupied. That has nothing to do with a gun - a home owner can call the police, or call out for help (and the neighbors call the police) which results in cars full of policemen who are not only armed, they are trained, surrounding the house. And that's a lot more threatening to a robber than one untrained civilian with a handgun. A robber can likely kill one untrained civilian and escape, since someone without training is unlikely to actually hit anything in a combat situation, but the odds of taking on many, trained police are much worse for them.
Statistically guns are much more likely to get the owner or someone else in their home killed than to save them from anything. So you might feel safer having a gun, but the odds aren't in your favor, and by carrying a gun you're (statistically) putting everyone around you at more risk. That's not to say that you're going to pull your gun out and shoot someone, but if you get angry or suicidal and you have a gun, you're more likely to shoot someone than if you don't have a gun. And those things happen more often than "hot" break-ins.
Perhaps you're an awesome gun owner. Statistically, however, guns in the home are 5x more likely to be used to kill someone in the home (suicide, murder) than to stop an intruder. So if nothing else an electronic gun lock will save lives by making the gun a bit harder to use.
Slashdot Mirror
If your network is pushing traffic into someone else's network for them to deliver you pay for that, because that's how their network infrastructure gets paid for.
If two networks push traffic to each other in equal quantities, which is typically between "peer" consumer ISPs, because users email each other, etc., then it's cheaper if you run the link for free than if you meter and bill each other.
If you're running a content site, pushing content out to the internet to deliver, and accepting nothing back but the occasional HTTP GET, then you have to pay for the transit to get your data to consumers. All web sites pay for their bandwidth, either explicitly or as a part of a site hosting package. And that's now the infrastructure of the internet is paid for. There's no reason that Netflix would be an exception, other than their desire to get massive volumes of transit for free.
The reason that Cogent's bad behavior is relevant is that they sign peering agreements with consumer ISP's, with the condition that the actual traffic be balanced between Cogent and the consumer ISP. Then they use that "peering" arrangement to push traffic onto the ISPs network, and instead of also receiving traffic back to deliver to other ISPs, they pushed out far more than they receive, so they're far from balanced traffic between peers. A few months of that, and the consumer ISP cuts Cogent off, forcing them to pay for the transit that they're consuming, or to balance their traffic. They've been doing this for many years, and been cut off by Verizon, AT&T, Comcast, France Telecom, ... pretty much every major ISP has had trouble with Cogent.
What happened here is that Cogent tried this trick with Comcast, and this time they tried to use their customer, Netflix, to force Comcast to let them get away with violating their peering agreement.
Safari uses WebKit, and the speed increase is enabled by default in WebKit for iOS and Mac OS X, so Safari should benefit from this speedup directly, presumably in the next Safari update.
I agree that Safari doesn't handle many tabs well. Though for me, neither does Chrome. Ever since they went to the "one process per tab" model (admittedly a good idea from the stability and security perspectives) both Chrome and Safari have kinda sucked with many tabs open, because each tab has much more overhead than they used to.
"How did Cogent violate peering agreements?"
Peering agreements are between peers, with balanced traffic. Cogent doesn't run balanced links to the "peers", they push tons of traffic into the "peer" networks, and accept a trivial amount of traffic back (basically just requests). And Netflix' proposed "peering" with Comcast is even more one-sided.
That's transit, not peering, and they should be paying for it.
That's why Cogent keeps getting cut off: Level3, Spring, Comcast, Verizon. France Telecom, ... there's a long history of Cogent making agreements and then pushing them aggressively until the other party cuts them off. And then issuing press releases blaming the other party for "violating net neutrality" or whatever. If you google for "cogent cut off for violating peering agreements" you'll get over 400,000 hits, because Congent's business model is to be cheaper than any other transit provider, and they can only do that by cheating and pushing their cost of business onto their "peers".
The problem with Cogent is that Cogent is a badly behaved bandwidth provider - they have very low prices because they don't like to pay for transit, they rely on peering arrangements which they then violate (by pushing data into the other network, when peering is supposed to be balanced traffic) and getting cut off. And many, many ISPs have cut off Cogent due to their consistently violating peering agreements, which is why Congent is such a cheap bandwidth provider - they have to charge low prices to get people to use their service that is constantly at risk of being shut off. This causes their customers to have bad service, or no service, which is why Netflix has having trouble delivering video through Cogent. The problem isn't Comcast, the problem is Netflix' use of a sleazy bandwidth provider.
But instead of switching to a better quality bandwidth provider, what Netflix was doing was running private data lines from their data centers into Comcast, bypassing Congent (which is fine), and demanding that Comcast give them the ports and transit for free (which is not fine). What Netflix should do is pay for transit from a bandwidth provider that hasn't abused all of the ISPS to the point where they get cut off. But Netflix doesn't want to pay for the bandwidth that they're consuming, so they're making threats and issuing press releases instead. Sleazy.
I think you're confused about the direction of transit. And you really shouldn't call me a liar just because you're confused about how networking is paid for. The direction of transit has nothing to do with who sends the request for data. The direction of transit is the direction of the flow of data volume, which (by a wide margin) flows from the content providers to consumers. Bandwidth providers, such as Netflix' provider (Cogent) receive data from content providers and delivers it out to consumer ISPs. Consumer ISPs, like Comcast, receive data from content providers and delivers it to consumers. The content providers pay their bandwidth providers, and consumers pay their ISPs, and that's how the internet gets paid for.
What Netflix tried to do was to bypass their bandwidth providers and run private links to Comcast, and then demanded that Comcast provide the transit for free.
Note that every web site on the internet pays for the bandwidth to deliver its content to anyone who browses the site. There's nothing magic about Netflix that they should be able to push their cost of doing business onto everyone else.
Actually, Netflix uses Cogent, a well known low-cost, unreliable bandwidth provider, not Level 3, a well-known, high-cost, high quality bandwidth provider. (As least, the routes I've seen) And Comcast didn't throttle Netflix. What actually happened was that Netflix decided that they didn't even want to pay Congent's low rates for bad bandwidth, they went to Comcast and proposed running direct lines, bypassing Congent, and they wanted Comcast to provide the ports and delivery for free.
Note that the bandwidth that Netflix was delivering through Cogent was being delivered just fine (as much as Congent is capable of delivering bandwidth). Comcast didn't interfere with Netflix at all, Netflix basically hosed their own deliveries in order to pressure Comcast to provide dedicated private links for free.
And no, this isn't the same as the caching servers that Netflix runs (and which Comcast has run for years).
No, by "free transit" I mean that Netflix didn't want to pay their ISP (Cogent) for the bandwidth, instead they wanted to bypass Congent and run data lines directly to Comcast and have Comcast deliver the data for free. Note that Comcast wasn't throttling Netflix, the "throttle" was that Netflix used a cheap, low quality ISP (Cogent) that delivered low quality bandwidth (i.e. Netflix caused the problems for their customers). Instead of paying for higher quality bandwidth, or Netflix tried to push Comcast to give them the delivery bandwidth for free, by calling it "peering".
Of course, Netflix delivering video to its customers isn't "peering".
Peering is for transit exchanges between consumer ISPs. That is, if (for example) Comcast and Verizon have customers communicating with each other in roughly balanced volumes, then Comcast and Verizon could agree to exchange the data for free instead of wasting time charging each other bills back and forth for that data. And that works because the traffic is balanced, so it gets the data delivered without wasting overhead of metering and billing each other.
But Netflix doesn't accept any traffic to deliver to consumers, they are a content provider that's pushing data (in huge quantities) onto consumer ISPs networks for them to deliver. So in their proposed "peering" relationship is entirely one-sided. And since they'd be the ones receiving all of the benefit of getting their data delivered, they pay for it. Like every other web site on the planet, Netflix has to pay for the bandwidth that they consume running their service.
You missed what was going on, then. What Netflix was trying to get for free was transit from Comcast. Specifically, Netflix wanted to bypass their ISP so they wouldn't have to pay for the transit, and instead run direct lines into Comcast's infrastructure, demanding that Comcast run the connections and transit the data for free. The transit that they were paying for was being delivered just fine the whole time. The only "fight" was that they demanded preferential treatment from Comcast. And it was particularly sleazy of Netflix' to call the transit "peering" because it clearly wasn't - peering is connecting between ISPs that have traffic between each other's customers, when balanced in both directions, so it's more efficient to deliver the traffic for free in both directions than to bill each other for the balanced traffic. But Netflix isn't an ISP, and their traffic isn't even remotely balanced - they push tons of traffic out into ISPs, and accept no traffic, so it's not balanced traffic between peers at all.
Perhaps you didn't know the details, so I'll forgive the snide comment. What Netflix tried to do with Comcast was to bypass Netflix' ISP and try to run direct lines. That'd be OK, but they tried to force Comcast to route all of Netflix content for free. That's a sleazy move by Netflix.
This is correct.
To elaborate a bit, reserved bandwidth is not the same as shared bandwidth.
The bandwidth that companies buy is reserved bandwidth, which is guaranteed capacity allocated to you and nobody else. That kind of bandwidth is expensive - let's say $10/Mbps (it's more in small quantities, less in large quantities, but let's make the math easy). So if you want 100 Mbps guaranteed, it'll cost you $1000/month. plus circuit fees, etc. In return for that money, you "own" the bandwidth, it's guaranteed available when you need it, and there's a Service Level Agreement with penalties if it's not there. And the cost is buy because to satisfy 100 customers, the ISP has to build 10,000 Mbps of capacity and keep it available, with redundancy.
The bandwidth that consumers buy is shared bandwidth, which is a capacity shared by everyone in the network. So if you get "up to 100 Mbps" you should expect to get 100 Mbps most of the time, but there are no guarantees. But you pay perhaps $100/month for that bandwidth (depending on your ISP, certainly less than $1,000). The way the ISP can sell shared bandwidth more cheaply is that they have a user behavioral model that tells them that if they have 100 customers that their total usage on average will peak at 1,000 Mbps, because not everyone will be home, online, and using max bandwidth. So they only have to build 1,000 Mbps of capacity, and because there are no SLAs, they don't need the redundancy that committed bandwidth would have.
So they're spending 1/10th as much to provision the "same bandwidth" for consumers as businesses, because it's not the same bandwidth. And that means that you can't expect committed bandwidth (SLA's, guaranteed capacity) if you didn't pay for it - if you pay cheap prices, you get what you pay for.
And that's not a bad thing. For most consumers, paying 1/10th as much for bandwidth is a great deal, and the bandwidth is almost always there when they want it. And if the internet connection is down for a few hours, it's an inconvenience, but they'll survive. But for businesses that rely on being reliably online (or they're out of businesses) paying 10x as much is worth it for the guarantees.
There are more variations.
If Netflix pays for their bandwidth to their ISP to serve their content, and you pay your ISP for your bandwidth to get to the internet (including Netflix), that's the deal. If your ISP slows down Netflix' content to try to get Netflix to pay for improved performance, that's wrong. If Netflix tries to get bandwidth for free, as they were trying to do with Comcast, that's wrong, too.
It's making something like 3D printing accessible to people who aren't into CAD. I have had a 3Doodler for a few months, and it's been a huge hit with my daughters, who are very artistic but not CAD-oriented. Even though it's the same extruder and plastic, the experience of drawing with plastic is very different from 3D printing with a printer, and it appeals to different people.
For some examples, see: http://kickrev.blogspot.com/20... .
It has buttons labeled "speed controls" so I'm pretty sure that the filament is motor driven. Manually driven filament would be way to uneven.
They also only say that it works on USB 3 or wall power, and they include a USB wall plug for anyone with USB 2.
Not sure how they got a motor small enough to fit into a pen and feed filament...
I double checked their Kickstarter, and they say that it comes with a USB cable and a USB power supply, that it works with USB 3 and they're looking into a USB 2 solution. So they're not saying that the LIX will work on any USB port, they're saying that it'll work on their USB power supply, and with USB 3 (which has a much higher power budget, optionally). Given that running an extruder off of a battery is a dumb idea (it likely draws as much or more power than your laptop), running on wall power makes sense. Do you really need to do freehand ABS extrusion on an airplane?
Give that, it's pretty much the same functionality as the (awesome) 3Doodler, but in a sleeker package. It uses straight filament, has a motor and an extruder. The main difference is that it runs on 1.75mm filament instead of 3mm filament. 1.75mm filament shouldn't change the power consumption - basically you need to transfer the same heat into the plastic to melt the same volume of plastic to extrude, whether it's coming in a 1.75mm or a 3mm diameter filament.
Caveat: if the LIX extrudes a thinner line of plastic, then either it'll draw faster, or use less power. The LIX says it has an 0.6mm nozzle. I'm not sure what the 3Doodler's nozzle size is - about the same size, I think. And 3Doodler just announced a range of accessories, including smaller and larger nozzles. So it'll be interesting to compare them.
I will say that, after a fair amount of testing, running the 3Doodler on curved filament from a spool didn't work out too well - it tends to jam in the straight path through the pen. That's not so much of an issue with ABS, but PLA is pretty stiff, at least the 3mm filament that 3Doodler uses, so I'd recommend using the straight filament that the vendors sell, not filament from a spool.
Correction, 500 mA @ 5 volt USB 1 ports, which is only 2.5 *watts*.
The 3Doodler exists, and people were making devices like that from spare 3D printer extruders for many years before that, so there's no doubt that you can melt and extrude ABS.
The only question with the Lix is whether they can convert power into melted ABS efficiently enough to do so from a USB port. Technically USB 3 can provide 100 watts of power, which is far more than is needed to melt and extrude plastic. So if USB 3, with its power budget, is their target, it's doable.
Where it's iffy is the "any USB port", which means going back to 500 mA @ 5 volt USB 1 ports, which is only 2.5 amps. If they are extremely efficient at heating the ABS (most 3D printers' extruders radiate significant heat, since power efficiency isn't the priority) it might be possible to do it. Though it might be pretty slow. The 3Doodler requires some patience, so if the Lix is a lot slower, I think they'd get a lot of complaints. Though "it's slow on very old USB ports, faster on newer USB ports" isn't a bad story.
To the people saying that you can't extrude plastic vertically - you can. I've posted pictures at http://kickrev.blogspot.com/20... . It takes some practice, and a steady hand, but it's very doable.
Kickstarters have delivered thousands of products successfully, so they're clearly not all scams. There have been a few scams that made it onto the site, which were shut down as people dug into their claims - the "crowd" doesn't passively hand over money, they dig with impressive thoroughness when they think they're being taken advantage of, and Kickstart shuts down projects as a result of crowd investigations. Of course, Kickstarter also filters out many projects (presumably including most scams), so if you look at Kickstarter, the projects generally look plausible. Not all a good idea, of course, but that doesn't make them a scam. The result, for me, is that of the large number of Kickstarts that I've backed, only one was a scam (or massive incompetence) - the vast majority deliver, and the ones that don't are people who (as far as I can tell) honestly got in over their heads and couldn't pull it off, which is the risk that you have to accept when contributing to a startup. Kickstarter is not a store.
I can compare to the VC route. I've done two VC-backed startups (both ended in successful acquisitions, woot!), so I can make a comparison. If anything, because there's usually a lot less money in a Kickstarter than a VC-backed company, there's less incentive to scam, and greater transparency from Kickstarters than from people pitching VCs. And because Kickstarters are mainly shooting for modest goals, rather than VC's "shooting for the moon", the success rate for Kickstarts is a lot higher than VC-backed startups.
Even though they're both ways to fund things, Kickstarter and VCs are very different worlds. Kickstarter's average successful project raises $40K, and nobody gets equity. Most VCs aren't interesting in any deal that doesn't have a lot more zeroes in it, and of course they get tons of equity in return. Anywhere there's money on the table people will try scamming, but both Kickstarter and VCs have mechanisms to protect them from abuse, that work well enough that overall the systems work.
I'm talking about the difference between theory and reality. In theory, guns might make people safer. In reality, however, people with guns get shot, and shoot other people, far more often than people without guns. So while they might feel safer, the numbers don't support their feelings. You see, in real life people aren't always rational - sometimes they get angry, or feel suicidal, etc., and when those feelings hit and they have a gun, people are much more likely to die than if they don't have a gun.
I'm not trying to "punish" anyone. I'm trying to keep people from making bad decisions due to lack of information about the risks that they're putting themselves and their loved ones in by keeping a gun in their home.
It sounds like we agree about requiring people to know how to use a gun before they can buy it, and in punishing those who commit crimes with guns. As are 90% of Americans polled. Sadly, the NRA is strongly opposed to both measures, and they have enough money to buy politicians to make sure that they don't do what the voters want.
Exactly - the founders made quite clear that they were referring to The People as a whole having a right to bear arms, through their state militia. They quite clearly opposed the idea that everyone had a right to unlimited guns - they killed anti-American rebels (e.g. Whiskey Rebellion). Remember, they founded the United States, they didn't hate it!
Let's look at it another way - gun deaths in the US are 10x higher per capita than other countries. Are gun owners unwilling to accept even minimal requirements for gun safety, even at the cost of many people's lives?
No, the 2nd Amendment gives states the right to organize a citizen militia armed with guns. It does not give all individuals the unrestricted right to buy guns. Guns were regulated when the US was founded, and have always been regulated. There have always been laws about what kinds of guns you can buy, who can buy them, where you can carry them, etc. The only debate is what the regulations should be. For example, the NRA's position is that felons should be able to buy unlimited quantities of any guns they like, anonymously. Most people oppose those positions, because they think it's a good idea to keep felons from buying guns, and would like to be able to track gun sales back to dealers in order to prevent them from selling guns to felons.
As for the "if you want to restrict guns, move somewhere that people agree with you", we already live there - 90% of Americans disagree with the NRA's most extreme positions. But apparently huge piles of cash from gun manufacturers affects politicians behaviors more than the opinions of the voters.
Really? A gun that has a battery that you have to check ever year is worse than a gun that can be taken from you and used to kill you? Are you sure about that?
If he hadn't had a gun, he wouldn't have done something stupid like confront robbers face-to-face. Instead he would have called the police and waited a few minutes for them to arrest the unarmed robbers.
By having he gun, he made a foolish choice that resulted in his death.
No, you're assuming your own conclusion. Robbers rob empty buildings because it's safer and easier than doing so when the building is occupied. That has nothing to do with a gun - a home owner can call the police, or call out for help (and the neighbors call the police) which results in cars full of policemen who are not only armed, they are trained, surrounding the house. And that's a lot more threatening to a robber than one untrained civilian with a handgun. A robber can likely kill one untrained civilian and escape, since someone without training is unlikely to actually hit anything in a combat situation, but the odds of taking on many, trained police are much worse for them.
Statistically guns are much more likely to get the owner or someone else in their home killed than to save them from anything. So you might feel safer having a gun, but the odds aren't in your favor, and by carrying a gun you're (statistically) putting everyone around you at more risk. That's not to say that you're going to pull your gun out and shoot someone, but if you get angry or suicidal and you have a gun, you're more likely to shoot someone than if you don't have a gun. And those things happen more often than "hot" break-ins.
Perhaps you're an awesome gun owner. Statistically, however, guns in the home are 5x more likely to be used to kill someone in the home (suicide, murder) than to stop an intruder. So if nothing else an electronic gun lock will save lives by making the gun a bit harder to use.