I mean, aside from all the squabbling over definitions.. if an embedded system wants to build it's functions into kernel-space, it'll have to abide by the GPL. That said....
Who cares? Embedded systems does NOT mean 'set-top boxes'...
Embedded systems generally do not need even a fraction of what the linux kernel offers, and they are generally so hardware specific that you don't *need* the source.. you don't have the tools to build it anyway.
What I'm trying to say is, nothing is broken here. If someone wants an OS they can extend at the kernel level, and not publish the results, they don't need to use linux...there are lots of other options, some open-source, some not.
This is not a 'flaw' in the GPL.. it just means that the GPL is not suitable for , in this case a license on an open-source extensible embedded kernel. It *IS* the license that linux and crew publish the Linux kernel under.... and THAT is THAT.... if you don't like it, don't use it.
Enterprise ready doesn't mean merely 'suitable for use by a big company with a big budget'.
Big companies use Win95 still.. does that make it enterprise-ready?
And just cause it runs print servers or departmental file servers doesn't make it enterprise ready either.
Do banking systems run on it? Real-time transaction processing for stock markets? Critical insturment control for nuclear power plants/other major industries? No....
That's what they mean by 'enterprise-ready' in this context; the ability to stake the data of your whole enterprise on it.
The original serial copy management system that by-law must be implemented on digital home audio recording devices , and is in use on CD (and in them mp3 format, but nobody uses it) that never really gets used (I'm sure some DAT drives use it) has 2 bits.
1 bit for 'copyright' and another bit for 'original'.
If the copyright bit is set, and the original bit is also set, the copy software is supposed to allow a copy, but turn off the original bit.
If the copyright bit is set, but the original bit is off, then the device/software is supposed to refuse to copy.
If the copyright bit is unset, then you can make all the copies you want.
See, what they were scared of with digital copies was that, a copy is as good as the original. This scheme was to prevent serial copying going on forever... it meant that sure, you yourself with the original could hand out hundreds of copies even, but those who you handed them to couldn't....
Of course, scms specifically exempted computers...
The problem we all have with these copy-control systems isn't the systems themselves. The average consumer doesn't *care*. They're glad to have the cool new tech.
The problem is the DMCA that makes it illegal for us to purchase gear and then modify it to avoid a copy control system, or to share information about how to do that.
It will be cracked, we will use the media for our own purposes, even storing music previously under copy protection. We will have ways of re-recording things without content control, and no content control system will keep us from moving the data into another medium.
It is unbreakable. Is he not describing a one-time pad?
If he is, you can't break it, period.
You could generate every possible key and run it against the ciphertext.. but this equates to , for a ciphertext of 8 characters, generating every possible compbination of 8 characters and deciding which is the 'real' message, just like an infinite number of monkeys...
IP space wasn't supposed to be a commodity, but it is now, due to ineffective planning (or whatever you want to call it).
An ISP should *only* ever enforce two rules.
1) How much bandwidth you can use.
2) Reserve the right to terminate your account if you cause them grief (spamming, etc..).
They shouldn't say 'don't run servers' 'only one computer' 'only for casual at-home use' etc.... they should simply make the bandwidth rules and prices reflect this.
Is this a result of the dot-com crash? I mean, I gather most of these huge clec's didn't actually make any money yet... they just banked on market-share and figuring out how to make money later?
Have they run out of public funds?
He didn't so much have a problem with the part of the contract you are referring to.
He had a problem with the clauses stating that 'any IP you come up with while in the employ of the company belongs to the company', meaning, he can't do anything, anywhere, on his own time without them owning it.
The clause you are referring to does make sense; the company is paying you to write code for them, so you shouldn't be able to add your 'own' previously developed code and then claim they owe you.
Can you explain what a 'negative' temperature would be, given that 0k is the absence of all temperature? I assume this is some kind of quantum theory thing...
can you elaborate a bit? I've never heard of 'negative heat' before.
1) This isn't patent infringement; it's a trademark dispute. They are *very* different.
2) The original license on SSH, which OpenSSH is based on, states basically that the name 'ssh' should not be used if the derivative work is incompatable with the ssh protocols specified in the RFCs. This implies that it's OKAY to use the name ssh if you ARE compatable.
3) Under trademark law, you *MUST* defend your mark. This mark is already highly dilute; there are many software packages known to the plaintiff here that use the ssh name, and he's known about openssh for quite some time.
4) ssh is the name of a publicly available protocol as specified in the rfc's, and the name of the common unix command used to implement such a protocol, regardless of vendor. It's already diluted. THe common techie does *not* automatically associate 'ssh' with meaning 'the product from company X'. They mean the protocol.. just like ftp or anything else.
That's why he can't enforce it.
It's not about what we can get away with. If he wanted to protect the name he should have made it clear from the beginning, and that hasn't happened.
What you are referring to is correct, it's called 'trade secret'. The recipe for KFC or Coke is a trade secret, and as long as it remains a secret, it can be protected quite extensively by the law. They could get the recipe back because it was possible to do so, as those who had it hadn't revealed it yet, in other words, it hadn't become public knowledge yet.
If the recipe for coke was wide public knowledge, because coke 'ignored' a threat to it for a week ro two, and everyone in the country and every newspaper had a copy in it, then they could no longer claim *ANY* trade secret protections on it. It's no longer a secret.
we're not simply reading from it, you have to be able to send keys to it or something in order to unlock the drive if it has a dvd in it. That's why... it can't be done through.
You may not see why the lack of enforcement should diminish their ability to prosecute, however, long-standing trademark law DOES.
Who even knew it was trademarked? Did they ever attempt to let anyone know this before? No.
SSH is the common word for a protocol; everyone knows that. It's already been what trademark law calles, 'diluted'. It's a common name now; once something is common like that, you can't take it back just because it's trademarked. Everyone already has something in their mind when they think of 'ssh', and it's not this guys product in particular, it's merely the protocols involved.
The import levy is in effect, and it's 5c per cd-r.
It is to be paid by any a) media manufactured for sale in the country, and b) media imported for sale. Canadians are free to import media from the US without paying the levy.. it only applies to import for sale.
That's why it makes no perceptible difference in the consumer price, like everyone thought it would.
The fees levied in Canada for CD-R is only 5c per cd-r... it doesn't at all make them 'double' the price.
CD-R-audio, the ones you *require* to use blackbox cd audio recorders (like lots have those, right?..) have a larger tarrif, because they are supposdly obviously for audio use. But who cares about them.
If it was a garage operation they didn't know about, that would be different. How could they enforce it if they didn't know about it? You're right there.
That is not the case here. OpenSSH is highly visible, and they even admit they KNEW about it a long time ago. It doesn't matter how big or small they are; if they are infringing, you must do something about it.
that the suit wasn't over the term 'x86' but over the '386, 80386, 486, 80486' etc individually...
Intel clamied they were 'trademarks', and that by making compatable chips and calling them '486' etc... they were infringing.
The court ruled that you can't trademark a number, a part number even.
That's why the 80586 was called the Pentium, and all advertising shifted to saying 'Pentium' instead of '586', because the word Pentium *Could* be trademarked.
I mean, if they've ignored him all along, that's different.
Though I'd still maintain that it's news to most of us that 'ssh' is a registered trademark (tm), given the open protocol specs by the same name, and the plethora of tools that use the ssh name.
You say "By your reasoning, any company with a trademark should be fascistic about enforcing their rights as soon and as often as possible. There can be no leniency or they will loose their trademark. That's not exactly a situation I want to come about (not that I have the least say in it, of course)."
Well.. that's *exactly* how trademark law DOES work. You enforce them, or you lose them. That's why you see so many stupid-sounding tradmark violation suits, because the lawyers are making sure they actively police their mark, because if they don't, they LOSE it, for good.
Slashdot Mirror
I mean, aside from all the squabbling over definitions.. if an embedded system wants to build it's functions into kernel-space, it'll have to abide by the GPL. That said....
Who cares? Embedded systems does NOT mean 'set-top boxes'...
Embedded systems generally do not need even a fraction of what the linux kernel offers, and they are generally so hardware specific that you don't *need* the source.. you don't have the tools to build it anyway.
What I'm trying to say is, nothing is broken here. If someone wants an OS they can extend at the kernel level, and not publish the results, they don't need to use linux...there are lots of other options, some open-source, some not.
This is not a 'flaw' in the GPL.. it just means that the GPL is not suitable for , in this case a license on an open-source extensible embedded kernel. It *IS* the license that linux and crew publish the Linux kernel under.... and THAT is THAT.... if you don't like it, don't use it.
Enterprise ready doesn't mean merely 'suitable for use by a big company with a big budget'.
Big companies use Win95 still.. does that make it enterprise-ready?
And just cause it runs print servers or departmental file servers doesn't make it enterprise ready either.
Do banking systems run on it? Real-time transaction processing for stock markets? Critical insturment control for nuclear power plants/other major industries? No....
That's what they mean by 'enterprise-ready' in this context; the ability to stake the data of your whole enterprise on it.
Why just one copy?
The original serial copy management system that by-law must be implemented on digital home audio recording devices , and is in use on CD (and in them mp3 format, but nobody uses it) that never really gets used (I'm sure some DAT drives use it) has 2 bits.
1 bit for 'copyright' and another bit for 'original'.
If the copyright bit is set, and the original bit is also set, the copy software is supposed to allow a copy, but turn off the original bit.
If the copyright bit is set, but the original bit is off, then the device/software is supposed to refuse to copy.
If the copyright bit is unset, then you can make all the copies you want.
See, what they were scared of with digital copies was that, a copy is as good as the original. This scheme was to prevent serial copying going on forever... it meant that sure, you yourself with the original could hand out hundreds of copies even, but those who you handed them to couldn't....
Of course, scms specifically exempted computers...
The problem we all have with these copy-control systems isn't the systems themselves. The average consumer doesn't *care*. They're glad to have the cool new tech.
The problem is the DMCA that makes it illegal for us to purchase gear and then modify it to avoid a copy control system, or to share information about how to do that.
If there's a market for it, let it come.
It will be cracked, we will use the media for our own purposes, even storing music previously under copy protection. We will have ways of re-recording things without content control, and no content control system will keep us from moving the data into another medium.
It is unbreakable. Is he not describing a one-time pad?
If he is, you can't break it, period.
You could generate every possible key and run it against the ciphertext.. but this equates to , for a ciphertext of 8 characters, generating every possible compbination of 8 characters and deciding which is the 'real' message, just like an infinite number of monkeys...
It is true that a one-time pad is not breakable.
Is very simple.
IP space wasn't supposed to be a commodity, but it is now, due to ineffective planning (or whatever you want to call it).
An ISP should *only* ever enforce two rules.
1) How much bandwidth you can use.
2) Reserve the right to terminate your account if you cause them grief (spamming, etc..).
They shouldn't say 'don't run servers' 'only one computer' 'only for casual at-home use' etc.... they should simply make the bandwidth rules and prices reflect this.
That, I understood. Appreciated.
Is this a result of the dot-com crash? I mean, I gather most of these huge clec's didn't actually make any money yet... they just banked on market-share and figuring out how to make money later?
Have they run out of public funds?
He didn't so much have a problem with the part of the contract you are referring to.
He had a problem with the clauses stating that 'any IP you come up with while in the employ of the company belongs to the company', meaning, he can't do anything, anywhere, on his own time without them owning it.
The clause you are referring to does make sense; the company is paying you to write code for them, so you shouldn't be able to add your 'own' previously developed code and then claim they owe you.
Because it's not about 'postal' or 'electronic'. Email is an anarchy...
It's about communication.
A company should not be able to 'harass' me with their communications if I do not wish them to, period. Just as a person can't.
Can you explain what a 'negative' temperature would be, given that 0k is the absence of all temperature? I assume this is some kind of quantum theory thing...
can you elaborate a bit? I've never heard of 'negative heat' before.
1) This isn't patent infringement; it's a trademark dispute. They are *very* different.
2) The original license on SSH, which OpenSSH is based on, states basically that the name 'ssh' should not be used if the derivative work is incompatable with the ssh protocols specified in the RFCs. This implies that it's OKAY to use the name ssh if you ARE compatable.
3) Under trademark law, you *MUST* defend your mark. This mark is already highly dilute; there are many software packages known to the plaintiff here that use the ssh name, and he's known about openssh for quite some time.
4) ssh is the name of a publicly available protocol as specified in the rfc's, and the name of the common unix command used to implement such a protocol, regardless of vendor. It's already diluted. THe common techie does *not* automatically associate 'ssh' with meaning 'the product from company X'. They mean the protocol.. just like ftp or anything else.
That's why he can't enforce it.
It's not about what we can get away with. If he wanted to protect the name he should have made it clear from the beginning, and that hasn't happened.
What you are referring to is correct, it's called 'trade secret'. The recipe for KFC or Coke is a trade secret, and as long as it remains a secret, it can be protected quite extensively by the law. They could get the recipe back because it was possible to do so, as those who had it hadn't revealed it yet, in other words, it hadn't become public knowledge yet.
If the recipe for coke was wide public knowledge, because coke 'ignored' a threat to it for a week ro two, and everyone in the country and every newspaper had a copy in it, then they could no longer claim *ANY* trade secret protections on it. It's no longer a secret.
And no.. that has nothing to do with this case.
You still have to make it know that it's a mark... you can still protect what you do without registering the mark. THis is trademark, not patent.
we're not simply reading from it, you have to be able to send keys to it or something in order to unlock the drive if it has a dvd in it. That's why... it can't be done through.
You may not see why the lack of enforcement should diminish their ability to prosecute, however, long-standing trademark law DOES.
Who even knew it was trademarked? Did they ever attempt to let anyone know this before? No.
SSH is the common word for a protocol; everyone knows that. It's already been what trademark law calles, 'diluted'. It's a common name now; once something is common like that, you can't take it back just because it's trademarked. Everyone already has something in their mind when they think of 'ssh', and it's not this guys product in particular, it's merely the protocols involved.
The import levy is in effect, and it's 5c per cd-r.
It is to be paid by any a) media manufactured for sale in the country, and b) media imported for sale. Canadians are free to import media from the US without paying the levy.. it only applies to import for sale.
That's why it makes no perceptible difference in the consumer price, like everyone thought it would.
The fees levied in Canada for CD-R is only 5c per cd-r... it doesn't at all make them 'double' the price.
CD-R-audio, the ones you *require* to use blackbox cd audio recorders (like lots have those, right?..) have a larger tarrif, because they are supposdly obviously for audio use. But who cares about them.
It's not a read only device, it's read-only media.
You can still send commands to the drive.. which involves 'writing'.
If it was a garage operation they didn't know about, that would be different. How could they enforce it if they didn't know about it? You're right there.
That is not the case here. OpenSSH is highly visible, and they even admit they KNEW about it a long time ago. It doesn't matter how big or small they are; if they are infringing, you must do something about it.
that the suit wasn't over the term 'x86' but over the '386, 80386, 486, 80486' etc individually...
Intel clamied they were 'trademarks', and that by making compatable chips and calling them '486' etc... they were infringing.
The court ruled that you can't trademark a number, a part number even.
That's why the 80586 was called the Pentium, and all advertising shifted to saying 'Pentium' instead of '586', because the word Pentium *Could* be trademarked.
I mean, if they've ignored him all along, that's different.
Though I'd still maintain that it's news to most of us that 'ssh' is a registered trademark (tm), given the open protocol specs by the same name, and the plethora of tools that use the ssh name.
You say "By your reasoning, any company with a trademark should be fascistic about enforcing their rights as soon and as often as possible. There can be no leniency or they will loose their trademark. That's not exactly a situation I want to come about (not that I have the least say in it, of course)."
Well.. that's *exactly* how trademark law DOES work. You enforce them, or you lose them. That's why you see so many stupid-sounding tradmark violation suits, because the lawyers are making sure they actively police their mark, because if they don't, they LOSE it, for good.
I'm not sure what you're asking...
The term 'cola' is not trademarked... so the 'cola' part has nothing to do with it.
Trademark only protects your 'mark' in a given 'trade' (hence the name).