Compare the amount of Mac sales to the amount of PC sales and if Apple moved over to AMD, I doubt Intel would give a second glance either. Apple probably buys around 10 percent of all laptop chips that Intel produces, and mostly goes for the more expensive ones, so I would estimate about 20 percent of dollar revenue. And Apple buys a good amount of expensive quad core server chips as well. And they don't buy any of the $50 low end chips that end up in your $399 PC. So financially, losing Apple would be a major hit for Intel.
Anyway, I wonder if this is a good thing for PR. If companies point to this to be even more reluctant to adopt F/OSS solutions, and make subcontractors indemnify themselves, and basically make everybody CTheirA even more tightly, it will likely be a bad thing for everybody involved; Open Source Software gets less support from the mainstream, services cost more (because of all that R&D poured into re-inventing this "wheel" thing everybody's talking about), and everybody misses out on the fruits of useful labor that could be shared. I think you underestimate the intelligence of most companies.
Most companies understand the GPL very well. Most understand that you can use GPL'd software without payment, but with legal consequences, they understand these consequences, and make an educated decision whether to use GPL'd software and source code or not in their own products.
The only exceptions would be companies run by complete morons (they might panic now), or companies who were willing to commit copyright infringement because they thought they could get away with it (they will now be trying to cover their tracks). But most companies are not affected by this case in any way.
Sure it the idea sounds scary that the government could potentially plant DNA evidence, but in this case I think the positive outweighs the negative. Imagine if they could have caught deranged serial killers like Ted Bundy or Ed Gein before they killed so many innocent people. Sure there are evil people in the government but I think most intend to help and not hurt the public. It wouldn't have stopped Harold Shipman.
It wouldn't stop anyone from going postal.
It also wouldn't have stopped Enron management, who may very well be responsible for the deaths of more people than many mass murderers (through people committing suicides, not having enough money to pay for medical treatment, and so on).
You can't be serious? Suppose your wife was raped and killed. The police could get DNA samples from her body, and know immediately who did it. Without DNA, and assuming no-one saw the guy break in and leave afterwards, and wore gloves to avoid leaving fingerprints, how exactly would he be caught? In many rape cases, the perpetrator is known, but it is difficult to impossible to prove that what happened was rape and not consentual sex.
In many other rape cases, the child is too afraid to say anything or nobody believes it anyway.
And anyone who starts an argument with "suppose your wife was raped and killed" is seriously sick in their mind, and their DNA should be collected with highest priority.
This was posted anonymously (0 points score, not posted by me) and nobody is going to read it, but it's interesting so I'll repost it:
I work for an ISP, and part of my job description is dealing with abuse complaints (including DMCA complaints.) The usual modus operandi when we receive a complaint is to verify that the address was in use, verify that traffic was coming from that port, and then send the user's name back. Someone else (not me) also contacts the user.
A few weeks ago, we got a request from a user who wanted to use the network to do research (it's a college town)--specifically research related to Bit Torrent. They wanted permission to conduct this research, and assured us that they would not be engaging in copyright infringement. They had written a special client which advertises hashes of the torrent, but never sends data.
Fast forward to about a week ago--two MPAA complaints against their IP address. We could also verify that there was no way that they were sending or receiving files--their traffic usage was actually pretty small.
My assumption based upon the data is that the company who sent the DMCA notification just looks at the IP addresses that the tracker advertises is a part of the swarm. They don't bother to download and verify that the torrent is what they think that it is, and it's in-line with other debacles of the sort (remember the usher mp3 one from several years back?), as well as some anomalies we'd seen where a notification time was shortly after a person had released their DHCP lease (sometimes the trackers don't immediately notice when a user has disconnected.) I'd love to say that I'm shocked, nay horrified that they would gather evidence in this way, but frankly, it makes good business sense. Why waste the time downloading every file? Almost everyone who gets tagged has been infringing copyright--there just aren't that many 'fake' files out there. And since almost everyone is actually doing it, they'll almost all settle. Those few who do not are easily handled by simply dropping the case.
For this reason, I would expect that blacklists like Peerguardian wouldn't really work all that well. Lots of people give me anecdotal evidence when I bring this up ("Well it works for me!") but that's kinda like saying that this elephant repellant is working because there aren't any elephants on my front lawn. I know a lot of people who engage in a lot of filesharing, and none of them have ever gotten DMCA complaints. Just based upon the number I see in my job, I know that they're not sending them to very many people, in general.
Yes, and parodies are themselves copyrighted, owned by those who created the parodies, not owned by the RIAA companies who made the originals. This is why the RIAA has $150,000 per infringement liability for every BritneySpearsSux.mp3 they download/upload that is not their original content. They can reasonably argue that the download happened with the consent of the copyright holder, if the copyright holder made it available for download on a P2P network. On the other hand, when MediaSentry downloads music owned by companies represented through RIAA, that also happens with the consent of the copyright holder, which means the upload is also legal.
It could very well be that Jimmy Wales is being swiftboated.
Likewise, the governor of New York is being called to resign because he allegedly saw a hooker. This has nothing whatsoever to do with his ability to govern the state, but sharp political rhetoric is being flung about nevertheless. The Bill Clinton affair also springs to mind.
All this means is that Jimmy Wales has some political opponents who are willing to fight dirty. He might be no saint, but Wikipedia seems to be doing pretty darn well for the most part. Your comparison is wildly off the mark. If a politician sees a hooker, that has very little to do with his ability to govern the state. If Jimmy Wales takes money to edit an article on Wikipedia, then this goes directly to the heart of Wikipedia.
The second part is devoid of any logic. If, for example, TheRegister publishes such damning accusations, then (1) that does not make them "political opponents" (Are you seriously claiming that Wales has "political opponents"? I think the editors of TheRegister just can't stand him; that has nothing to do with politics). Second, it doesn't "only mean they are willing to fight dirty". There are quite a few possibilities, and four of these possibilities are: 1. They fight dirty. Or 2. They are mistaken. Or 3. They are exaggerating. Or 4. They are exactly on the mark.
You're analogy is poor, because dropping packets when a network is busy is part of the standard. Also, you failed to capture the fact that the rest packets are forged. The finally problem with you analogy is that it didn't mention cars. You are looking at this at the wrong level. The individual packets are not the letters. The complete file transfer equals the letter. If a few packets are dropped, even if a few percent of all packets are deliberately dropped, then the protocol copes with it by resending the packets (that would be an analogy to the postman missing a few letters before he starts his round; they will be delivered on the next day). The single forged packed makes sure that the complete file transfer doesn't happen, which is the same as throwing the letter away.
Answering the question "what can the FCC do": I would assume that they could take their license away, as the final threat.
I don't think the Comcast situation has much to do with net neutrality. Lack of "net neutrality" would mean that a service provider slows down some traffic and not other traffic. So your bittorrent might take 12 hours instead of 1, but work without problems. But that is not what Comcast does: They actively manipulate the traffic that goes through their system, sending fake abort messages to bittorrent clients. That, I think, could be very much in violation of whatever license they need.
If I sent you a letter and it arrived in five days instead of one day, I would complain. If the post office deliberately threw away my letters, I would complain a lot louder.
It doesn't even succeed against known threats. They have regular security screenings where a TSA agent sneaks through a fake bomb disguised as a back brace or something innocuous. Less than a 50% success rate at stopping it. If "the terrorists" actually get to that point, it's more likely than not that TSA will let them through. I think what you are missing here is the psychological aspect. They watch out for people who behave in a certain way. Someone with a bomb in his backpack that he wants to explode in a mass of people, taking himself to heaven in the process, behaves very much different from someone who has a fake bomb in their back trying to get one over the security guards. I don't want security to check for TSA agents sneaking in with fake bombs, I want them to look out for people with real bombs.
Which is a good thing. Distributing workload to computers that would be idleing otherwise is efficent. Not to mention, that whole thing scales well so that the distributor can deal with a slashdot effect. You are not distributing workload to computers. You are distributing the transport of data, and that is most likely not a good thing.
Different pipes have different costs. The pipe that goes from your home to the ISP and back is the most expensive one you can find. A national broadcaster has a much bigger and more cost-effective pipe available. When the BBC started transmitting programs through the internet, ISPs were not much interested in caching the content, because the delivery from BBC to the ISP has minimal cost. But they complain very loud about the cost that they have to transport the data on to the end user, and Bittorrent doubles that cost.
It would be different if you could distribute computational workloads to computers. There you are right, a computer sitting idle would have just slightly increased cost in electricity if it was used to perform some computation and I would think that even the most efficient server has higher total cost than the additional cost for electricity on a less efficient home computer.
I'll pay the extra $0.02 for each movie/CD/hour of tv I download, vs the cost associated with creating the physical DVD/CD/cable channel and getting it from Mexico/Taiwan/Time Warner to my house. It isn't $0.02 for each movie. In the UK, most ISPs use the BT (British Telecom) network and buy bandwidth on it to reach end users. With an estimated 66 percent average usage of the available bandwidth, downloads to your home cost the ISP about £0.55 to £0.60 per GB, that would be about $1.10 to $1.20. One 55 minute show that I downloaded from iTMS had a file size of 593 MB, which makes it about £0.33 or $0.66 for that show. The ISP carries that cost, but in the end what you pay every month has to be enough to cover the cost of the ISP, so you have an effective cost of $0.66.
The problem with Bittorrent is that the upload goes through the same expensive network. So you add another $0.66 for the upload, plus whatever inefficiencies bittorrent introduces (not much, I hope). But when Apple sends the same stream to me, their cost to deliver the stream is much less, because they use the biggest, meanest and cheapest pipe you can find to push their data out.
I don't agree with that... I'm paying for my upload even if I don't use it. I'm not paying more if I spend my whole time torrenting files... Your ISP pays for the cost initially.
However, your ISP must make a profit or go bankrupt. If the ISP's cost grows, and he doesn't want to go bankrupt (which doesn't serve anyone), then he has the choices: Stop you from using as much bandwidth, get rid of you as a customer, charge more for your account, or charge more for every account. In the end, you pay.
I would have liked to see an analysis of the actual total distribution cost - not the cost to the originator, but the total.
In the UK, cost of internet data consists of two parts: The cost of getting the data to your ISP, and the cost of getting the data from the ISP to your home, usually using bandwidth bought at wholesale prices from BT (British Telecom). The cost for the ISP to send data to your home is around £0.60 per Gigabyte, But the cost to get data from a huge source to the ISP is much lower. For example, getting a movie from the BBC server to your ISP has negligible cost, compared to the cost of getting the same movie from the ISP to your home. A Bittorrent would obviously send data from many, many homes to ISPs, and then from the ISPs to different homes. In other words, the data goes through the expensive route twice instead of once. I would think that the actual cost is actually almost twice as high using Bittorrent.
An interesting question is: Who pays for it? In the end, your ISP pays the cost. The ISP will of course calculate your monthly payments so that they will come out ahead, and if you use torrents a lot they might convince you to get a more expensive package with more bandwidth. So in the end you will end up paying the cost.
I would have liked to see an analysis of the actual total distribution cost - not the cost to the originator, but the total.
In the UK, cost of internet data consists of two parts: The cost of getting the data to your ISP, and the cost of getting the data from the ISP to your home, usually using bandwidth bought at wholesale prices from BT (British Telecom). The cost for the ISP to send data to your home is around £0.60 per Gigabyte, But the cost to get data from a huge source to the ISP is much lower. For example, getting a movie from the BBC server to your ISP has negligible cost, compared to the cost of getting the same movie from the ISP to your home.
A Bittorrent would obviously send data from many, many homes to ISPs, and then from the ISPs to different homes. In other words, the data goes through the expensive route twice instead of once. I would think that the actual cost is actually almost twice as high using Bittorrent.
An interesting question is: Who pays for it? In the end, your ISP pays the cost. The ISP will of course calculate your monthly payments so that they will come out ahead, and if you use torrents a lot they might convince you to get a more expensive package with more bandwidth. So in the end you will end up paying the cost.
As horrible as it may sound, unless the iPhone can accept pushed mail from an exchange server - any hope of breaking into the business market is all but dashed. Interestingly, that is exactly what Phil Schiller demonstrated - email pushed from an exchange server.
The interesting thing for your IT department is that the iPhone gets its email from the _exchange server_ itself, not from a server that connects to a server that connects to the exchange server, so with the iPhone you don't need any additional costly hardware and additional costly software licenses to get your push email, unlike with some other phones.
Actually, I was able to install the SDK on my G4 using only a couple extra manual steps. Because it's not an intel machine it skipped a few critical packages. I merely installed them manually and I was then able to compile some sample applications and run them in the iPhone simulator. You probably won't be able to load your programs onto an iPhone but at least you can do development work using the simulator. That's interesting. That would imply that the complete iPhone simulator is written in some high-level language (like C) and must be an ARM interpreter, not a compiler. I would have thought that a dual processor ARM chip is powerful enough that you need an assembly translator to run it in real-time, even on a Core 2 Duo. Something like Rosetta, but for ARM->Intel instead of PPC->Intel.
Ever written code for these "other phone substrates"? It's such a monumental pain in the arse that, assuming developers can be bothered at all, they'll certainly reimplement vast swathes of the application from the ground up due to fundamental (and often archaic) platform differences anyway. This SDK will more than likely clean up if only because it's a derivative of OS X and thus programming for it'll be a dream. Is very any other phone where the software development VP would go live on stage and demonstrate writing a (tiny) application, compiling it, downloading it to the phone and running it, in front of a live audience?
Steve Jobs presentation from yesterday is available on the Apple site. Could anyone who complains about the lack of Microsoft Exchange compatibility please watch the keynote first. Most of the posts so far can be answered by saying "You may not have watched the keynote yesterday, but..."
You may think it's reasonable, but I don't see how you think it's even possibly allowed by the GPLv3: "Installation Information" for a User Product means any methods, procedures, authorization keys, or other information required to install and execute modified versions of a covered work in that User Product from a modified version of its Corresponding Source. Doesn't apply in this situation. If Apple shipped the iPhone with GPL'd software, then Apple would have to provide you with source code, a way to compile modified source code, and the ability to install modified code on the iPhone. Even in that case I'm not sure that they would have to provide that possibility for free.
In this case, however, the developer writes software, and unfortunately has to pay some money to get it installed on people's iPhones. You have the same problem after getting the source and modifying it. But it is not because of anything the developer did, except if the developer was Apple. So Apple could be in trouble if they ported gnu chess to the iPhone. Anyone else wouldn't.
Is this GPL compatible? If I offer a GPL app on this store and provide source, the user can't use that source to modify the app without paying a fee, right? Is this a problem under the GPLv2? v3? I don't think that is a problem. You can download the free SDK. You can then develop an application based on GPL code. Unfortunately you have to pay $99 to get it distributed. But that isn't your fault, it is Apple's fault. So I buy your application (GPL doesn't prevent you from charging money for it), and ask you for the source code. I then modify the source code with the free SDK. Now I am just in exactly the same version as you were: To distribute it, I'll have to pay some money. Not your fault. You are not preventing me from downloading the application to the iPhone.
So not only do I need OS X to develop for the iPhone I also need an Intel processor? That's... BS. What about those with OS X and older Power PC computers? You can buy a Mac Mini for £399. You'll probably find one cheaper on eBay. Every hour an Apple developer wastes on getting the SDK to work on PowerPC machines is an hour that is lost for improving the Intel SDK. Apple probably thinks that a developer who hasn't bought a new Macintosh in the last two years is not going to be a great loss for them.
I was under the impression that a judge in civil cases has a lot of discretion. There is plenty of discovery orders in civil cases. While the judge may not agree to say "Let the other side see...", he may say "Show me, I want to see...". In any case, some discovery can be filed as sealed documents. That would seem to apply here. The judge can't keep anything like that to himself, he would have to allow access to the defendant's lawyers and expert witnesses (but not to the defendant and the public). Lawyers and expert witnesses would obviously be in big trouble if anything leaked out.
On the other hand, there is now precedent that you can't hide behind "proprietary methods"; I think there was the case of a manufacturer of breathalyzer equipment that ran into this problem. Of course they can refuse to open up their "proprietary methods", but then any evidence based on these proprietary methods would be invalid. In case of the breathalyzer equipment, nobody could actually force them to open up their code, but in practice every case based on their equipment would have been dropped, and the police would never again have bought their equipment.
The problem with the Air is that I don't know how I'd install XP on it, and I need XP for proper MS Office utils. You need the USB DVD drive according to an Apple support article. Probably the easiest way would be to buy the MacBook Air at an Apple Store and I am sure they would lend you the drive if you arrive there with cash for the MBA in one hand and a Windows XP CD in the other hand.
Slashdot Mirror
Most companies understand the GPL very well. Most understand that you can use GPL'd software without payment, but with legal consequences, they understand these consequences, and make an educated decision whether to use GPL'd software and source code or not in their own products.
The only exceptions would be companies run by complete morons (they might panic now), or companies who were willing to commit copyright infringement because they thought they could get away with it (they will now be trying to cover their tracks). But most companies are not affected by this case in any way.
It wouldn't stop anyone from going postal.
It also wouldn't have stopped Enron management, who may very well be responsible for the deaths of more people than many mass murderers (through people committing suicides, not having enough money to pay for medical treatment, and so on).
In many other rape cases, the child is too afraid to say anything or nobody believes it anyway.
And anyone who starts an argument with "suppose your wife was raped and killed" is seriously sick in their mind, and their DNA should be collected with highest priority.
A few weeks ago, we got a request from a user who wanted to use the network to do research (it's a college town)--specifically research related to Bit Torrent. They wanted permission to conduct this research, and assured us that they would not be engaging in copyright infringement. They had written a special client which advertises hashes of the torrent, but never sends data.
Fast forward to about a week ago--two MPAA complaints against their IP address. We could also verify that there was no way that they were sending or receiving files--their traffic usage was actually pretty small.
My assumption based upon the data is that the company who sent the DMCA notification just looks at the IP addresses that the tracker advertises is a part of the swarm. They don't bother to download and verify that the torrent is what they think that it is, and it's in-line with other debacles of the sort (remember the usher mp3 one from several years back?), as well as some anomalies we'd seen where a notification time was shortly after a person had released their DHCP lease (sometimes the trackers don't immediately notice when a user has disconnected.) I'd love to say that I'm shocked, nay horrified that they would gather evidence in this way, but frankly, it makes good business sense. Why waste the time downloading every file? Almost everyone who gets tagged has been infringing copyright--there just aren't that many 'fake' files out there. And since almost everyone is actually doing it, they'll almost all settle. Those few who do not are easily handled by simply dropping the case.
For this reason, I would expect that blacklists like Peerguardian wouldn't really work all that well. Lots of people give me anecdotal evidence when I bring this up ("Well it works for me!") but that's kinda like saying that this elephant repellant is working because there aren't any elephants on my front lawn. I know a lot of people who engage in a lot of filesharing, and none of them have ever gotten DMCA complaints. Just based upon the number I see in my job, I know that they're not sending them to very many people, in general.
Likewise, the governor of New York is being called to resign because he allegedly saw a hooker. This has nothing whatsoever to do with his ability to govern the state, but sharp political rhetoric is being flung about nevertheless. The Bill Clinton affair also springs to mind.
All this means is that Jimmy Wales has some political opponents who are willing to fight dirty. He might be no saint, but Wikipedia seems to be doing pretty darn well for the most part. Your comparison is wildly off the mark. If a politician sees a hooker, that has very little to do with his ability to govern the state. If Jimmy Wales takes money to edit an article on Wikipedia, then this goes directly to the heart of Wikipedia.
The second part is devoid of any logic. If, for example, TheRegister publishes such damning accusations, then (1) that does not make them "political opponents" (Are you seriously claiming that Wales has "political opponents"? I think the editors of TheRegister just can't stand him; that has nothing to do with politics). Second, it doesn't "only mean they are willing to fight dirty". There are quite a few possibilities, and four of these possibilities are: 1. They fight dirty. Or 2. They are mistaken. Or 3. They are exaggerating. Or 4. They are exactly on the mark.
Answering the question "what can the FCC do": I would assume that they could take their license away, as the final threat.
I don't think the Comcast situation has much to do with net neutrality. Lack of "net neutrality" would mean that a service provider slows down some traffic and not other traffic. So your bittorrent might take 12 hours instead of 1, but work without problems. But that is not what Comcast does: They actively manipulate the traffic that goes through their system, sending fake abort messages to bittorrent clients. That, I think, could be very much in violation of whatever license they need.
If I sent you a letter and it arrived in five days instead of one day, I would complain. If the post office deliberately threw away my letters, I would complain a lot louder.
Different pipes have different costs. The pipe that goes from your home to the ISP and back is the most expensive one you can find. A national broadcaster has a much bigger and more cost-effective pipe available. When the BBC started transmitting programs through the internet, ISPs were not much interested in caching the content, because the delivery from BBC to the ISP has minimal cost. But they complain very loud about the cost that they have to transport the data on to the end user, and Bittorrent doubles that cost.
It would be different if you could distribute computational workloads to computers. There you are right, a computer sitting idle would have just slightly increased cost in electricity if it was used to perform some computation and I would think that even the most efficient server has higher total cost than the additional cost for electricity on a less efficient home computer.
The problem with Bittorrent is that the upload goes through the same expensive network. So you add another $0.66 for the upload, plus whatever inefficiencies bittorrent introduces (not much, I hope). But when Apple sends the same stream to me, their cost to deliver the stream is much less, because they use the biggest, meanest and cheapest pipe you can find to push their data out.
However, your ISP must make a profit or go bankrupt. If the ISP's cost grows, and he doesn't want to go bankrupt (which doesn't serve anyone), then he has the choices: Stop you from using as much bandwidth, get rid of you as a customer, charge more for your account, or charge more for every account. In the end, you pay.
Same post again, with line breaks:
I would have liked to see an analysis of the actual total distribution cost - not the cost to the originator, but the total.
In the UK, cost of internet data consists of two parts: The cost of getting the data to your ISP, and the cost of getting the data from the ISP to your home, usually using bandwidth bought at wholesale prices from BT (British Telecom). The cost for the ISP to send data to your home is around £0.60 per Gigabyte, But the cost to get data from a huge source to the ISP is much lower. For example, getting a movie from the BBC server to your ISP has negligible cost, compared to the cost of getting the same movie from the ISP to your home. A Bittorrent would obviously send data from many, many homes to ISPs, and then from the ISPs to different homes. In other words, the data goes through the expensive route twice instead of once. I would think that the actual cost is actually almost twice as high using Bittorrent.
An interesting question is: Who pays for it? In the end, your ISP pays the cost. The ISP will of course calculate your monthly payments so that they will come out ahead, and if you use torrents a lot they might convince you to get a more expensive package with more bandwidth. So in the end you will end up paying the cost.
I would have liked to see an analysis of the actual total distribution cost - not the cost to the originator, but the total. In the UK, cost of internet data consists of two parts: The cost of getting the data to your ISP, and the cost of getting the data from the ISP to your home, usually using bandwidth bought at wholesale prices from BT (British Telecom). The cost for the ISP to send data to your home is around £0.60 per Gigabyte, But the cost to get data from a huge source to the ISP is much lower. For example, getting a movie from the BBC server to your ISP has negligible cost, compared to the cost of getting the same movie from the ISP to your home. A Bittorrent would obviously send data from many, many homes to ISPs, and then from the ISPs to different homes. In other words, the data goes through the expensive route twice instead of once. I would think that the actual cost is actually almost twice as high using Bittorrent. An interesting question is: Who pays for it? In the end, your ISP pays the cost. The ISP will of course calculate your monthly payments so that they will come out ahead, and if you use torrents a lot they might convince you to get a more expensive package with more bandwidth. So in the end you will end up paying the cost.
The interesting thing for your IT department is that the iPhone gets its email from the _exchange server_ itself, not from a server that connects to a server that connects to the exchange server, so with the iPhone you don't need any additional costly hardware and additional costly software licenses to get your push email, unlike with some other phones.
Steve Jobs presentation from yesterday is available on the Apple site. Could anyone who complains about the lack of Microsoft Exchange compatibility please watch the keynote first. Most of the posts so far can be answered by saying "You may not have watched the keynote yesterday, but..."
In this case, however, the developer writes software, and unfortunately has to pay some money to get it installed on people's iPhones. You have the same problem after getting the source and modifying it. But it is not because of anything the developer did, except if the developer was Apple. So Apple could be in trouble if they ported gnu chess to the iPhone. Anyone else wouldn't.
On the other hand, there is now precedent that you can't hide behind "proprietary methods"; I think there was the case of a manufacturer of breathalyzer equipment that ran into this problem. Of course they can refuse to open up their "proprietary methods", but then any evidence based on these proprietary methods would be invalid. In case of the breathalyzer equipment, nobody could actually force them to open up their code, but in practice every case based on their equipment would have been dropped, and the police would never again have bought their equipment.