While it's true that Berman hasn't done much lately, I'd say he's the primary reason that TNG made it beyond the first couple of seasons (which IMHO were some of the worst Trek episodes ever). It was only after Berman took over completely that TNG hit the mark.
IBM would have to behave itself for many more years before they make up for a long history of cutthroat behavior.
Even today they don't really embrace openness as much as you think. For example, they bought Rational and killed Rational Visual Test, a popular software testing product because it competed with the much more expensive Rational Robot. Not only have they discountinued development and support, they won't even allow you to buy additional licences for it. If they really were OSS true believers wouldn't they release the source?
"any recompilation and chances are forced by the different architectures, not Linux."
Of course you could use that statement to wiggle out of any argument against seamless scaling; even Windows could make that claim. Your not going to convince me that Linux or Windows is seamlessly scalable.
"That's what "seamlessly scale" means when talking about an OS, and it's the best any OS is going to achieve."
Certainly it's possible that an OS could be written that is more scalable than Linux. In any case even if Linux is "the best any OS is going to achieve" in scalability, that doesn't make it seamlessly scalable.
"We are discussing the relative merits of Windows CE/XP and Linux."
No. YOU have been discussing the relative merits of Windows CE/XP. I simply questioned your claim that Linux scaled seamlessly from small systems on up. I'm not addressing Windows CE/XP simply because nobody has made the "seamlessly scale" argument for it as you have for Linux.
"With Linux, you can run an MMU-less port like ucLinux, or you can use any of a dozen other embedded operating systems with POSIX APIs."
In other words, in real-world small systems you can use a modified version of Linux or other OS's that are not derived from it. Thus Linux does not seamlessly scale from small to large systems. QED.
"It means exactly what any reasonable person would think it means: the Linux kernel runs on a small ARM board with a couple of megs of RAM, on a 16G 64bit machine, or a compute cluster with 256 nodes."
I guess when you said "small" you were referring to the footprint. A system with "a couple of megs of RAM" is not small by embedded systems standards.
"The same software, libraries, etc., can run across that entire spectrum of machines."
Many embedded systems don't have memory management or a processor with privilege levels . How can Linux run on such a system without massive changes?
I really don't understand this argument. We require users to have special-purpose software like browsers to view web pages, but we insist that using a text editor for page creation is somehow "powerful".
Perhaps we should use only paper because we can prepare content without the "fuss" of a text editor and computer.
ASCII is simply an abstraction of bits to characters. It works on a system only because that system has the software that supports it. Binary XML would be exactly the same.
"Would it have been "hundreds or thousands of tweaks" on the other OS?"
What other OS are you referring to? I'm not picking on Linux, I'm making the general point that traditional OS's are not the most effective choice for embedded systems.
I think it's redundant to say an embedded system with custom hardware. What other kind is there?
It's not generally necessary to modify an OS just to support custom hardware. Historically, most embedded OS's were proprietary and not being able to modify the OS has never been a great limitation.
Perhaps the problem is that since general purpose OS's like Linux weren't designed for embedded systems, more tweaking is required.
"In an embedded system, it's not such a big issue if you have to GPL your code, since it won't be any use without the hardware you are embedding it in."
If that's true, there's not much value to the "community" in forcing companies to release the source either. Perhaps revising the GPL accordingly would help promote OSS/Free software in embedded systems.
"A friend of mine works for a company that makes embedded systems and they chose a Linux kernel to drive it since they have to make hundreds of tweaks in the kernel code to compensate for custom hardware (that they build in-house)."
It sounds to me like your friend's company chose the wrong OS if they needed to make "hundreds of tweaks" to it.
My recommendation is to forget about Linux and Windows CE and use a OS that was designed from the ground up for embedded use.
Note that RS-422 has a higher number than RS-232. Technology improves over time and you can't judge designs across generations. RS-232 was first created in 1960 and wasn't bad for its time.
The problem is that RS232 is (was) being used for things it was not designed for. It was never meant to be a general purpose serial communications standard.
"If you do think that truth is uncorrelated with opinion, then how do you explain society's seeming to advance to hold more "true" ideas about the world over history?"
Well, look at my example again. Let's say that in the old days 95% people believed the earth was flat and 5% believed it was round. Today, the percentages are probably flipped.
So the opinion "variable" has changed by 95% and the earth's shape has not changed (more or less). Please explain the mathematical correlation.
"Could the spurning of stock options for cash send a message to the employer that this particular employee doesn't like the company's long-term perspectives"
I suspect that companies that have this narrow point of view probably won't offer a cash option since they believe it inconceivable that they could fail or that anyone wouldn't want to share in the incredible wealth that is just around the corner. In other words, a company that is going to fail.
I agree that there's no consistent relationship between profit and innovation. However, I'm not sure if Amazon is particulary innovative either.
As far as profit is concerned, it's perfectly natural for investers to be more interested in that then pure innovation or creativity. It's also natural for publications that cater to investers to take an interest in profits as well.
In general companies that aren't profitable will go out of business, and will no longer have the opportunity to be innovative.
"I am assuming here that probability of truth of a proposition is proportional to the # of people who believe in the proposition"
That assumption is the problem with your argument.
For most of man's existence, almost everyone believed that the earth was flat and that the sun circled around it. There's no fixed relationship between opinion and fact.
"For example, exclusive OEM bundle agreements are one aspect of aggressive marketing."
I'm not an expert so I can't judge if OEM bundle agreements should be classified as marketing. It sounds more like negotiation on the terms of a sale to me.
"For high-budget corporate customers, an impression of "dazzlement" can be a negative, as it signals a product meant for artists and radicals"
As far as advertising is concerned, IBM's services Ads on TV are far more interesting than anything MS has done and high-budget corporate customers are their bread-and-butter.
Slashdot Mirror
While it's true that Berman hasn't done much lately, I'd say he's the primary reason that TNG made it beyond the first couple of seasons (which IMHO were some of the worst Trek episodes ever). It was only after Berman took over completely that TNG hit the mark.
IBM would have to behave itself for many more years before they make up for a long history of cutthroat behavior.
Even today they don't really embrace openness as much as you think. For example, they bought Rational and killed Rational Visual Test, a popular software testing product because it competed with the much more expensive Rational Robot. Not only have they discountinued development and support, they won't even allow you to buy additional licences for it. If they really were OSS true believers wouldn't they release the source?
"any recompilation and chances are forced by the different architectures, not Linux."
Of course you could use that statement to wiggle out of any argument against seamless scaling; even Windows could make that claim. Your not going to convince me that Linux or Windows is seamlessly scalable.
"That's what "seamlessly scale" means when talking about an OS, and it's the best any OS is going to achieve."
Certainly it's possible that an OS could be written that is more scalable than Linux. In any case even if Linux is "the best any OS is going to achieve" in scalability, that doesn't make it seamlessly scalable.
"We are discussing the relative merits of Windows CE/XP and Linux."
No. YOU have been discussing the relative merits of Windows CE/XP. I simply questioned your claim that Linux scaled seamlessly from small systems on up. I'm not addressing Windows CE/XP simply because nobody has made the "seamlessly scale" argument for it as you have for Linux.
"With Linux, you can run an MMU-less port like ucLinux, or you can use any of a dozen other embedded operating systems with POSIX APIs."
In other words, in real-world small systems you can use a modified version of Linux or other OS's that are not derived from it. Thus Linux does not seamlessly scale from small to large systems. QED.
"It means exactly what any reasonable person would think it means: the Linux kernel runs on a small ARM board with a couple of megs of RAM, on a 16G 64bit machine, or a compute cluster with 256 nodes."
I guess when you said "small" you were referring to the footprint. A system with "a couple of megs of RAM" is not small by embedded systems standards.
"The same software, libraries, etc., can run across that entire spectrum of machines."
Many embedded systems don't have memory management or a processor with privilege levels . How can Linux run on such a system without massive changes?
Most Slasdotters like Everquest?
Yes, read the minds of potential employers. It's great for interviews too. Anything else is just a shot in the dark.
I really don't understand this argument. We require users to have special-purpose software like browsers to view web pages, but we insist that using a text editor for page creation is somehow "powerful".
Perhaps we should use only paper because we can prepare content without the "fuss" of a text editor and computer.
ASCII is simply an abstraction of bits to characters. It works on a system only because that system has the software that supports it. Binary XML would be exactly the same.
"That would make it into the BSD license."
That certainly wouldn't bother me, but I was suggesting a different treatment of embedded systems vs. non-embedded systems under the GPL.
"Would it have been "hundreds or thousands of tweaks" on the other OS?"
What other OS are you referring to? I'm not picking on Linux, I'm making the general point that traditional OS's are not the most effective choice for embedded systems.
I think it's redundant to say an embedded system with custom hardware. What other kind is there?
It's not generally necessary to modify an OS just to support custom hardware. Historically, most embedded OS's were proprietary and not being able to modify the OS has never been a great limitation.
Perhaps the problem is that since general purpose OS's like Linux weren't designed for embedded systems, more tweaking is required.
You keeping using these words: scales, seamlessly, small. I don't thing they mean, what you think they mean.
"In an embedded system, it's not such a big issue if you have to GPL your code, since it won't be any use without the hardware you are embedding it in."
If that's true, there's not much value to the "community" in forcing companies to release the source either. Perhaps revising the GPL accordingly would help promote OSS/Free software in embedded systems.
"A friend of mine works for a company that makes embedded systems and they chose a Linux kernel to drive it since they have to make hundreds of tweaks in the kernel code to compensate for custom hardware (that they build in-house)."
It sounds to me like your friend's company chose the wrong OS if they needed to make "hundreds of tweaks" to it.
My recommendation is to forget about Linux and Windows CE and use a OS that was designed from the ground up for embedded use.
It's simple really. If most Slashdotters like it, it's Unix-like, if they don't it's not.
Note that RS-422 has a higher number than RS-232. Technology improves over time and you can't judge designs across generations. RS-232 was first created in 1960 and wasn't bad for its time.
The problem is that RS232 is (was) being used for things it was not designed for. It was never meant to be a general purpose serial communications standard.
"If you do think that truth is uncorrelated with opinion, then how do you explain society's seeming to advance to hold more "true" ideas about the world over history?"
Well, look at my example again. Let's say that in the old days 95% people believed the earth was flat and 5% believed it was round. Today, the percentages are probably flipped.
So the opinion "variable" has changed by 95% and the earth's shape has not changed (more or less). Please explain the mathematical correlation.
"Could the spurning of stock options for cash send a message to the employer that this particular employee doesn't like the company's long-term perspectives"
I suspect that companies that have this narrow point of view probably won't offer a cash option since they believe it inconceivable that they could fail or that anyone wouldn't want to share in the incredible wealth that is just around the corner. In other words, a company that is going to fail.
10 minutes of content will do the job.
I agree that there's no consistent relationship between profit and innovation. However, I'm not sure if Amazon is particulary innovative either.
As far as profit is concerned, it's perfectly natural for investers to be more interested in that then pure innovation or creativity. It's also natural for publications that cater to investers to take an interest in profits as well.
In general companies that aren't profitable will go out of business, and will no longer have the opportunity to be innovative.
"I am assuming here that probability of truth of a proposition is proportional to the # of people who believe in the proposition"
That assumption is the problem with your argument.
For most of man's existence, almost everyone believed that the earth was flat and that the sun circled around it. There's no fixed relationship between opinion and fact.
Has Amazon made a profit yet? I'm not talking about profitable quarters. Has Amazon made more in revenue over its lifetime than the money invested?
"For example, exclusive OEM bundle agreements are one aspect of aggressive marketing."
I'm not an expert so I can't judge if OEM bundle agreements should be classified as marketing. It sounds more like negotiation on the terms of a sale to me.
"For high-budget corporate customers, an impression of "dazzlement" can be a negative, as it signals a product meant for artists and radicals"
As far as advertising is concerned, IBM's services Ads on TV are far more interesting than anything MS has done and high-budget corporate customers are their bread-and-butter.