For starters, multicast assumes a broadcast model, and not an on-demand model. If 1,000,000 customers want to watch 1,000,000 different things, or the people who want to watch the same thing are not along the same path, then it won't work.
Secondly, how do you bill for multicast? The content provider is using one upstream into their ISP. The customers are using 1,000,000 downstreams from their ISPs. Both parties are footing the bill on their usage. But, what about when the content provider's ISP has to branch the single content stream into five other ISPs? Peering revenue gets all messed up suddenly. The content provider is being billed for 1x bits under the assumption that's all it will actually cost the ISP to deliver 1x bits. But now the content provider is being billed for 1x bits but the ISP has to deliver 5x bits. Does the ISP engage in the horrible nightmare of tracking every multicast stream to figure out how many times it branches, to which peers, and how many bits it transfers over each branch (because of clients joining and leaving the tree)?
Another thing to consider is the technical cost of maintaining multicast state in all the necessary routers in the entire Internet. We haven't achieved that sort of storage capacity at the random access speeds necessary for worldwide general multicast support.
There are reasons multicast was not turned on in the public Internet. In many ways, it doesn't work once more than a few people want to be doing it. The backbones and ISPs all know about multicast. The decision was made not to turn it on. I'm sure there are additional reasons that I'm unaware of, but those three are the ones that come to mind right now.
I think many/. readers will be thinking that you pay for your bandwidth, so you can pay more, and get more bandwidth. But the reality is that pipes are physically/technically limited. You pay for your water, and if you leave your faucet on you'll get more water and pay more. But you can only get so much; there is a maximum limit that the pipes going to your home will carry. It is the same way for serving massive video content.
Let's say you want to serve content to 100 customers, each of whom has a 3Mbps downlink. So that's 300Mbps. Easy enough to get that at a hosting center. But now let's say you want to serve content to 1,000,000 customers. You need 3,000,000Mbps, or 3Tbps sustained bandwidth. Where can you buy that, no matter what the price? Do you think providing 3Tbps of bandwidth is as simple or as cheap as 1,000,000 x 3Mbps? If that sort of thing were true, you could buy a Petabyte of disk space for a linear increase in cost as well.
Profit without product, service, or other material trappings.
Religious organizations figured that one out a while back. The hard part is finding someone to be your prophet/martyr. Unfortunately, you run a 50/50 chance of being branded a cultist. In which case you will become the prophet/martyr.
I'm in a similar situation, but for me the real problem is that the current generation of content and hardware is just sucky. In particular, you cannot output 1080p from the player, and you cannot get Dolby TrueHD (which is only available on one disc) or DTS-HD (I think that's the name) on the digital outputs to anything that can decode it. The current Toshiba player will only decode two channels of TrueHD to the analog outputs, and 5.1 out of the 7.1 possible channels on the other formats. And I suspect the analog audio components of the current hardware is crap.
What I need is: 1) A high-quality receiver or processor that takes the high-definition digital audio input on HDMI, Firewire, or something. As of now this doesn't exist. 2) A high-quality 1080p front projector that doesn't cost $10,000 (preferrably with a built-in HQV video processor). This also doesn't exist. 3) A hardware player that outputs 1080p over HDMI/DVI (HQV video processor if not in projector). Doesn't exist. 4) Content that is mastered off the original film, rather than a transcode from an existing DVD to 1080p. Does exist, I think..
In other words, I want to be listening to an orchestra from a pretty decent seat in the house, not the nosebleed section.
Well, except Chinese characters are stand-alone. So a dictionary wouldn't help, and you might end up with several different plausible sentences on each pole. And you'd never end up with nonsense words. Possible nonsense sentences. Which one is the right one? Nonsense sentences might not be nonsense to other people.
That's also how it is in the U.S. Many, many Americans have no idea Echelon is real, and not a plot item in the TV show Alias, or that Osama Bin Laden was a U.S. operative trained to perform terrorist activities against the Soviet Union, or the magnitude of privacy losses due to acts passed after 9/11, etc.
I think you've gotten Open Standards and Open Source mixed up. Open Standards is like those fittings and making sure everyone uses the same ones. Open Source and Closed Source would be like two different types of power tools. If you buy/build the open source power tool, you can do things to it that you wouldn't be able to with the closed source power tool (e.g. change saw blades on a circular saw) if you put in your own effort. While with the closed source one, you might get a different level/experience of support, but you wouldn't be able to tinker with it to change its behavior (although it might still support changing saw blades).
The least obtrusive game copy protections, IMHO, were those that required the manuals. But they were easy enough to defeat programmatically (SoftICE...), too.
Or easy enough to defeat using paper and pencil. What is most annoying is if you accidentally misplace said manual.
I also used to make copies of my floppies to run games off the copied disks, instead of the originals. This ensured 1) I didn't accidentally write to a game floppy disk with the write tab enabled 2) The copied disk could undergo rough treatment and I wouldn't be unhappy
I still do #2 with music CDs. I don't with movie DVDs because they aren't in frequent use in comparison.
Or, you could say that this is the strength of taxes being used to fund public science, rather than the strength of open source software. Now imagine if the estimated end-cost of the Iraqi invasion (US$2 trillion over the years + 3,000 US citizens + 30,000 Iraq citizens) was being put towards public science?
OpenSuSE is not supported by Novell. It is supported by the community. You can purchase a retail version of SuSE to get support. Mac OS X does indeed come with free telephone support, and in-person support at Apple Stores. Novell and the OpenSuSE community also does _not_ support OpenOffice. You would have to communicate with the OpenOffice community for that support. YaST is not even close to as user-friendly as the Mac OS X System Preferences or Software Update features. And you can still get very confused or mess things up with YaST.
One opinion I've heard from a coworker is that remaining in development will make your salary hit a ceiling. (The only exceptions to this that I know of are for IBM Fellows or STSMs, who are paid a base salary + a percentage of their product revenue/profit.) The only way to further increase your salary is to enter management positions, or other positions where you have control over money distribution. Plus, managers tend to be considered less replaceable than developers. I don't know if there's any real validity to this theory.
As a caucasian American in my major (EE), I have always been in a minority. I'd estimate that between 60-75% of the students in my classes are students from outside the country: India, China, Indonesia, etc. Does this study even consider taking that into account? Glancing it over briefly, it sure doesn't seem so.
You realize that unless you actually ask each of those students what their citizenship is, you are making assumptions based on their ethnicity (i.e. physical appearance). There do happen to be a lot of non-Caucasian Americans in America.
The stock market is based on the idea that one item is worth more to someone than it is to someone else. But the only reason the price fluctuates is because there's a scarcity of the item. If everyone could buy as many shares of Google that they wanted, without having to require someone else to have lost that share, then the whole system would fall apart. So I don't think the stock market model can apply to a commodity that is essentially of unlimited number. Plus, the value of an item is equal to all owners of that item. Person X and person Y who both own one share of Google will end up earning or losing an equal amount of money from a given point in time.
To answer the question, is a less popular song worth $0.99? It certainly is to the person who is buying it at that price. By the proposed logic, a song which perhaps I do not like at all would be priced much higher than what I am personally willing to pay for it. Perhaps I am willing to pay $1.99 for a song by Orbital, but only $0.25 for a song by Norah Jones. My prices for those items are inversely proportional to their popularity. To make them proportional to popularity might result in less income (although perhaps more sales) for Orbital and vice versa for Norah Jones. I'm not an economist so I don't know the right math to apply here.
So maybe you want to try personalized pricing. That's only going to invite major consumer backlash as people discover they're being charged more for a song than the person sitting next to them. Best case scenario is some sort of trading community where song purchases are traded so that everyone can buy their songs at $0.25. And now you're back to the original flat-pricing scheme. Worst case scenario is everyone drops iTMS because now Apple is being evil and greedy.
And I assume then you are willing to pay the additional cost associated with employing real people instead of an automated system? Those costs can either be passed down to you in the original product or service price, or you can be charged for the phone call. Or, somehow you can convince the shareholders to take a lower return on their investment so you can get better service when you call. There's a reason things have gone this direction, and it does have a lot to do with what the market asked for.
Titleserv asserts that it owns copies of the programs because it paid Krause a substantial sum to develop them and has an undisputed right to possess and use them permanently.
I'm copying this from the parent, so hopefully it's verbatim. I wonder what it means in this case to own copies of the program. Work for hire typically means the company owns _all_ copies of the program, and it is licensed for use by others. Consultant work typically means the consultant owns _all_ copies of the program (or photograph, or whatever) and the company has a license to use it.
But this wording seems to imply that some copies are owned by the company, and some copies are owned by the consultant. In other words, that there can be two owners. A copy is now being referred to in the same sense as a physical copy would be, such as a photograph. The right to derivitive works is solely in the rights of the copyright holder. Are these two owners copyright owners? Or are they owners in the sense of a physical object, such that I am allowed to destroy, draw on, etc. a photograph.
If you look at what Windows and OSX has going for it, you'll quickly note that it's simple and easy to do just about anything.
This should be changed to say, "If you look at what Windows and Mac OS X have going for them, you'll quickly note that it's simple and easy to do just about anything (well on Mac OS X, or poorly on Windows)." People put up with crap on Windows that no one in their right mind should have to put up with. It's old rhetoric around here, but people will make up all sorts of excuses for using a piece of crap if they think that's what they are supposed to be using.
For plans on how to go about buying and selling the moon, read The Man Who Sold the Moon. I think Robert A. Heinlein should be like a quote source in Congress and schools or something.
If I tell you I installed your software on 400,000 machines, and ask you for $60k. How likely is it you will assume I was able to accomplish that legitimately?
Actually, I take that back. Adware gets put into games and other downloads all the time now. In that situation, it's quite possible I could distribute the adware to 400,000 machines quickly.
Part of the appeal of the IPods is that they do what they do *well*. Interface, and sound quality. Now I'm not sure how they made it so that every kid under the age of 30 *has* to have one...that's another story.
How old _are_ you?
Among my species, thirty-year-olds generally aren't considered children.
Right now I'm sitting here with a 2 inch long scratch on my tum... uh.. stom.. uh.. crap factory because last night my clutzy-ass-cat took a swipe at the cord to my sweat pants.
Somehow I don't think you've died due to disembowelment yet.
Seriously though, if you ever watch a cat playing, or allow one to climb on you, you'd know their claws are for either climbing, traction like cleats, piercing and holding onto prey, and executing power slides around sharp corners on hardwood floors (claws are involved somehow, I just haven't figured out how yet).
The same arguments of a "protection racket" can easily be applied towards software vendors that also charge for services/consulting. Everyone knows that those companies are making huge bucks on their consulting fees. Much more than on their software sales. (Thus, the push for many software vendors to start providing consulting services.) If the software was easy, bug-free, and provided features and interoperability, then there wouldn't be any consultants required for installation and integration.
I do not think the level of violence in the US is acceptable. However, one thing to consider is that the US is extremely heterogenous, and the social system encourages selfish behavior. No other country in the world contains so many people living near each other that are so different from each other, culturally, physically, languages, social norms, religions, etc. And there is very much a look out for yourself attitude among people here.
I imagine turning 10% of Germany into Chinese people (really Chinese, not just genetically), another 10% into Mexicans, another 10% into Indians, another 10% into Africans, and you'll have some social disruption going on as well.
Slashdot Mirror
For starters, multicast assumes a broadcast model, and not an on-demand model. If 1,000,000 customers want to watch 1,000,000 different things, or the people who want to watch the same thing are not along the same path, then it won't work.
Secondly, how do you bill for multicast? The content provider is using one upstream into their ISP. The customers are using 1,000,000 downstreams from their ISPs. Both parties are footing the bill on their usage. But, what about when the content provider's ISP has to branch the single content stream into five other ISPs? Peering revenue gets all messed up suddenly. The content provider is being billed for 1x bits under the assumption that's all it will actually cost the ISP to deliver 1x bits. But now the content provider is being billed for 1x bits but the ISP has to deliver 5x bits. Does the ISP engage in the horrible nightmare of tracking every multicast stream to figure out how many times it branches, to which peers, and how many bits it transfers over each branch (because of clients joining and leaving the tree)?
Another thing to consider is the technical cost of maintaining multicast state in all the necessary routers in the entire Internet. We haven't achieved that sort of storage capacity at the random access speeds necessary for worldwide general multicast support.
There are reasons multicast was not turned on in the public Internet. In many ways, it doesn't work once more than a few people want to be doing it. The backbones and ISPs all know about multicast. The decision was made not to turn it on. I'm sure there are additional reasons that I'm unaware of, but those three are the ones that come to mind right now.
I think many /. readers will be thinking that you pay for your bandwidth, so you can pay more, and get more bandwidth. But the reality is that pipes are physically/technically limited. You pay for your water, and if you leave your faucet on you'll get more water and pay more. But you can only get so much; there is a maximum limit that the pipes going to your home will carry. It is the same way for serving massive video content.
Let's say you want to serve content to 100 customers, each of whom has a 3Mbps downlink. So that's 300Mbps. Easy enough to get that at a hosting center. But now let's say you want to serve content to 1,000,000 customers. You need 3,000,000Mbps, or 3Tbps sustained bandwidth. Where can you buy that, no matter what the price? Do you think providing 3Tbps of bandwidth is as simple or as cheap as 1,000,000 x 3Mbps? If that sort of thing were true, you could buy a Petabyte of disk space for a linear increase in cost as well.
Profit without product, service, or other material trappings.
Religious organizations figured that one out a while back. The hard part is finding someone to be your prophet/martyr. Unfortunately, you run a 50/50 chance of being branded a cultist. In which case you will become the prophet/martyr.
I'm in a similar situation, but for me the real problem is that the current generation of content and hardware is just sucky. In particular, you cannot output 1080p from the player, and you cannot get Dolby TrueHD (which is only available on one disc) or DTS-HD (I think that's the name) on the digital outputs to anything that can decode it. The current Toshiba player will only decode two channels of TrueHD to the analog outputs, and 5.1 out of the 7.1 possible channels on the other formats. And I suspect the analog audio components of the current hardware is crap.
What I need is:
1) A high-quality receiver or processor that takes the high-definition digital audio input on HDMI, Firewire, or something. As of now this doesn't exist.
2) A high-quality 1080p front projector that doesn't cost $10,000 (preferrably with a built-in HQV video processor). This also doesn't exist.
3) A hardware player that outputs 1080p over HDMI/DVI (HQV video processor if not in projector). Doesn't exist.
4) Content that is mastered off the original film, rather than a transcode from an existing DVD to 1080p. Does exist, I think..
In other words, I want to be listening to an orchestra from a pretty decent seat in the house, not the nosebleed section.
Well, except Chinese characters are stand-alone. So a dictionary wouldn't help, and you might end up with several different plausible sentences on each pole. And you'd never end up with nonsense words. Possible nonsense sentences. Which one is the right one? Nonsense sentences might not be nonsense to other people.
That's also how it is in the U.S. Many, many Americans have no idea Echelon is real, and not a plot item in the TV show Alias, or that Osama Bin Laden was a U.S. operative trained to perform terrorist activities against the Soviet Union, or the magnitude of privacy losses due to acts passed after 9/11, etc.
I think you've gotten Open Standards and Open Source mixed up. Open Standards is like those fittings and making sure everyone uses the same ones. Open Source and Closed Source would be like two different types of power tools. If you buy/build the open source power tool, you can do things to it that you wouldn't be able to with the closed source power tool (e.g. change saw blades on a circular saw) if you put in your own effort. While with the closed source one, you might get a different level/experience of support, but you wouldn't be able to tinker with it to change its behavior (although it might still support changing saw blades).
The least obtrusive game copy protections, IMHO, were those that required the manuals. But they were easy enough to defeat programmatically (SoftICE...), too.
Or easy enough to defeat using paper and pencil. What is most annoying is if you accidentally misplace said manual.
I also used to make copies of my floppies to run games off the copied disks, instead of the originals. This ensured
1) I didn't accidentally write to a game floppy disk with the write tab enabled
2) The copied disk could undergo rough treatment and I wouldn't be unhappy
I still do #2 with music CDs. I don't with movie DVDs because they aren't in frequent use in comparison.
Or, you could say that this is the strength of taxes being used to fund public science, rather than the strength of open source software. Now imagine if the estimated end-cost of the Iraqi invasion (US$2 trillion over the years + 3,000 US citizens + 30,000 Iraq citizens) was being put towards public science?
Please define "geek cool" versus "nerd cool". And then explain how Microsoft fits into it at all. TIA.
OpenSuSE is not supported by Novell. It is supported by the community. You can purchase a retail version of SuSE to get support. Mac OS X does indeed come with free telephone support, and in-person support at Apple Stores. Novell and the OpenSuSE community also does _not_ support OpenOffice. You would have to communicate with the OpenOffice community for that support. YaST is not even close to as user-friendly as the Mac OS X System Preferences or Software Update features. And you can still get very confused or mess things up with YaST.
One opinion I've heard from a coworker is that remaining in development will make your salary hit a ceiling. (The only exceptions to this that I know of are for IBM Fellows or STSMs, who are paid a base salary + a percentage of their product revenue/profit.) The only way to further increase your salary is to enter management positions, or other positions where you have control over money distribution. Plus, managers tend to be considered less replaceable than developers. I don't know if there's any real validity to this theory.
As a caucasian American in my major (EE), I have always been in a minority. I'd estimate that between 60-75% of the students in my classes are students from outside the country: India, China, Indonesia, etc. Does this study even consider taking that into account? Glancing it over briefly, it sure doesn't seem so.
You realize that unless you actually ask each of those students what their citizenship is, you are making assumptions based on their ethnicity (i.e. physical appearance). There do happen to be a lot of non-Caucasian Americans in America.
The stock market is based on the idea that one item is worth more to someone than it is to someone else. But the only reason the price fluctuates is because there's a scarcity of the item. If everyone could buy as many shares of Google that they wanted, without having to require someone else to have lost that share, then the whole system would fall apart. So I don't think the stock market model can apply to a commodity that is essentially of unlimited number. Plus, the value of an item is equal to all owners of that item. Person X and person Y who both own one share of Google will end up earning or losing an equal amount of money from a given point in time.
To answer the question, is a less popular song worth $0.99? It certainly is to the person who is buying it at that price. By the proposed logic, a song which perhaps I do not like at all would be priced much higher than what I am personally willing to pay for it. Perhaps I am willing to pay $1.99 for a song by Orbital, but only $0.25 for a song by Norah Jones. My prices for those items are inversely proportional to their popularity. To make them proportional to popularity might result in less income (although perhaps more sales) for Orbital and vice versa for Norah Jones. I'm not an economist so I don't know the right math to apply here.
So maybe you want to try personalized pricing. That's only going to invite major consumer backlash as people discover they're being charged more for a song than the person sitting next to them. Best case scenario is some sort of trading community where song purchases are traded so that everyone can buy their songs at $0.25. And now you're back to the original flat-pricing scheme. Worst case scenario is everyone drops iTMS because now Apple is being evil and greedy.
And I assume then you are willing to pay the additional cost associated with employing real people instead of an automated system? Those costs can either be passed down to you in the original product or service price, or you can be charged for the phone call. Or, somehow you can convince the shareholders to take a lower return on their investment so you can get better service when you call. There's a reason things have gone this direction, and it does have a lot to do with what the market asked for.
Titleserv asserts that it owns copies of the programs because it paid Krause a substantial sum to develop them and has an undisputed right to possess and use them permanently.
I'm copying this from the parent, so hopefully it's verbatim. I wonder what it means in this case to own copies of the program. Work for hire typically means the company owns _all_ copies of the program, and it is licensed for use by others. Consultant work typically means the consultant owns _all_ copies of the program (or photograph, or whatever) and the company has a license to use it.
But this wording seems to imply that some copies are owned by the company, and some copies are owned by the consultant. In other words, that there can be two owners. A copy is now being referred to in the same sense as a physical copy would be, such as a photograph. The right to derivitive works is solely in the rights of the copyright holder. Are these two owners copyright owners? Or are they owners in the sense of a physical object, such that I am allowed to destroy, draw on, etc. a photograph.
Hm...confusing.
If you look at what Windows and OSX has going for it, you'll quickly note that it's simple and easy to do just about anything.
This should be changed to say, "If you look at what Windows and Mac OS X have going for them, you'll quickly note that it's simple and easy to do just about anything (well on Mac OS X, or poorly on Windows)." People put up with crap on Windows that no one in their right mind should have to put up with. It's old rhetoric around here, but people will make up all sorts of excuses for using a piece of crap if they think that's what they are supposed to be using.
For plans on how to go about buying and selling the moon, read The Man Who Sold the Moon. I think Robert A. Heinlein should be like a quote source in Congress and schools or something.
If I tell you I installed your software on 400,000 machines, and ask you for $60k. How likely is it you will assume I was able to accomplish that legitimately?
Actually, I take that back. Adware gets put into games and other downloads all the time now. In that situation, it's quite possible I could distribute the adware to 400,000 machines quickly.
Part of the appeal of the IPods is that they do what they do *well*. Interface, and sound quality. Now I'm not sure how they made it so that every kid under the age of 30 *has* to have one...that's another story.
How old _are_ you?
Among my species, thirty-year-olds generally aren't considered children.
Right now I'm sitting here with a 2 inch long scratch on my tum... uh.. stom.. uh.. crap factory because last night my clutzy-ass-cat took a swipe at the cord to my sweat pants.
Somehow I don't think you've died due to disembowelment yet.
Seriously though, if you ever watch a cat playing, or allow one to climb on you, you'd know their claws are for either climbing, traction like cleats, piercing and holding onto prey, and executing power slides around sharp corners on hardwood floors (claws are involved somehow, I just haven't figured out how yet).
The same arguments of a "protection racket" can easily be applied towards software vendors that also charge for services/consulting. Everyone knows that those companies are making huge bucks on their consulting fees. Much more than on their software sales. (Thus, the push for many software vendors to start providing consulting services.) If the software was easy, bug-free, and provided features and interoperability, then there wouldn't be any consultants required for installation and integration.
Are they going to be teaching at a college or university? Or at local schools? That's certainly a big difference in pay grade.
Not sure if you can catch/spread the plague by eating an infected corpse. Seems unlikely this would move through the food chain.
You're not familiar with zombies, are you. Gwarrh! Braaiiins!
I do not think the level of violence in the US is acceptable. However, one thing to consider is that the US is extremely heterogenous, and the social system encourages selfish behavior. No other country in the world contains so many people living near each other that are so different from each other, culturally, physically, languages, social norms, religions, etc. And there is very much a look out for yourself attitude among people here.
I imagine turning 10% of Germany into Chinese people (really Chinese, not just genetically), another 10% into Mexicans, another 10% into Indians, another 10% into Africans, and you'll have some social disruption going on as well.