The telcos used lead-acid batteries as a filter for the line in addition to back-up. If you have an analog signal riding on top of a constant current source like with POTS, you need an *extremely* stable DC level or else people will hear buzzing. For a very long time (and maybe even now), the cheapest and easiest way to do this was to run the whole system off batteries.
So I think this could be classified as "good business sense", though for reasons completely unrelated to reliability.
It should be noted that this work is purely theoretical. What they have done is show that there is a much more physically realizable way to way a repulsive Casimir effect than the previous schemes, using a material with negative refraction over some range of important frequencies (this is a similar problem as making a cloaking device, but with a harder range of the spectrum). In practice, the effect would be small and the material hard to make, but the idea is interesting.
This actually makes a very good point. Some arguably secure coding styles (microkernels, for instance) involve a fantastic number of syscalls, as operations trampoline through kernel space.
On the other end of things, the way to get the fewest possible number of syscalls is to implement the entire web server in the kernel (in a single function, as the OP wrote). Then you just call the handle_http_request() syscall and walk away. This is, of course, the least secure and most dangerous possible way to implement a web server.
The only thing with which number of system calls actually correlates is request handling speed -- barring other performance issues, context switches take some amount of time, which is why microkernels typically have poor performance. Given the massively different software architectures involved, however, I would imagine that any important performance differences lie elsewhere.
Sun is moving ahead with their SPARC servers, and just taped out a successor to the Niagara. If you'd read the article, you'd know they are replacing their (quite excellent) AMD servers with Intel ones, not SPARC with anything. Sun has quite happily been selling both architectures for some time now.
It was never about performance per se -- there are plenty of faster things out there than the Core 2 Duo. IBM will be happy to sell you some of them, as will Sun or Fujitsu. Or Cray. All for the low price of $600k a machine.
The issue is that IBM makes supercomputers, and Motorola makes cellphones, and they design their chips accordingly. Apple, making neither of these things, couldn't persuade either of them to make a low-power, fast, cheap CPU useful for a laptop and continue updating it with such a small market. Intel, on the other hand, spends most of their engineering effort trying to solve exactly this problem, and so has its business interests aligned with Apple's, as opposed to IBM or Motorola, who didn't really care about them at all, and would happily spend their R&D money on designing things like this chip instead of making a G5 that would fit in a laptop.
Slashdot Mirror
The telcos used lead-acid batteries as a filter for the line in addition to back-up. If you have an analog signal riding on top of a constant current source like with POTS, you need an *extremely* stable DC level or else people will hear buzzing. For a very long time (and maybe even now), the cheapest and easiest way to do this was to run the whole system off batteries.
So I think this could be classified as "good business sense", though for reasons completely unrelated to reliability.
Here's the paper, courtesy of arXiv:
http://www.arxiv.org/abs/quant-ph/0608115
It should be noted that this work is purely theoretical. What they have done is show that there is a much more physically realizable way to way a repulsive Casimir effect than the previous schemes, using a material with negative refraction over some range of important frequencies (this is a similar problem as making a cloaking device, but with a harder range of the spectrum). In practice, the effect would be small and the material hard to make, but the idea is interesting.
I believe the nouveau driver is actually based on the Haiku one, which has 3D support for a limited range of nVidia products (up to GeForce 4).
This actually makes a very good point. Some arguably secure coding styles (microkernels, for instance) involve a fantastic number of syscalls, as operations trampoline through kernel space.
On the other end of things, the way to get the fewest possible number of syscalls is to implement the entire web server in the kernel (in a single function, as the OP wrote). Then you just call the handle_http_request() syscall and walk away. This is, of course, the least secure and most dangerous possible way to implement a web server.
The only thing with which number of system calls actually correlates is request handling speed -- barring other performance issues, context switches take some amount of time, which is why microkernels typically have poor performance. Given the massively different software architectures involved, however, I would imagine that any important performance differences lie elsewhere.
Sun is moving ahead with their SPARC servers, and just taped out a successor to the Niagara. If you'd read the article, you'd know they are replacing their (quite excellent) AMD servers with Intel ones, not SPARC with anything. Sun has quite happily been selling both architectures for some time now.
It was never about performance per se -- there are plenty of faster things out there than the Core 2 Duo. IBM will be happy to sell you some of them, as will Sun or Fujitsu. Or Cray. All for the low price of $600k a machine.
The issue is that IBM makes supercomputers, and Motorola makes cellphones, and they design their chips accordingly. Apple, making neither of these things, couldn't persuade either of them to make a low-power, fast, cheap CPU useful for a laptop and continue updating it with such a small market. Intel, on the other hand, spends most of their engineering effort trying to solve exactly this problem, and so has its business interests aligned with Apple's, as opposed to IBM or Motorola, who didn't really care about them at all, and would happily spend their R&D money on designing things like this chip instead of making a G5 that would fit in a laptop.
For those of you wanting high quality and a connection that works:
p ngt if
http://banshee.uchicago.edu/~nathanw/2230_6163_2.
http://banshee.uchicago.edu/~nathanw/2230_6163_3.