From Wikipedia: "Many advocates for net neutrality represent it as the application of common carrier rules to Internet Service Providers (ISPs) and broadband telecommunications carriers, thereby requiring ISPs to manage all Internet traffic on equal terms."
You are correct that net neutrality is not about government control. It's about ISP conrol. When I said that other countries realize that the net should remain neutral, I was really refering to the fact that the EU has passed (or is about to pass) laws to keep the net 'neutral' in Europe. You are right that there are other countries that censor content, that's a little different than what most people refer to as 'net neutrality', but it is a similar (and probably worse) thing. I probably should have said specifically the EU and I believe Japan and some other asian nations (except for China/North Korea) are for net neutrality as well.
Actually, I don't think this would effect you the way you think it does. It mainly has to do with the last mile service provider. In other words, the ISPs would like to filter their content so that whoever pays them off will make it go to the end users faster. So, unless you have an American ISP, this will not effect you. Many of the bigger American websites have servers in other cournties as well, so when you go to American websites you could very well be redirecting to a closer server anyways. Google definitly does this.
Because if the allied forces had not defeated the axis powers in WWII, Germany would have gained control over Europe. Germany had actually invaded France and taken it over by the time the US entered the war. Typically, when people refer to what would have happened if the allies had lost the war, they say that "We'd be speaking German". This is really just a figure of speach though. Russia was on the side of the alies in WWII, so they won.
This actually has nothing to do with net neutrality. ICANN is only responsible for internet names. Net neutrality is a matter of US law (other countries realize that the net should remain neutral so it's not an issue elsewhere).
I hope you get the loan free while you're in school. If that's the case, I'd totally do what you're doing. It's free money. Even if you invest in money markets you're going to make a small profit. I would advise using a low cost money market like Vanguard Prime Money market: Vanguard.com. If you're willing to tollerate risk, you could invest in something like the S&P 500. I know many people would be freaked out by using borrowed money to invest in the stock market, but if you are dollar cost averaging, your risk goes down. Since I assume they give you the money in payments as opposed to lump sums, that means you'll be dollar cost averaging your purchases. Keep in mind, if you do this, your account could go down. It could possibly go down a lot. If you started in 2000, you'd probably regret doing this. But, if you plan to contribute to a 401k after you graduate, I really see this just as some prepayments to the 401k. After you graduate, pay down your student debt in monthly installments coming out of this S&P fund. If you do this over a 10 year period, you'd be dollar cost averaging out of your S&P 500 fund as well. This would be a pretty good way to do this. The thing to be careful about is if you discontinue this program due to losing money. If you are they type of person that cannot stand to lose money, just go with the money market fund because chances are you'll sell at the bottom.
You mean when you don't devote all the country's resources to war, you can actually spend money on developing infrastructure at home and abroad that improves the lives of citizens?!?? AMAAAAAZING!!
Yeah, you're right. Maybe the US should have done this during WWII? Where would France be if that had happened?
They have a strong belief that it [A tiered system] is needed in order to make sure that ISPs have the tools and funds to expand their already overtaxed networks.
If the ISPs' networks are seriously overtaxes, why don't they raise their prices? There is seriously no need to charge both consumers and suppliers of content. The suppliers are already charged to get their content on the net by ISPs anyways. They have two opportunties to raise their prices, why do they want this third charge? It can only be because they want to control what content is available (e.g. block out people that are compeeting with them in terms of voip and video). Anyways, I'm not sure legislation is the solution. I guess what we need to do is read the service agreements we have with broadband providers. I certainly would be more likely to subscribe to a broadband provider that gives me unlimited access to files as opposed to one that decides what content to let me see at what speed.
First of all the Sun thin clients are the price they are because Sun makes money on the server you have to buy to make it all work. So the actual price is significantly higher than $300/seat.
The Sun client works with x86 and sparc. So, you could load the server software on the $300 pcs you mentioned.
you've discussed a theoretical scenario that is supposed to increase the performance of the system should I be able to make a deal with all those neighbors I haven't met, but the price isn't any lower per seat.
It's not theoretical. Configurations like this have been tested. 30 pcs is not a whole lot. It's a lot lower cost than the equivelent power system. Like I said in my previous post for an 8 cpu system, you'd need a lot more money. Probably at least $2000 if not more. We're talking performance per $ here. Yes, you can probably buy a lowend pc for marginally less than the system I've proposed. That's because no one is mass producing the stuff needed to strip out the fat of the cost of the system I've proposed. If this sort of thing was mass produced the cost would clearly be reduced. This is how prototypes work. They are more expensive than the actual product. Don't take it so literally that you would actually go knock on your neighbors door and ask them if they want to buy a pc with you. This is where a company like Google comes in. They could share the connection among all the users automatically.
The bottom line is that there are technical, social, legal, and evironment issues involved with what you propose. I'm going to wait until at least some of these issues are resolved before betting on such an approach.
I don't see any social/legal/environmental issues? It's no different from getting a web hosting account today actually. Techincal issues, yes. I said that from the begining. You need a reliable amount of bandwidth and low latency. Also, someone needs to setup a system for sharing the systems that is close to the user globally. This is an enormous engineering problem to say the least, but it's something that's doable. Given that the product that I'm describing doesn't exist yet, I'd say it's a good idea to wait.
Dude...this isn't that complicated. The point of network computing is that by sharing computing resources you need less of them and get better performance. This is because of the massive amount of wasted CPU cycles that we're all aware of. There are thin clients available today for under $300 if you want a current exact example: http://www.sun.com/sunray/sunray1/. Since almost no one buys thin clients, the price is relatively high. Typically mass production would bring the cost down.
So lets look at it this way. Say you and your 100 closest neighbors all decided to get these $300 thin clients from Sun and you all pooled your money together to buy a grid of 30 lowend pcs for $300 (the price you mentioned) each that the thin clients would connect to that all run as a grid. The total amount you spent as a group would be $39,000. This is $390 per person total, lets just make it $400 to make it a round number. It could be marginally higher due to ethernet cables etc, but this works really well if you live in an apartment complex or a condo. So, for $400 (just a little bit more expensive than the cheapest of the cheap at dell), what do you get: You get a maxiumum of 30 cpus that are shared by 100 people. Since my previous estimate was that the average user uses no more than 1 cpu hour per day, you get access to this grid. Now, lets assume you don't want to let users use more than 25% of the total computing resources for their usage bursts. That means during your peak usage (opening programs, etc) you would have access to up to 7.5 cpus at the same time. So, in this example, for about $400 (slightly over the very cheapest box available) you get a 7.5 cpu system. I don't believe you can even buy a quad cpu system for less than a couple thousand dollars (certainly not less than $1000).
Nobody is going to do any real work on a cell phone.
It doesn't have to be a cell phone. I said, "It could be on the order of the price of a cell phone". This could be a desktop or better yet, it could something the size of a cell phone with blue tooth to connect your keyboard to.
The cost of the display isn't going to magically drop just because the device is connected to the network.
Of course it won't. In fact to me the display is the most important part of the computer. But displays are another story. We were talking about computers meaning the actual box (no monitor). So, for a comparison, many people buy a box for $500 - $1000 (not including the price of the monitor). Cellphones cost about $100 - $300. This is the savings I'm talking about. Yes, you will still need to buy a monitor and nice speakers and a printer. None of that will change just because most of your computing moves onto the net. Now, imagine this $100 - $300 device (aka box) connecting into this computing grid to do all your computation. This $100 - $300 dollar device will replace your $500 - $1000 device and give the appearance of being thousands of times faster due to the fact that whenever you do any computation, you can instantly provision thousands of CPUs just to your task. This is the idea.
Many companies already lease their PC's so there's nothing new there either.
Sure they do, but that means that they have a physical box sent out to their office where some system admin assembles it and maintains it. Companies spend a ton of money on administration and they will be the first and biggest winners of grid computing. Imagine a corporation that is expanding. Instead of buying new boxes for each new employee and replacing as they become out of date being able to tap into a grid and only use the amount of CPU time they really need. All the corporations would need is the basic $100 - $300 device I mentioned. This device could be used for a administrative assistant to check email, or for an engineer to run cad programs. They'd never need to upgrade the $100 - $300 device and they'd just get a bigger bill from the utility computing company that they are subscribing to if the engineer needs to run more simulations.
Sure I did, and even if you take the most conservative estimates for these things we're wasting a ton of cpu time. Again, look at how many wasted CPU cycles you have. Even my high estimate is that the average user uses 1 hour of cpu time per day. This means you're only using 1/24th the power you have in front of you and yes, I'm sure there are a few people who use more than 1 hour cpu time, but I said average. My job involves using the computer almost all day and I rarely use 100% CPU. It's mostly when I open a program for 1 - 2 seconds I'm using the CPU at 100%. Right now, as I'm typing, I'm using less than 1% of the CPU.
what is the net effect? You're still going to end up buying a hardware device that is either going to be neutered PC that will be $50 cheaper at best, or a specialized thin-client machine that will probably cost more than a PC.
Well, actually, there would be a number of advantages to this setup in terms of both price and performance:
1.) You could buy a client system for super cheap because basically all it needs to do is display data (and play sound) that it gets from the computing grid it's connected to. All other processing could be done on the grid (as soon as the latency and bandwidth issues I mentioned in my previous post are addressed). It could be on the order of the price of a cell phone (if not in 2006, definitly by 2010 given the current rate of progress in cell phone cpu power). Like many cell phone plans, there would be free models with particular service contracts. So, it would be cheap.
2.) This device which costs about the same as a cell phone would give you access to this entire grid of computers that is shared by millions of users. During your spike usage (see my previous post for details on what this is), you could get access to say 1000 CPUs that would run what you need run in 1000th of the time it typically takes to run the code if you had a single system. Even though you're still only using the same CPU time you would on your individual PC, it would seem as if you are using a computer 1000 times faster.
So, no it wouldn't be a neutered PC, it would be something way more powerful than any PC you can currently buy and it would cost about as much as cell phone. Remember cell phones get cheaper every year as well.
All for a device that turns into a lump of coal if your Internet connection fails. No thanks.
First of all, for most people their computer is essentially a lump of coal when the internet connection fails. I really can think of very few activities that I can do without connecting to the net. Secondly, as I said in my first post this requires two things:
1.) A reliable 5 mbps+ connection (emphasise reliable)
2.) Network latency no greater than 10 ms (about the minimum that the human eye can detect).
Yes, we don't have either of these things (for the most part) now. But we will. There are two technologies that will solve this: Wireless mesh networks and wimax connections. I don't have time to explain why this will work, but basically what it comes down to is: picture all the computers within a mile radius being potential wireless routers for you to connect to the net. You'd be simultaneously connected to hundreds of them and if any of them go down, you'd be connected to hundreds of other net connections at the same time. That's the idea. Google wimax and then google wireless mesh networks for more info.
So the guys at Slate thinks that the combined computing power of Google's umpteen million users is less than the power of their server farm? Unlikely, even for Google's impressive data centers.
But, what you forgot is that you're not using 99% of your computer's resources. What I mean is: look down at your performance meter. Probably what you'll see is that you're using around 1% of your cpu when you are actively typing at your computer. Sure occassionally this spikes up to 100% for a few seconds when you open an application, but for the most part your computer usage is very low. Ok, yeah yeah maybe it's higher for people who play graphic intensive games, etc but that's not the average user. Also, think about how often you actually use your computer. Most people don't sit in front of their computer 24/7. Even a super geek probably only sits in front of a computer for around 10 - 15 hours. So what's the total CPU time you actually use in a given day? I'd bet the average user who checks email and surfs the web and plays the occasional game uses well under 1 CPU hour per day. In fact, it might be as little as 10 minutes of CPU time per day for a lot of people (think your parents). So, when you say that Google's CPU power doesn't come close to the CPU power sitting in front of it's user base, you are right, but since their user base probably doesn't even use 1% of their CPU power, there's a compelling reason to switch more and more of the CPU cycles to the network because Google (or insert any other network computing company here) can get a lot closer to their maximum capacity by sharing their computing resources among it's huge user base.
Here's another way to think about it: What if you and 10 of your friends could pool your computing resources to make one super-computer? Maybe you could buy only 5 computers and save half the cost that you intended to spend. When you start your web browser, you could instantly get all 5 computing resources dedicated to your task of opening the browser. Then, after it's open your using about 1% of the CPU. Since you don't need more than 1% of the CPU to scroll through a web page and type, you don't notice a difference but when your usage spikes your browser INSTANTLY opens. In effect you have made it seem as if you are using the system that is 5 times faster than what you paid (at a 50% discount) for because you're sharing resources. Now, when your friend opens up the browser while you're typing he gets 99% of the total resources to open the browser. You don't even notice this though since you're typing and everything appears as normal. This is the power of using shared resources. Of course with only 10 users you could run into a problem if all 10 of you are on the computer at the same time ripping cds, talking over skype and watching streaming video all at once, but that's why it makes more sense for a company like Google to do this. With Millions of users the average CPU usage will not fluctuate much. Very quickly they'll figure out what the peak usage is. There of course will be peaks and troughs like there probably will be more users on at 2 pm as opposed to 4 am, but the sharing is still extremely efficient.
So, dollar for dollar this is why shared network computing makes sense and I have almost no doubt that it will be the norm eventually. To a large degree it depends on bandwidth usage and latency. When latency is below 10 ms round trip (the smallest percivable delay to the human eye) and you have more than 5 mBit/s bandwidth consistantly (dvd quality video), it will make no sense to have a local computer around (except possibly for privacy concerns). The privacy concerns can be solved be using encryption and locally storing your private keys though so even that is not a problem that cannot be resolved.
The best way for Windows users to compute untroubled (or less troubled) by malware is to switch to Mac OS X
Or in a more general sense: the best way to be safer from viruses is to use a platform that is not the mainstream one. Mac OS X is one example of something that could be used. Also, Linux, Free BSD, Solaris and various other platforms would work.
How pathetic are the teachers you know that they would seriously believe that they are seriously threatened by something like this?
The icon showed a gun pointing to a head, a bullet leaving the gun, and blood splattering from the head. It included the words "Kill Mr. VanderMolen,". You don't consider this a threat?
This is completely disrepectful. The kid needs to learn respect for his teachers. I'm not saying he needs to go to jail or anything, but there should be consequences for his actions. This is part of school and growing up in general. I would not expect a teacher to overlook this complete lack of respect. Any teacher that reacted the way you mentioned should not be allowed to teach.
Especially when the 'death threat' was not even communicated to him!
Obviously, he found out about it. So, someone communicated it to him.
Correct, death threats are not free speach. The classic example is that freedom of speach doesn't allow you to scream fire in a movie theatre.
First Amendment:
Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.
The amendment even mentions the right "peaceably to assemble". Even though that is probably a reference to protests, the spirit of the law is clearly to allow do things as long as the are peacful. Also, what about the teacher he threatened? He has the right to life, liberty and the pursuit of happiness as well. Could a death threat disturb this right? Of course it could. If I was a teacher I would not be happy about it at all and expect the school to do what they have done.
Google plans to charge merchants a 2.2% commission on a sale, plus 30 cents per transaction using its payment service, according to people briefed on Google's pricing. That is higher than Pay-Pal's lowest published rate of a 1.9% commission plus 30 cents per transaction.
I'm surprised to see that they're charging a higher rate than Paypal.
A problem with your idea: the patent troll would contract with a real company to develop their "technology" (ie. patented ideas).
What would be wrong with that? If a company comes up with an idea and then contracts with another company to develop their product, we're ok. What we want to stop is companies that just come up with ideas and never pursue them and then sue anyone else who actually does the work of developing that technology.
Good luck proving that they aren't doing anything with their patents.
In the case of the company that this article discusses, I believe it would be quite easy to prove that they aren't pursuing the majority of their patents because their whole model is just to come up with ideas and patent them. In the case of a big company like IBM or something, yeah you have a point, but I think the biggest offenders of the whole patent thing has been companies that have no real business model other than Suing people (i.e. SCO). Obviously, this solution wouldn't fix problems like Amazon's one click patent or the hyperlink patent, but hey, it's a start and it's easy to implement.
Require companies to show that they have/or had the intent to develop the technologies they patented.
These patent trolls would have a hard time showing that they actually intended to develop these technologies and thus their patents would be useless. On the otherhand, startups (and other inventors) would be able to use the system as it was intended to be used by disclosing their ideas in a protected way.
Agreed. Linus never intended to control the distributions. While the distros like Red Hat and Suze are important, he's always cared about the Kernel. I don't think this has changed much. He's still clearly in control of the Kernel which is an important role. People who think the Kernel doesn't matter don't understand what the Kernel does.
Yeah, I've seen the predictions about accelerating temperature changes and like you I am somewhat skeptical. I guess my point would be that none of these predictions take into consideration the fact that nanotechnology will enable the manufacture of ultra cheap solar panels that are cost effective within 20 years. Even if we use as much oil and coal as possible we will run out eventually. What's more likely is that no one will even consider using fossil fuels since the solar alternative becomes much more attractive.
The internet distribution cost Could be very low. with suitable multicast servers if everyone watched at a predefined time etc. and network maintainence fees don't go away they just shift from the tv bill to the internet bill.
I'm talking about Google's costs (not total costs). Since YOU already pay your internet bill, it doesn't need to be factored into Google's cost. This is why I think their will be a lot more content available.
Anyways, I can't find a single point in your entire comment that i found to be in any way correct.
Turst me, it's all correct. You must have misinterperted it if you think otherwise. You probably just disagree with my point about it being cheaper. But, it IS cheaper for _Google_ to provide video than it is for Cable companies because Google doesn't have to provide internet service for you because you already pay for it. It's not one of their costs. They only pay for their internet banwidth and servers. This is a lot easier than maintaining a network, sending out service men to fix problems, etc. If you disagree with another point, be specific and I'll gladly tell you why you're wrong.
Slashdot Mirror
From Wikipedia: "Many advocates for net neutrality represent it as the application of common carrier rules to Internet Service Providers (ISPs) and broadband telecommunications carriers, thereby requiring ISPs to manage all Internet traffic on equal terms."
You are correct that net neutrality is not about government control. It's about ISP conrol. When I said that other countries realize that the net should remain neutral, I was really refering to the fact that the EU has passed (or is about to pass) laws to keep the net 'neutral' in Europe. You are right that there are other countries that censor content, that's a little different than what most people refer to as 'net neutrality', but it is a similar (and probably worse) thing. I probably should have said specifically the EU and I believe Japan and some other asian nations (except for China/North Korea) are for net neutrality as well.
Actually, I don't think this would effect you the way you think it does. It mainly has to do with the last mile service provider. In other words, the ISPs would like to filter their content so that whoever pays them off will make it go to the end users faster. So, unless you have an American ISP, this will not effect you. Many of the bigger American websites have servers in other cournties as well, so when you go to American websites you could very well be redirecting to a closer server anyways. Google definitly does this.
Because if the allied forces had not defeated the axis powers in WWII, Germany would have gained control over Europe. Germany had actually invaded France and taken it over by the time the US entered the war. Typically, when people refer to what would have happened if the allies had lost the war, they say that "We'd be speaking German". This is really just a figure of speach though. Russia was on the side of the alies in WWII, so they won.
This actually has nothing to do with net neutrality. ICANN is only responsible for internet names. Net neutrality is a matter of US law (other countries realize that the net should remain neutral so it's not an issue elsewhere).
I hope you get the loan free while you're in school. If that's the case, I'd totally do what you're doing. It's free money. Even if you invest in money markets you're going to make a small profit. I would advise using a low cost money market like Vanguard Prime Money market: Vanguard.com. If you're willing to tollerate risk, you could invest in something like the S&P 500. I know many people would be freaked out by using borrowed money to invest in the stock market, but if you are dollar cost averaging, your risk goes down. Since I assume they give you the money in payments as opposed to lump sums, that means you'll be dollar cost averaging your purchases. Keep in mind, if you do this, your account could go down. It could possibly go down a lot. If you started in 2000, you'd probably regret doing this. But, if you plan to contribute to a 401k after you graduate, I really see this just as some prepayments to the 401k. After you graduate, pay down your student debt in monthly installments coming out of this S&P fund. If you do this over a 10 year period, you'd be dollar cost averaging out of your S&P 500 fund as well. This would be a pretty good way to do this. The thing to be careful about is if you discontinue this program due to losing money. If you are they type of person that cannot stand to lose money, just go with the money market fund because chances are you'll sell at the bottom.
I think you mean German.
You mean when you don't devote all the country's resources to war, you can actually spend money on developing infrastructure at home and abroad that improves the lives of citizens?!?? AMAAAAAZING!!
Yeah, you're right. Maybe the US should have done this during WWII? Where would France be if that had happened?
wtf does that imply? and wtf does it have to do with CSS? and wtf is this stuff that matters?
Vista is a hot key word. He probably wants this article to show up on searches for Vista.
They have a strong belief that it [A tiered system] is needed in order to make sure that ISPs have the tools and funds to expand their already overtaxed networks.
If the ISPs' networks are seriously overtaxes, why don't they raise their prices? There is seriously no need to charge both consumers and suppliers of content. The suppliers are already charged to get their content on the net by ISPs anyways. They have two opportunties to raise their prices, why do they want this third charge? It can only be because they want to control what content is available (e.g. block out people that are compeeting with them in terms of voip and video). Anyways, I'm not sure legislation is the solution. I guess what we need to do is read the service agreements we have with broadband providers. I certainly would be more likely to subscribe to a broadband provider that gives me unlimited access to files as opposed to one that decides what content to let me see at what speed.
For some info on this: http://www.peacemagazine.org/archive/v19n4p06.htm
First of all the Sun thin clients are the price they are because Sun makes money on the server you have to buy to make it all work. So the actual price is significantly higher than $300/seat.
The Sun client works with x86 and sparc. So, you could load the server software on the $300 pcs you mentioned.
you've discussed a theoretical scenario that is supposed to increase the performance of the system should I be able to make a deal with all those neighbors I haven't met, but the price isn't any lower per seat.
It's not theoretical. Configurations like this have been tested. 30 pcs is not a whole lot. It's a lot lower cost than the equivelent power system. Like I said in my previous post for an 8 cpu system, you'd need a lot more money. Probably at least $2000 if not more. We're talking performance per $ here. Yes, you can probably buy a lowend pc for marginally less than the system I've proposed. That's because no one is mass producing the stuff needed to strip out the fat of the cost of the system I've proposed. If this sort of thing was mass produced the cost would clearly be reduced. This is how prototypes work. They are more expensive than the actual product. Don't take it so literally that you would actually go knock on your neighbors door and ask them if they want to buy a pc with you. This is where a company like Google comes in. They could share the connection among all the users automatically.
The bottom line is that there are technical, social, legal, and evironment issues involved with what you propose. I'm going to wait until at least some of these issues are resolved before betting on such an approach.
I don't see any social/legal/environmental issues? It's no different from getting a web hosting account today actually. Techincal issues, yes. I said that from the begining. You need a reliable amount of bandwidth and low latency. Also, someone needs to setup a system for sharing the systems that is close to the user globally. This is an enormous engineering problem to say the least, but it's something that's doable. Given that the product that I'm describing doesn't exist yet, I'd say it's a good idea to wait.
Dude...this isn't that complicated. The point of network computing is that by sharing computing resources you need less of them and get better performance. This is because of the massive amount of wasted CPU cycles that we're all aware of. There are thin clients available today for under $300 if you want a current exact example: http://www.sun.com/sunray/sunray1/. Since almost no one buys thin clients, the price is relatively high. Typically mass production would bring the cost down.
So lets look at it this way. Say you and your 100 closest neighbors all decided to get these $300 thin clients from Sun and you all pooled your money together to buy a grid of 30 lowend pcs for $300 (the price you mentioned) each that the thin clients would connect to that all run as a grid. The total amount you spent as a group would be $39,000. This is $390 per person total, lets just make it $400 to make it a round number. It could be marginally higher due to ethernet cables etc, but this works really well if you live in an apartment complex or a condo. So, for $400 (just a little bit more expensive than the cheapest of the cheap at dell), what do you get: You get a maxiumum of 30 cpus that are shared by 100 people. Since my previous estimate was that the average user uses no more than 1 cpu hour per day, you get access to this grid. Now, lets assume you don't want to let users use more than 25% of the total computing resources for their usage bursts. That means during your peak usage (opening programs, etc) you would have access to up to 7.5 cpus at the same time. So, in this example, for about $400 (slightly over the very cheapest box available) you get a 7.5 cpu system. I don't believe you can even buy a quad cpu system for less than a couple thousand dollars (certainly not less than $1000).
Nobody is going to do any real work on a cell phone.
It doesn't have to be a cell phone. I said, "It could be on the order of the price of a cell phone". This could be a desktop or better yet, it could something the size of a cell phone with blue tooth to connect your keyboard to.
The cost of the display isn't going to magically drop just because the device is connected to the network.
Of course it won't. In fact to me the display is the most important part of the computer. But displays are another story. We were talking about computers meaning the actual box (no monitor). So, for a comparison, many people buy a box for $500 - $1000 (not including the price of the monitor). Cellphones cost about $100 - $300. This is the savings I'm talking about. Yes, you will still need to buy a monitor and nice speakers and a printer. None of that will change just because most of your computing moves onto the net. Now, imagine this $100 - $300 device (aka box) connecting into this computing grid to do all your computation. This $100 - $300 dollar device will replace your $500 - $1000 device and give the appearance of being thousands of times faster due to the fact that whenever you do any computation, you can instantly provision thousands of CPUs just to your task. This is the idea.
Many companies already lease their PC's so there's nothing new there either.
Sure they do, but that means that they have a physical box sent out to their office where some system admin assembles it and maintains it. Companies spend a ton of money on administration and they will be the first and biggest winners of grid computing. Imagine a corporation that is expanding. Instead of buying new boxes for each new employee and replacing as they become out of date being able to tap into a grid and only use the amount of CPU time they really need. All the corporations would need is the basic $100 - $300 device I mentioned. This device could be used for a administrative assistant to check email, or for an engineer to run cad programs. They'd never need to upgrade the $100 - $300 device and they'd just get a bigger bill from the utility computing company that they are subscribing to if the engineer needs to run more simulations.
You've baked alot of assumptions in your argument
Sure I did, and even if you take the most conservative estimates for these things we're wasting a ton of cpu time. Again, look at how many wasted CPU cycles you have. Even my high estimate is that the average user uses 1 hour of cpu time per day. This means you're only using 1/24th the power you have in front of you and yes, I'm sure there are a few people who use more than 1 hour cpu time, but I said average. My job involves using the computer almost all day and I rarely use 100% CPU. It's mostly when I open a program for 1 - 2 seconds I'm using the CPU at 100%. Right now, as I'm typing, I'm using less than 1% of the CPU.
what is the net effect? You're still going to end up buying a hardware device that is either going to be neutered PC that will be $50 cheaper at best, or a specialized thin-client machine that will probably cost more than a PC.
Well, actually, there would be a number of advantages to this setup in terms of both price and performance:
1.) You could buy a client system for super cheap because basically all it needs to do is display data (and play sound) that it gets from the computing grid it's connected to. All other processing could be done on the grid (as soon as the latency and bandwidth issues I mentioned in my previous post are addressed). It could be on the order of the price of a cell phone (if not in 2006, definitly by 2010 given the current rate of progress in cell phone cpu power). Like many cell phone plans, there would be free models with particular service contracts. So, it would be cheap.
2.) This device which costs about the same as a cell phone would give you access to this entire grid of computers that is shared by millions of users. During your spike usage (see my previous post for details on what this is), you could get access to say 1000 CPUs that would run what you need run in 1000th of the time it typically takes to run the code if you had a single system. Even though you're still only using the same CPU time you would on your individual PC, it would seem as if you are using a computer 1000 times faster.
So, no it wouldn't be a neutered PC, it would be something way more powerful than any PC you can currently buy and it would cost about as much as cell phone. Remember cell phones get cheaper every year as well.
All for a device that turns into a lump of coal if your Internet connection fails. No thanks.
First of all, for most people their computer is essentially a lump of coal when the internet connection fails. I really can think of very few activities that I can do without connecting to the net. Secondly, as I said in my first post this requires two things:
1.) A reliable 5 mbps+ connection (emphasise reliable)
2.) Network latency no greater than 10 ms (about the minimum that the human eye can detect).
Yes, we don't have either of these things (for the most part) now. But we will. There are two technologies that will solve this: Wireless mesh networks and wimax connections. I don't have time to explain why this will work, but basically what it comes down to is: picture all the computers within a mile radius being potential wireless routers for you to connect to the net. You'd be simultaneously connected to hundreds of them and if any of them go down, you'd be connected to hundreds of other net connections at the same time. That's the idea. Google wimax and then google wireless mesh networks for more info.
So the guys at Slate thinks that the combined computing power of Google's umpteen million users is less than the power of their server farm? Unlikely, even for Google's impressive data centers.
But, what you forgot is that you're not using 99% of your computer's resources. What I mean is: look down at your performance meter. Probably what you'll see is that you're using around 1% of your cpu when you are actively typing at your computer. Sure occassionally this spikes up to 100% for a few seconds when you open an application, but for the most part your computer usage is very low. Ok, yeah yeah maybe it's higher for people who play graphic intensive games, etc but that's not the average user. Also, think about how often you actually use your computer. Most people don't sit in front of their computer 24/7. Even a super geek probably only sits in front of a computer for around 10 - 15 hours. So what's the total CPU time you actually use in a given day? I'd bet the average user who checks email and surfs the web and plays the occasional game uses well under 1 CPU hour per day. In fact, it might be as little as 10 minutes of CPU time per day for a lot of people (think your parents). So, when you say that Google's CPU power doesn't come close to the CPU power sitting in front of it's user base, you are right, but since their user base probably doesn't even use 1% of their CPU power, there's a compelling reason to switch more and more of the CPU cycles to the network because Google (or insert any other network computing company here) can get a lot closer to their maximum capacity by sharing their computing resources among it's huge user base.
Here's another way to think about it: What if you and 10 of your friends could pool your computing resources to make one super-computer? Maybe you could buy only 5 computers and save half the cost that you intended to spend. When you start your web browser, you could instantly get all 5 computing resources dedicated to your task of opening the browser. Then, after it's open your using about 1% of the CPU. Since you don't need more than 1% of the CPU to scroll through a web page and type, you don't notice a difference but when your usage spikes your browser INSTANTLY opens. In effect you have made it seem as if you are using the system that is 5 times faster than what you paid (at a 50% discount) for because you're sharing resources. Now, when your friend opens up the browser while you're typing he gets 99% of the total resources to open the browser. You don't even notice this though since you're typing and everything appears as normal. This is the power of using shared resources. Of course with only 10 users you could run into a problem if all 10 of you are on the computer at the same time ripping cds, talking over skype and watching streaming video all at once, but that's why it makes more sense for a company like Google to do this. With Millions of users the average CPU usage will not fluctuate much. Very quickly they'll figure out what the peak usage is. There of course will be peaks and troughs like there probably will be more users on at 2 pm as opposed to 4 am, but the sharing is still extremely efficient.
So, dollar for dollar this is why shared network computing makes sense and I have almost no doubt that it will be the norm eventually. To a large degree it depends on bandwidth usage and latency. When latency is below 10 ms round trip (the smallest percivable delay to the human eye) and you have more than 5 mBit/s bandwidth consistantly (dvd quality video), it will make no sense to have a local computer around (except possibly for privacy concerns). The privacy concerns can be solved be using encryption and locally storing your private keys though so even that is not a problem that cannot be resolved.
The best way for Windows users to compute untroubled (or less troubled) by malware is to switch to Mac OS X
Or in a more general sense: the best way to be safer from viruses is to use a platform that is not the mainstream one. Mac OS X is one example of something that could be used. Also, Linux, Free BSD, Solaris and various other platforms would work.
How pathetic are the teachers you know that they would seriously believe that they are seriously threatened by something like this?
The icon showed a gun pointing to a head, a bullet leaving the gun, and blood splattering from the head. It included the words "Kill Mr. VanderMolen,". You don't consider this a threat?
This is completely disrepectful. The kid needs to learn respect for his teachers. I'm not saying he needs to go to jail or anything, but there should be consequences for his actions. This is part of school and growing up in general. I would not expect a teacher to overlook this complete lack of respect. Any teacher that reacted the way you mentioned should not be allowed to teach.
Especially when the 'death threat' was not even communicated to him!
Obviously, he found out about it. So, someone communicated it to him.
Correct, death threats are not free speach. The classic example is that freedom of speach doesn't allow you to scream fire in a movie theatre.
First Amendment: Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.
The amendment even mentions the right "peaceably to assemble". Even though that is probably a reference to protests, the spirit of the law is clearly to allow do things as long as the are peacful. Also, what about the teacher he threatened? He has the right to life, liberty and the pursuit of happiness as well. Could a death threat disturb this right? Of course it could. If I was a teacher I would not be happy about it at all and expect the school to do what they have done.
Google plans to charge merchants a 2.2% commission on a sale, plus 30 cents per transaction using its payment service, according to people briefed on Google's pricing. That is higher than Pay-Pal's lowest published rate of a 1.9% commission plus 30 cents per transaction.
I'm surprised to see that they're charging a higher rate than Paypal.
A problem with your idea: the patent troll would contract with a real company to develop their "technology" (ie. patented ideas).
What would be wrong with that? If a company comes up with an idea and then contracts with another company to develop their product, we're ok. What we want to stop is companies that just come up with ideas and never pursue them and then sue anyone else who actually does the work of developing that technology.
Good luck proving that they aren't doing anything with their patents.
In the case of the company that this article discusses, I believe it would be quite easy to prove that they aren't pursuing the majority of their patents because their whole model is just to come up with ideas and patent them. In the case of a big company like IBM or something, yeah you have a point, but I think the biggest offenders of the whole patent thing has been companies that have no real business model other than Suing people (i.e. SCO). Obviously, this solution wouldn't fix problems like Amazon's one click patent or the hyperlink patent, but hey, it's a start and it's easy to implement.
Require companies to show that they have/or had the intent to develop the technologies they patented. These patent trolls would have a hard time showing that they actually intended to develop these technologies and thus their patents would be useless. On the otherhand, startups (and other inventors) would be able to use the system as it was intended to be used by disclosing their ideas in a protected way.
Agreed. Linus never intended to control the distributions. While the distros like Red Hat and Suze are important, he's always cared about the Kernel. I don't think this has changed much. He's still clearly in control of the Kernel which is an important role. People who think the Kernel doesn't matter don't understand what the Kernel does.
Personally, I'm guessing that "janet-on" is a bot someone made to try to accumulate karma
Yeah, either that or a rat brain/nanodot combo made to try to accumulate karma.
Yeah, I've seen the predictions about accelerating temperature changes and like you I am somewhat skeptical. I guess my point would be that none of these predictions take into consideration the fact that nanotechnology will enable the manufacture of ultra cheap solar panels that are cost effective within 20 years. Even if we use as much oil and coal as possible we will run out eventually. What's more likely is that no one will even consider using fossil fuels since the solar alternative becomes much more attractive.
The internet distribution cost Could be very low. with suitable multicast servers if everyone watched at a predefined time etc. and network maintainence fees don't go away they just shift from the tv bill to the internet bill.
I'm talking about Google's costs (not total costs). Since YOU already pay your internet bill, it doesn't need to be factored into Google's cost. This is why I think their will be a lot more content available.
Anyways, I can't find a single point in your entire comment that i found to be in any way correct.
Turst me, it's all correct. You must have misinterperted it if you think otherwise. You probably just disagree with my point about it being cheaper. But, it IS cheaper for _Google_ to provide video than it is for Cable companies because Google doesn't have to provide internet service for you because you already pay for it. It's not one of their costs. They only pay for their internet banwidth and servers. This is a lot easier than maintaining a network, sending out service men to fix problems, etc. If you disagree with another point, be specific and I'll gladly tell you why you're wrong.