Yes, for soldering a third and forth hand would be magnificent. No more operating a pair of pliers between the little finger and the inside of your palm.
If they have figured out a good way to interface with the nervous system, why complicate things by having a hand between the brain and the code? Imagine the typing speed with a USB cord plugged straight into the arm instead.
never have I seen a software company where experts were valued above salesmen. When a salesman makes a big contract, it's like he is the king of the world. Whole company has to kneel before him (just a metaphor).
I think this is because it is so much easier for management to connect the incomes of a particular sale to the salesman, than the much more indirect effect of skilled engineers. This is how they see it: Take away an expert and not much appears to happen, take away the saleman and you wouldn't have that million dollar deal. Obviously the saleman is worth more to the company.
The problem is, of course, that the reason the saleman was able to make the deal, was because the engineers had been working on the product over a long period of time. But that is much less conspicuous.
I think we are talking past each other... I have never said that people won't pay for on-line journals. All I have said is that if there is no reason to pay for something, then there is no reason for it to exist. The question by the Royal Society ("Why would you pay to subscribe to a journal if the papers appear free of charge [elsewhere]?") seemed to stipulate that printed journals have an intrinsic value which must be protected by artificial "trade" barriers that force people to pay for them.
The fact that many, as you say, will pay for on-line equivalents of the service even though it is available for free only makes the reasoning of the Royal Society even more silly.
I think you might be missing my point slightly. I also pay for add-free stuff and donate money for good causes (although I'm not a slashdot subscriber), but I think this is different; let me explain.
Maybe I misunderstood this whole issue, but wasn't the complaint that researchers publish their research papers themselves for free online? Since the journals usually aren't paying the researchers money for the papers anyway, I really can't see a problem with this. If it means that noone will buy journals, then they are meaningless (as opposed to good causes and add-free comics). If, on the other hand, the journals provide additional value such as peer review, that the original authors can't when they publish online, then people will continue to buy journals -- the product one would be paying for would then actually be the peer review rather than the content.
Why would you pay to subscribe to a journal if the papers appear free of charge?
Actually, that statement is silly whichever way you turn it. Either journals give you some value (peer-review?) that publication via the Internet cannot, in which case that is why I pay to subscribe, or they are in every way equivalent to Internet publication, in which case the question becomes rhetorical: I wouldn't pay.
Hehe, no I am a computer scientist. I have studied some university level plasma physics and a small amount of quantum mechanics though, but I would far from call myself a physicist.
Nah, at least he was within the correct order of magnitude. Probably a sufficiently good approximation for this discussion, especially considering the accuracy of the ingoing parameters. It all depends on the purpose of the calculation. A physicist knows that.
Where were the parents? Slacking off? Don't say that parents can't keep an eye on their children all the time becuase that is bs. Both my parents worked and they still managed to keep me from inserting things up my nose and jumping out of windows.
The fact that you and all of us managed to survive our childhood proves absolutely nothing. If you had not you wouldn't be here to make that comment.
Kids don't immediately go and commit suicide the first second their parents look away. Kids occasionally do stupid things when their parents aren't around, however, but in 99.9% of the cases it turns out OK, and nobody blames the parents. If you take a large enough population, there will always be cases where kids kill themselves, but it doesn't necessarily mean the parents were neglectful (not more than the average parent anyway).
I find that with each waking moment advertising is getting more invasive and more offensive. It needs to stop. But I don't think I should have to pay to make it stop.
In fact, we are actually paying for them to be there, through the products that are advertised.
I am referring to (what is exposed as) system memory vs. disk related memory. From a software development point of view, we have these two logical memories, which are implemented using multiple physical memory types; as you were saying: [CPU register -- L1 cache -- L2 cache -- (L3 cache possibly) -- system memory] vs. [(drive controller cache possibly) -- disk cache -- disk].
In my opinion, there should be only one logical memory. Since I'm no hardware or kernel developer I don't care too much about exactly how it is implemented, but I can see that the only persistent memory type (i.e. hard drive) is too slow for all of these to be merged into one single coherent logical memory efficiently.
A return to core memory would be nice if it is possible. I can't help feeling like the whole idea of secondary (i.e. disk) memory with files and file systems is an ugly hack.
Yes, obviously you can implement all kinds of persistent things with ordinary hard drives. And people do. However, what I am talking about is orthogonal persistence. The programmer shouldn't have to implement code to handle persistence in every single application. It should be built into the system at a much lower level, so that the programmer only sees one coherent block of memory. The RAM would be just as invisible as the CPU cache is today. Then everything in the entire system would be "saved" automatically, including the carret position, window locations etc. etc. And that would require faster persistent memory technology.
I agree. Compared to RAM, hard drives are insanely slow. If there were a reasonably large, fast persistent storage system we could finally get rid of the explicit barrier between primary and secondary memory and have one single transparent memory hierarchy, from registers in the CPU down to whatever replaces hard drives. Persistence would be orthogonal and boot/load times close to zero. Users wouldn't have to be exposed to the concept of in-memory copies of data, loading and saving of files and documents -- if you change a document it remains changed, just as a real world document would.
Well, perhaps I should have said "more discontent and angry" instead...;)
But the sad part is that most (non-tech savvy) people haven't really been angry with MS for all the BSoDs. Somehow, the general public seem to have adopted the notion that computers are "supposed" to work as badly as Windows has done over the years. Probably because Windows is all they have ever seen, in terms of operating systems.
So you're talking about replacing 480K jobs, with 20K jobs. Sounds great to me, they just have to work 24 times as hard.
Only if what those 480k developers are doing is strictly disjunct. In reality, every wheel is reinvented hundreds of times, so that is hardly the case.
In fact, I suspect there are a lot of people out there that thinks "Only 80 megs? That software can't possibly be as good as MS Office with it's mighty 1.2 gigs of features". It's the same thing with prices -- there are several examples where increasing the price of a product actually have had a positive impact on sales. It's silly ofcourse, but people aren't generally very rational.
If you assume that the failures are spread out evenly across time, a 1000-disk system will have a failure every 500 hours, or about every 3 weeks!
For the sake of your argument I suppose that assumption could be considered fair. If one were to do a somewhat more sophisticated analysis, a better model for hard drive failures is the Bathtub curve. It represents the result of a combination of three types of failures: infant mortality (flaws in the manufacturing), random failures and wear-out failures.
The failure rates, in fact, will likely be fractal -- you'll have long periods without failures, or with few failures, and then a bunch of failures will occur in a short period of time, seemingly all at once.
I think what you are referring to is how multiple observations of a uniformly distributed stochastic variable generally look. It doesn't have anything to do with fractals, though.
Slashdot Mirror
The site is not responding (big surprise). But from the description it sounds like it is similar to Music-Map.
Yes, for soldering a third and forth hand would be magnificent. No more operating a pair of pliers between the little finger and the inside of your palm.
If they have figured out a good way to interface with the nervous system, why complicate things by having a hand between the brain and the code? Imagine the typing speed with a USB cord plugged straight into the arm instead.
Maybe they should market it as a free heater instead.
"Buy the new Xbox 360 now and get a stylish living room heater at no additional cost!"
I think this is because it is so much easier for management to connect the incomes of a particular sale to the salesman, than the much more indirect effect of skilled engineers. This is how they see it: Take away an expert and not much appears to happen, take away the saleman and you wouldn't have that million dollar deal. Obviously the saleman is worth more to the company.
The problem is, of course, that the reason the saleman was able to make the deal, was because the engineers had been working on the product over a long period of time. But that is much less conspicuous.
The fact that many, as you say, will pay for on-line equivalents of the service even though it is available for free only makes the reasoning of the Royal Society even more silly.
So basically I think we agree.
Maybe I misunderstood this whole issue, but wasn't the complaint that researchers publish their research papers themselves for free online? Since the journals usually aren't paying the researchers money for the papers anyway, I really can't see a problem with this. If it means that noone will buy journals, then they are meaningless (as opposed to good causes and add-free comics). If, on the other hand, the journals provide additional value such as peer review, that the original authors can't when they publish online, then people will continue to buy journals -- the product one would be paying for would then actually be the peer review rather than the content.
In Sweden, we have land rise (since the last ice age) to compensate for the sea rise, and we already are further north. Ha! ;)
Actually, that statement is silly whichever way you turn it. Either journals give you some value (peer-review?) that publication via the Internet cannot, in which case that is why I pay to subscribe, or they are in every way equivalent to Internet publication, in which case the question becomes rhetorical: I wouldn't pay.
Yes, they will probably hide a rootkit in their own OS and hardware. That makes sense.
Hehe, no I am a computer scientist. I have studied some university level plasma physics and a small amount of quantum mechanics though, but I would far from call myself a physicist.
He didn't calculate the volume. The calculation was 1MW / (40*40*6) ft^2 which is about 104 W/ft^2.
Nah, at least he was within the correct order of magnitude. Probably a sufficiently good approximation for this discussion, especially considering the accuracy of the ingoing parameters. It all depends on the purpose of the calculation. A physicist knows that.
The fact that you and all of us managed to survive our childhood proves absolutely nothing. If you had not you wouldn't be here to make that comment.
Kids don't immediately go and commit suicide the first second their parents look away. Kids occasionally do stupid things when their parents aren't around, however, but in 99.9% of the cases it turns out OK, and nobody blames the parents. If you take a large enough population, there will always be cases where kids kill themselves, but it doesn't necessarily mean the parents were neglectful (not more than the average parent anyway).
In fact, we are actually paying for them to be there, through the products that are advertised.
In my opinion, there should be only one logical memory. Since I'm no hardware or kernel developer I don't care too much about exactly how it is implemented, but I can see that the only persistent memory type (i.e. hard drive) is too slow for all of these to be merged into one single coherent logical memory efficiently.
A return to core memory would be nice if it is possible. I can't help feeling like the whole idea of secondary (i.e. disk) memory with files and file systems is an ugly hack.
Yes, obviously you can implement all kinds of persistent things with ordinary hard drives. And people do. However, what I am talking about is orthogonal persistence. The programmer shouldn't have to implement code to handle persistence in every single application. It should be built into the system at a much lower level, so that the programmer only sees one coherent block of memory. The RAM would be just as invisible as the CPU cache is today. Then everything in the entire system would be "saved" automatically, including the carret position, window locations etc. etc. And that would require faster persistent memory technology.
I agree. Compared to RAM, hard drives are insanely slow. If there were a reasonably large, fast persistent storage system we could finally get rid of the explicit barrier between primary and secondary memory and have one single transparent memory hierarchy, from registers in the CPU down to whatever replaces hard drives. Persistence would be orthogonal and boot/load times close to zero. Users wouldn't have to be exposed to the concept of in-memory copies of data, loading and saving of files and documents -- if you change a document it remains changed, just as a real world document would.
But the sad part is that most (non-tech savvy) people haven't really been angry with MS for all the BSoDs. Somehow, the general public seem to have adopted the notion that computers are "supposed" to work as badly as Windows has done over the years. Probably because Windows is all they have ever seen, in terms of operating systems.
Because not doing so makes their customers discontent and angry?
Did anyone else read that as Uninterruptible Power Supply?
I actually pondered for a brief second on what a "down" was...
Only if what those 480k developers are doing is strictly disjunct. In reality, every wheel is reinvented hundreds of times, so that is hardly the case.
In fact, I suspect there are a lot of people out there that thinks "Only 80 megs? That software can't possibly be as good as MS Office with it's mighty 1.2 gigs of features". It's the same thing with prices -- there are several examples where increasing the price of a product actually have had a positive impact on sales. It's silly ofcourse, but people aren't generally very rational.
For the sake of your argument I suppose that assumption could be considered fair. If one were to do a somewhat more sophisticated analysis, a better model for hard drive failures is the Bathtub curve. It represents the result of a combination of three types of failures: infant mortality (flaws in the manufacturing), random failures and wear-out failures.
The failure rates, in fact, will likely be fractal -- you'll have long periods without failures, or with few failures, and then a bunch of failures will occur in a short period of time, seemingly all at once.
I think what you are referring to is how multiple observations of a uniformly distributed stochastic variable generally look. It doesn't have anything to do with fractals, though.