Microsoft has said that Open Source is communist and Anti-American! How can the business community survive, now that their broken algorithms have been published? We're doooooooomed!
You're absolutely right. On the other hand, people might very well say "a break in my resume will look very bad when it comes to getting a new job, I'll spend my time on open source rather than watch TV".
It won't speed up the hard drive, but it should reduce the latency of a context switch (something like 21 microseconds, isn't it?) and it should also reduce the latency involved in going through the various layers of the kernel.
Yes, this isn't much in comparison to the speed of the drive, but that's not the point. I didn't say it would speed it up by a lot, merely that it would speed up.
I don't know what the latency is within the kernel in the VFS layer or within the different filesystems (ignoring mechanical delays whether from reading the data or any metadata needed due to the FS algorithm), but I can be certain it won't be zero. I can also be certain that much of this latency won't be synchronized with the disk spinning, so it's not going to vanish in a spout of parallelism. Although I can't see any reason why this would be impossible if the FS and hard drive were designed in tandem. That's not the way it's usually done, though.
The practical upshot is that using ICC and getting the 8% savings in the kernel might give you a 0.0008% improvement in performance (assuming no savings via the drive cache). Not a whole lot, certainly not enough to show on any but the most sensitive of disk I/O performance gauges, but it's still a saving.
If the drive has an on-board RAM cache large enough to eliminate consideration of the mechanical components, then I/O savings would return to the more normal 8%.
I would imagine that it means for the kernel. We would then need to factor in how much time user applications spend in the kernel. Anything that is I/O-intensive is kernel-intensive. Anything that is malloc-intensive may be kernel-intensive if you're using a VM-based memory pool rather than a pre-allocated one.
I'm also wondering how this would compare to using Cilk++ and #defining the few keywords it has to the standard keywords when using vanilla GCC or ICC.
Perhaps there should be a table showing the relative performance of the different kernel subsystems under different compilation methods.
They also promised they could make people fly. I'm wondering if this has anything to do with the UK Government saying Open Source is good. There might not be too many departments willing to switch if it might become a hot potato and it would seem that UK Government Windows contracts must equate to a hell of a lot more revenue than lost car navigation system sales.
Saying they are planning something != they actually are. If DNSSEC isn't something that will generate revenue, expect "unforeseen delays" to cause the roll-out to be postponed indefinitely - or scrapped entirely. Press releases cost very little, and by the time 2011 rolls around, nobody will remember them anyway.
The particular case is unlikely to ever be resolved to anyone's satisfaction, but the concept of having a blogger head an investigation is an interesting one. I'm not sure it's necessarily a good idea - a blogger lacks the contacts and insider information to be effective, but having a blogger somewhere on the team should improve communications.
This is only my opinion, but I'd wonder about the notion of splitting any investigating team into three sub-groups - the "regulars"/insiders, the independent experts, and the bloggers. The experts would have the means to call the bluff of the insiders (or, if it's a sincere investigation, provide better guidance than the insiders would have on their own) and the bloggers would have the means to communicate effectively.
My reasoning goes something like this. In the US and the UK, it's well-known that standard committees either start off corrupt or end up corrupt, that there's just too much pressure that doesn't get seen and never has to be answered for. Experts who, unlike bloggers, can actually be highly constructive within the committee are no less subject to pressure because their real identities will be well-known.
Bloggers, especially if they take on more of a jury role with a juror's anonymity within the system, would be anonymous to those outside pressures and therefore would make an ideal safety valve. It's hard to coerce people you don't know.
That depends. If the co-workers involved were of suitable age, gender and looks, and in need of an office fling, sending such an e-mail might have very worthwhile results.
No, all it proves is that he's not a person. This leaves all advanced alien lifeforms, artificial intelligences and supernatural phenomena as possibilities.
I think that was the article I submitted, expressly with the intent that it would catch the attention of people like yourself who could contribute to the auditing. This improves the quality of the submissions, perhaps identifies flaws in algorithms, and in general leads to a better contest between better competing implementations.
Personally, I'm gloating a little because the functions I considered to have such cool names (eg: Blue Midnight Wish) all came through clean and are also listed on the SHA3 Zoo as clean - for now. On the other hand, the ones listed on one or both sites as flawed have boring names.
I picture a "space cleaner" as using a method analogous to the shielding on the Giotto probe for Halley's Comet. That used alternating layers of aluminium foil and kevlar to vaporize incoming particles. It was knocked off-course by something the size of a pea, but it survived.
Now, imagine something built along the same sort of design but with more layers and where the only steering requirement is that it stays in orbit. You might even be able to get away with a simple ion drive for that.
This vehicle would not "sweep" debris out the way, into the atmosphere, or into some sort of container. Rather, it would follow a much simpler principle of ramming into the junk. As long as the shielding holds up, the debris will either be destroyed or contained within the kevlar.
This might work, because you only have to care about the very small debris. The large fragments can be detected and avoided. The fragments too small to be detected are the dangerous pieces of space junk, and are also the most numerous.
Such a vehicle might even be reusable, as you can send a shuttle up with a spare fuel cylinder for the ion drive and some fresh shielding.
If this turned out to be viable, you could then adapt the approach to eliminate larger space junk - simply lob explosives up into orbit to reduce the large chunks into many small fragments. So long as you always cleaned up more fragments than you generated, you'd remain ahead of the game and could clean orbit quite quickly.
Maybe you're looking in the wrong aisle in the supermarket. Or maybe you need to let yourself be inspired by the story of Mary Anning, who had discovered two new species of dinosaur by age 20. There are well-known rich fossil beds in many countries, who knows what you might find.
Hell, new species of insect are sometimes discovered by people visiting their local fossil store and buying a piece of amber. It's not as impressive as a mammoth, sure, but it's still spectacular. So, yes, although I wasn't thinking in those terms (which you know), I am highly optimistic that one of Slashdot's own (perhaps you) WILL make an important discovery in fossils.
Anything Mary Anning could do, you can do. Sure, more species are known now, but more fossil beds are known now and you have far superior mobility and technology at your disposal.
I shall expect a detailed report of the finding of Spineboyasaurus in, oh, shall we say the next couple of months?
A MOSIX or Kerrighed cluster might very well be a worthwhile thing to set up. And one that size could look good on a resume. (If you regard each CPU as a node, you'd be responsible for configuring and administrating a 68-node high-performance system.)
What would you run on such a cluster? Well, once you determine how fast it is, you can split the cluster into N virtual clusters, and rent out the compute cycles to smaller colleges, clubs and societies that need CPU power, and so on.
Ok, you can only handle about 4 customers before the cluster would no longer be powerful enough to be worth it, but that's four people paying you for stuff that otherwise you're essentially paying (in terms of space, not cash) to store.
So long as the income exceeds the cost of the electricity + net connection + self-employment tax, over the course of the year, you've profited directly.
If you even just get one or two short-term contracts for virtual clustering, that's again one hell of an item on your resume that very few others will have, which is likely to catch the interest of others.
Catastrophic collision? Hmmm. Actually, that probably is about right. Even the ISS is pretty much guaranteed to be destroyed in a catastrophic collision within the next 5-10 years, and that's in a relatively safe place.
*begs forgiveness for confusing Australia and New Zealand. I know how much the two countries hate it, and I really don't want the All Blacks scrunching me up and using me for a practice rugger ball in retaliation*
Slashdot Mirror
Microsoft has said that Open Source is communist and Anti-American! How can the business community survive, now that their broken algorithms have been published? We're doooooooomed!
The obvious next question is, can she use daisy-chaining techniques to turn the flowers into a beowulf cluster for the use of Pirate Bay?
You're absolutely right. On the other hand, people might very well say "a break in my resume will look very bad when it comes to getting a new job, I'll spend my time on open source rather than watch TV".
It won't speed up the hard drive, but it should reduce the latency of a context switch (something like 21 microseconds, isn't it?) and it should also reduce the latency involved in going through the various layers of the kernel.
Yes, this isn't much in comparison to the speed of the drive, but that's not the point. I didn't say it would speed it up by a lot, merely that it would speed up.
I don't know what the latency is within the kernel in the VFS layer or within the different filesystems (ignoring mechanical delays whether from reading the data or any metadata needed due to the FS algorithm), but I can be certain it won't be zero. I can also be certain that much of this latency won't be synchronized with the disk spinning, so it's not going to vanish in a spout of parallelism. Although I can't see any reason why this would be impossible if the FS and hard drive were designed in tandem. That's not the way it's usually done, though.
The practical upshot is that using ICC and getting the 8% savings in the kernel might give you a 0.0008% improvement in performance (assuming no savings via the drive cache). Not a whole lot, certainly not enough to show on any but the most sensitive of disk I/O performance gauges, but it's still a saving.
If the drive has an on-board RAM cache large enough to eliminate consideration of the mechanical components, then I/O savings would return to the more normal 8%.
I would imagine that it means for the kernel. We would then need to factor in how much time user applications spend in the kernel. Anything that is I/O-intensive is kernel-intensive. Anything that is malloc-intensive may be kernel-intensive if you're using a VM-based memory pool rather than a pre-allocated one.
I'm also wondering how this would compare to using Cilk++ and #defining the few keywords it has to the standard keywords when using vanilla GCC or ICC.
Perhaps there should be a table showing the relative performance of the different kernel subsystems under different compilation methods.
That's nothing! Now, when crocodiles swim in circles, it'll generate an electric current. Shocking, I tell you!
How many of them in congress?
Well, Microsoft's usual approach to entering a new market is to cripple someone's small company, buy out what's left and rebadge it.
They also promised they could make people fly. I'm wondering if this has anything to do with the UK Government saying Open Source is good. There might not be too many departments willing to switch if it might become a hot potato and it would seem that UK Government Windows contracts must equate to a hell of a lot more revenue than lost car navigation system sales.
Saying they are planning something != they actually are. If DNSSEC isn't something that will generate revenue, expect "unforeseen delays" to cause the roll-out to be postponed indefinitely - or scrapped entirely. Press releases cost very little, and by the time 2011 rolls around, nobody will remember them anyway.
The particular case is unlikely to ever be resolved to anyone's satisfaction, but the concept of having a blogger head an investigation is an interesting one. I'm not sure it's necessarily a good idea - a blogger lacks the contacts and insider information to be effective, but having a blogger somewhere on the team should improve communications.
This is only my opinion, but I'd wonder about the notion of splitting any investigating team into three sub-groups - the "regulars"/insiders, the independent experts, and the bloggers. The experts would have the means to call the bluff of the insiders (or, if it's a sincere investigation, provide better guidance than the insiders would have on their own) and the bloggers would have the means to communicate effectively.
My reasoning goes something like this. In the US and the UK, it's well-known that standard committees either start off corrupt or end up corrupt, that there's just too much pressure that doesn't get seen and never has to be answered for. Experts who, unlike bloggers, can actually be highly constructive within the committee are no less subject to pressure because their real identities will be well-known.
Bloggers, especially if they take on more of a jury role with a juror's anonymity within the system, would be anonymous to those outside pressures and therefore would make an ideal safety valve. It's hard to coerce people you don't know.
Just a thought.
That depends. If the co-workers involved were of suitable age, gender and looks, and in need of an office fling, sending such an e-mail might have very worthwhile results.
No, all it proves is that he's not a person. This leaves all advanced alien lifeforms, artificial intelligences and supernatural phenomena as possibilities.
I think that was the article I submitted, expressly with the intent that it would catch the attention of people like yourself who could contribute to the auditing. This improves the quality of the submissions, perhaps identifies flaws in algorithms, and in general leads to a better contest between better competing implementations.
Personally, I'm gloating a little because the functions I considered to have such cool names (eg: Blue Midnight Wish) all came through clean and are also listed on the SHA3 Zoo as clean - for now. On the other hand, the ones listed on one or both sites as flawed have boring names.
I picture a "space cleaner" as using a method analogous to the shielding on the Giotto probe for Halley's Comet. That used alternating layers of aluminium foil and kevlar to vaporize incoming particles. It was knocked off-course by something the size of a pea, but it survived.
Now, imagine something built along the same sort of design but with more layers and where the only steering requirement is that it stays in orbit. You might even be able to get away with a simple ion drive for that.
This vehicle would not "sweep" debris out the way, into the atmosphere, or into some sort of container. Rather, it would follow a much simpler principle of ramming into the junk. As long as the shielding holds up, the debris will either be destroyed or contained within the kevlar.
This might work, because you only have to care about the very small debris. The large fragments can be detected and avoided. The fragments too small to be detected are the dangerous pieces of space junk, and are also the most numerous.
Such a vehicle might even be reusable, as you can send a shuttle up with a spare fuel cylinder for the ion drive and some fresh shielding.
If this turned out to be viable, you could then adapt the approach to eliminate larger space junk - simply lob explosives up into orbit to reduce the large chunks into many small fragments. So long as you always cleaned up more fragments than you generated, you'd remain ahead of the game and could clean orbit quite quickly.
Maybe you're looking in the wrong aisle in the supermarket. Or maybe you need to let yourself be inspired by the story of Mary Anning, who had discovered two new species of dinosaur by age 20. There are well-known rich fossil beds in many countries, who knows what you might find.
Hell, new species of insect are sometimes discovered by people visiting their local fossil store and buying a piece of amber. It's not as impressive as a mammoth, sure, but it's still spectacular. So, yes, although I wasn't thinking in those terms (which you know), I am highly optimistic that one of Slashdot's own (perhaps you) WILL make an important discovery in fossils.
Anything Mary Anning could do, you can do. Sure, more species are known now, but more fossil beds are known now and you have far superior mobility and technology at your disposal.
I shall expect a detailed report of the finding of Spineboyasaurus in, oh, shall we say the next couple of months?
It depends on whether the Mammoth is European or African and what its air velocity is when hurled by a coconut.
That was covered in a story some time back on the earliest evidence of microbial life.
I think you'll find those are pygmy woolly mammoths, and that the fossil finds are of a much older subspecies.
Correction: Wow, that's a find of a lame mammoth.
I thought the largest deposit of fossils was in Washington DC.
A MOSIX or Kerrighed cluster might very well be a worthwhile thing to set up. And one that size could look good on a resume. (If you regard each CPU as a node, you'd be responsible for configuring and administrating a 68-node high-performance system.)
What would you run on such a cluster? Well, once you determine how fast it is, you can split the cluster into N virtual clusters, and rent out the compute cycles to smaller colleges, clubs and societies that need CPU power, and so on.
Ok, you can only handle about 4 customers before the cluster would no longer be powerful enough to be worth it, but that's four people paying you for stuff that otherwise you're essentially paying (in terms of space, not cash) to store.
So long as the income exceeds the cost of the electricity + net connection + self-employment tax, over the course of the year, you've profited directly.
If you even just get one or two short-term contracts for virtual clustering, that's again one hell of an item on your resume that very few others will have, which is likely to catch the interest of others.
Catastrophic collision? Hmmm. Actually, that probably is about right. Even the ISS is pretty much guaranteed to be destroyed in a catastrophic collision within the next 5-10 years, and that's in a relatively safe place.
That is a deep insult to all Neanderthal-kind. They, at least, made things that people could use (and wanted to).
*begs forgiveness for confusing Australia and New Zealand. I know how much the two countries hate it, and I really don't want the All Blacks scrunching me up and using me for a practice rugger ball in retaliation*