Probably the fact that the business development guy you'd need to talk to at ESPN wouldn't take your phone call if he couldn't find your ticker symbol.
That and it's keyd off IP address blocks, and you don't have your own.
That's why they'd rather go with a model where the actual costs are hidden from you.
Look at the model that espn360.com is taking. You can't subscribe to espn360.com. You have to get your ISP to subscribe. Then the cost is hidden from you in your internet bill and they only have to bully your ISP over price, not every individual user.
Expect online movies to follow the same model. You'll either get the cost folded into the bill for something else you already pay for, and see it as a few dollars a month increase for a service most customers won't use, or it will be pay-per-view, but still billed with something you already have.
Once they have something like this going, they can slowly ratchet the prices up, and it'll be years before people catch on.
It's still the case with most telcos though. Even many of the ones that require you to buy the circuit through them still allow you to choose whichever carrier you like for your internet traffic. SBC is a good example of this, though they don't really advertise it.
If you get a frame-relay connection, or a direct T1, you pay the telecom for the copper, but anybody you'd like can route your bits. This has been true for decades. Before, during, and now after the line leasing regulations. The only thing that's changed is the price. (It went down. A lot. Leasing a line was about $2k a month in the early '90s, and is between $150 and $750 now depending on where you live.)
You're talking about last mile communications. This story is about long haul communications.
Either way, don't expect the good third parties to go away anytime soon. The smart ones got contracts to go along with the regulations. Not that the local telcos aren't required by regulation to share their lines, they're still bound by the decade+ long contractual obligations to the CLECs. And even though they pretend to be upset by it, they're actually making a tidy profit on the bundling of voice services, and the generally higher cost they charge to the CLEC over what they can get out of a direct sale. They might talk big, but don't expect them to try very hard to get the likes of Worldcom and Covad out of their COs.
You could switch to DSL, where you could get whatever provider you liked.
Or you could get a T1.
There are lots of ways not to be at the mercy of some large corporation. You'll just have to fork over the cash yourself instead of having the costs spread out over their whole customer pool.
Everybody is an equal on the internet... Corporation or end user alike. But that means everybody has to pay the same too.
It seems to me that what he was saying was that preventing somebody from realizing potential income is not theft. Just because it's not theft doesn't mean it's not illegal. It doesn't seem to me that he was making the point you think he was making.
I'd rather replace all the GNU software than all those device drivers....
But now that I think about it, other than libc and bash I don't really use much GNU software. The software running on my system right now comes from x.org, Mozilla, Apache, Sun, myself, and a few independant developers. There are plenty of shells and libc implementations, so going GNU-free wouldn't be that hard.
It's the drivers that are hard to replace. Even if you run lots of GNU.
I'm really, really struggling to think of a realistic situation in which it would be useful to take a single function, by itself, out of one program, and use it in a completely different program.
Struggling? You must not develop software then.. Sure, you may modify it some, but there's lots of basic functionality that every type of program needs.
Can you give an example of a realistic scenario where you could usefully copy someone else's code into your program, but still make a reasonable case that you weren't producing a derivative work?
Sure... If you pull in the linked list header from the kernel and use the macros in your database application, is your software a derivitive of the linux kernel?
Want more complex? How about a data compression routine? A checksum algorithm? ANy library out there that is GPLd that should probably be LGPLd?
As long as you don't forget to reproduce the copyright notice, list of conditions, and disclaimer in the documentation and/or other materials provided with your software, you mean.
The modified BSD license is the regular BSD license without that advertising clause.
You're seriously being obtuse on this point, by defining the word "use" as whatever you want it to mean. I could as easily say that the Microsoft EULA allows me to make copies of a Windows XP CD and sell them or give them away, because I have a license to "use" the software and I define "use" as whatever I want it to mean.
I'm not being obtuse, I'm being accurate. The Microsoft EULA defines "use" very specifically, and either way, that point is moot, since you don't have to code.
Why is that? because they allow a company to hijack your work.
I'm not sure you can say that is why. It might be why, but it might also be that the mere newness of Linux, and it's lack of features, attracted developers in the way open BSD code couldn't. That's why I started coding for Linux; not because of the GPL.
Dec, SGI, IBM, HP, even Sun were willing to support Linux because it could not be taken away
They were willing to support BSD based operating systems too. What do you think Tru64, Irix, AIX, and Solaris were based on? This is probably the best example of why the GPL is good for Linux though.
"Use" and "distribution" are two ENTIRELY separate things.
They are seperate, but not entirely. They are not mutually exclusive. Use isn't distribution, but distribution is a form of use. From a developers perspective, copying a small segment of code from somewhere completely unrelated into his/her own software may be the only way they intend to use a particular program.
They are only entirely seperate if you use a very limited definition of "use" (which the GPL does).
Please don't think that I'm saying this is bad. I don't know how anything I'm saying can be FUD when I'm not passing judgement.
Distributing derivative works is not reasonably known as "freedom of use".
It only seems unreasonable if you think of derivitive works as containing another application completely, or being based on an existing application. What if you just wanted to use a single function in your program that does something completely different? Is that still a derivative work? I'm sure that legally the answer would be yes, but it's an interesting mental exercise.
So in effect your argument is saying that the only safe software to use is BSD-licensed software.
Hell no. I'm just saying that those licenses are more friendly to corporate developers. There are safe ways to use GPL software, but using modified BSD licensed software can be done with no questions asked.
We do need to speculate, for two reasons. First, it's impossible to tell where modified BSD licensed code ends up. Second, Linux started from scratch, which left lots of interesting development to do that was already complete on BSD. That brought a lot of talented developers to the table that would have been bored working with BSDs existing framework.
It's impossible to tell what Linux would look like under a different license... So it's impossible to tell if it would be better or not.
Indeed. That's why I said it's not clear that it would be better.
I disagree that BSD takeup lags behind Linux. BSD licensed code ends up everywhere. Places you wouldn't even think to look. It just isn't called BSD anymore when it gets there. Again, up for debate/personal opinion whether this is good or not.
The GPL claims to protect the user's freedoms, but that's plain wrong. The GPL protects other users freedoms at the expense of any one individual's ability to use a piece of code completely freely. Corporate lawyers have a hard time coming to terms with that, and for good reason. The GPL is as much an ethical statement as a license, and it's not something that a commercial producer of software should take lightly. The modified Artistic license and modified BSD license are much more user friendly, and if Linux and most Linux software used those instead adoption would probably be greater. It's not clear that would be better though.
I wrote a (unfortunatly, closed source) filesystem that was remarkably similar to this once. Generally these types of filesystems are used when you're constantly writing new data. You're going to be eating the space anyway, but you want the reliability of syncronous writes with the performance of asyncronous cached writes. Reading from these filesystems is incredibly slow in comparison.
The version I wrote took advantage of the client's bursty IO pattern and used the slow periods to offload the data to an ext2 filesystem on a seperate disk. Hopefully your system memory was large enough that the offload to the secondary filesystem happened without any disk reads. Once that was done, the older sections of log could be re-used.... But only once the disk filled up and wrapped back to the beginning, because you want to keep your writes (essentially... There's other timing tricks you can play to get more speed) sequential.
There's been lots of research done on this method of write structuring. Look for papers on the "TRAIL" project (also closed source), for example.
Slashdot Mirror
My point was to tell a joke... You know... To make people laugh.
I guess I failed.
Forget that, ever seen the same thing on an old Sun Station? You know, like from back before Microsoft had login capability in their OS?
There are also some specious arguments and claims:
* Microsoft is the only one who "gets it" about how technology is an interrelated puzzle.
That's not so suspicious. I don't know many people who would tell you that getting software to work together should be puzzling...
Probably the fact that the business development guy you'd need to talk to at ESPN wouldn't take your phone call if he couldn't find your ticker symbol.
That and it's keyd off IP address blocks, and you don't have your own.
That's why they'd rather go with a model where the actual costs are hidden from you.
Look at the model that espn360.com is taking. You can't subscribe to espn360.com. You have to get your ISP to subscribe. Then the cost is hidden from you in your internet bill and they only have to bully your ISP over price, not every individual user.
Expect online movies to follow the same model. You'll either get the cost folded into the bill for something else you already pay for, and see it as a few dollars a month increase for a service most customers won't use, or it will be pay-per-view, but still billed with something you already have.
Once they have something like this going, they can slowly ratchet the prices up, and it'll be years before people catch on.
It's still the case with most telcos though. Even many of the ones that require you to buy the circuit through them still allow you to choose whichever carrier you like for your internet traffic. SBC is a good example of this, though they don't really advertise it.
Since you've been modded up, I'm surprised that nobody has bothered to explain to you yet that the web isn't the internet.
How does that matter?
If you get a frame-relay connection, or a direct T1, you pay the telecom for the copper, but anybody you'd like can route your bits. This has been true for decades. Before, during, and now after the line leasing regulations. The only thing that's changed is the price. (It went down. A lot. Leasing a line was about $2k a month in the early '90s, and is between $150 and $750 now depending on where you live.)
You're talking about last mile communications. This story is about long haul communications.
Either way, don't expect the good third parties to go away anytime soon. The smart ones got contracts to go along with the regulations. Not that the local telcos aren't required by regulation to share their lines, they're still bound by the decade+ long contractual obligations to the CLECs. And even though they pretend to be upset by it, they're actually making a tidy profit on the bundling of voice services, and the generally higher cost they charge to the CLEC over what they can get out of a direct sale. They might talk big, but don't expect them to try very hard to get the likes of Worldcom and Covad out of their COs.
You could switch to DSL, where you could get whatever provider you liked.
Or you could get a T1.
There are lots of ways not to be at the mercy of some large corporation. You'll just have to fork over the cash yourself instead of having the costs spread out over their whole customer pool.
Everybody is an equal on the internet... Corporation or end user alike. But that means everybody has to pay the same too.
Additionally, your little example there is terrible. A better analogy would be:
This tomato is not a vegetable.
That tomato is not a vegetable.
All tomatos aren't vegetables.
Still rediculous logic wise, but analogous to his example, and still comes to the correct conclusion.
The whole point was that the logic was terrible!
It seems to me that what he was saying was that preventing somebody from realizing potential income is not theft. Just because it's not theft doesn't mean it's not illegal. It doesn't seem to me that he was making the point you think he was making.
I'd rather replace all the GNU software than all those device drivers....
But now that I think about it, other than libc and bash I don't really use much GNU software. The software running on my system right now comes from x.org, Mozilla, Apache, Sun, myself, and a few independant developers. There are plenty of shells and libc implementations, so going GNU-free wouldn't be that hard.
It's the drivers that are hard to replace. Even if you run lots of GNU.
I'm surprised that people are surprised about this. Where do they think the water that drips out the end of their tailpipe comes from?
I'm really, really struggling to think of a realistic situation in which it would be useful to take a single function, by itself, out of one program, and use it in a completely different program.
Struggling? You must not develop software then.. Sure, you may modify it some, but there's lots of basic functionality that every type of program needs.
Can you give an example of a realistic scenario where you could usefully copy someone else's code into your program, but still make a reasonable case that you weren't producing a derivative work?
Sure... If you pull in the linked list header from the kernel and use the macros in your database application, is your software a derivitive of the linux kernel?
Want more complex? How about a data compression routine? A checksum algorithm? ANy library out there that is GPLd that should probably be LGPLd?
As long as you don't forget to reproduce the copyright notice, list of conditions, and disclaimer in the documentation and/or other materials provided with your software, you mean.
The modified BSD license is the regular BSD license without that advertising clause.
You're seriously being obtuse on this point, by defining the word "use" as whatever you want it to mean. I could as easily say that the Microsoft EULA allows me to make copies of a Windows XP CD and sell them or give them away, because I have a license to "use" the software and I define "use" as whatever I want it to mean.
I'm not being obtuse, I'm being accurate. The Microsoft EULA defines "use" very specifically, and either way, that point is moot, since you don't have to code.
Why is that? because they allow a company to hijack your work.
I'm not sure you can say that is why. It might be why, but it might also be that the mere newness of Linux, and it's lack of features, attracted developers in the way open BSD code couldn't. That's why I started coding for Linux; not because of the GPL.
Dec, SGI, IBM, HP, even Sun were willing to support Linux because it could not be taken away
They were willing to support BSD based operating systems too. What do you think Tru64, Irix, AIX, and Solaris were based on? This is probably the best example of why the GPL is good for Linux though.
"Use" and "distribution" are two ENTIRELY separate things.
They are seperate, but not entirely. They are not mutually exclusive. Use isn't distribution, but distribution is a form of use. From a developers perspective, copying a small segment of code from somewhere completely unrelated into his/her own software may be the only way they intend to use a particular program.
They are only entirely seperate if you use a very limited definition of "use" (which the GPL does).
Please don't think that I'm saying this is bad. I don't know how anything I'm saying can be FUD when I'm not passing judgement.
Isn't that obvious?
Distributing derivative works is not reasonably known as "freedom of use".
It only seems unreasonable if you think of derivitive works as containing another application completely, or being based on an existing application. What if you just wanted to use a single function in your program that does something completely different? Is that still a derivative work? I'm sure that legally the answer would be yes, but it's an interesting mental exercise.
So in effect your argument is saying that the only safe software to use is BSD-licensed software.
Hell no. I'm just saying that those licenses are more friendly to corporate developers. There are safe ways to use GPL software, but using modified BSD licensed software can be done with no questions asked.
We do need to speculate, for two reasons. First, it's impossible to tell where modified BSD licensed code ends up. Second, Linux started from scratch, which left lots of interesting development to do that was already complete on BSD. That brought a lot of talented developers to the table that would have been bored working with BSDs existing framework.
It's impossible to tell what Linux would look like under a different license... So it's impossible to tell if it would be better or not.
This is an incorrect statement.
You must be unclear on the meaning of the word "incorrect." Otherwise, you could never follow that sentence with this sequence of words:
You may use GPL code in whatever way you please, except
Indeed. That's why I said it's not clear that it would be better.
I disagree that BSD takeup lags behind Linux. BSD licensed code ends up everywhere. Places you wouldn't even think to look. It just isn't called BSD anymore when it gets there. Again, up for debate/personal opinion whether this is good or not.
It's funny, that you can say I'm absolutely wrong followed directly by saying exactly wy I'm absolutely right.
The GPL claims to protect the user's freedoms, but that's plain wrong. The GPL protects other users freedoms at the expense of any one individual's ability to use a piece of code completely freely. Corporate lawyers have a hard time coming to terms with that, and for good reason. The GPL is as much an ethical statement as a license, and it's not something that a commercial producer of software should take lightly. The modified Artistic license and modified BSD license are much more user friendly, and if Linux and most Linux software used those instead adoption would probably be greater. It's not clear that would be better though.
I wrote a (unfortunatly, closed source) filesystem that was remarkably similar to this once. Generally these types of filesystems are used when you're constantly writing new data. You're going to be eating the space anyway, but you want the reliability of syncronous writes with the performance of asyncronous cached writes. Reading from these filesystems is incredibly slow in comparison.
The version I wrote took advantage of the client's bursty IO pattern and used the slow periods to offload the data to an ext2 filesystem on a seperate disk. Hopefully your system memory was large enough that the offload to the secondary filesystem happened without any disk reads. Once that was done, the older sections of log could be re-used.... But only once the disk filled up and wrapped back to the beginning, because you want to keep your writes (essentially... There's other timing tricks you can play to get more speed) sequential.
There's been lots of research done on this method of write structuring. Look for papers on the "TRAIL" project (also closed source), for example.