Regarding cheap gigabit ethernet controllers Intel has a rather new one (I think) "INTEL PRO/1000 MT DESKTOP ADAPTER". Here in Sweden it's only ~$10 more than a comparable 10/100 card.
The reason why it's harder to go fast with parallel connections compared to serial is that you need to ensure that all information arrives at the same time. If you have a 40 pin bus then at the other end you must be sure that all 40 pins are stable before you "grab" the data from the bus. Naturally the problem becomes more and more difficult as the speed of the bus increases. (Because that mainly mean that you can't wait as long for the connections to stabalize.)
This problem is avoided with serial connections.
And as someone already pointed out, this is why Serial ATA is on the rise now.
Yes I'm sure that a lot of people are honestly trying to do their best when it comes to this type of research. But you need to realize that there is no single definition of "intelligence" neither is there any way to really measure it.
It's not like measuring the time it takes for an apple to fall to the ground. Because in psychology you don't know what the apple and ground is, and you can't measure time.
The same thing is happening in other "soft" sciences. Interesting examples in the field on anthropology can be found in the book "Lila" by Robert M Pirzig. (Same guy who wrote Zen and the Art of Motorcycle Maintainence.)
The scientific method is a powerful tool, but like any good tool you might try to apply it to cases where it doesn't apply.
And it should perhaps be mentioned that I'm not knowledgeable about all that much research in the area. I have read some about cognition and that's really interesting IMHO. The researchers there are generally not stupid/brave enough to say that they now have found "intelligence" they tend to discuss more specific problems. Like what happens in your brain when you see different objects, or read different words. Then the scientific method is valid.
And I have yet to take an IQ test which I afterwords felt actually did a good job of exploring my intelligence. (Yeah, it's just because I do poorly, har har.;-)
Yes! Although I would have wanted to do away with the entire paper handling procedure this is what software voting should be.
Most people seem to be stuck into doing software voting the same as paper voting. It would seem that a lot of examples have proven that this is often not a good choice. If you go software/electronic why not use that to give the voters options they have not had before?
The way I though of it would be that every vote would be hashed so that afterwards I could (more or less) download all votes in my district and check if my vote was in that bundle. I could also verify the result/myself/.
If you want to do software voting then go all the way. Not some half assed attempt at a "paperless voting".
If you start a project "for fun" then focus groups are overkill. Now if someone uses your program and finds a missing feature and submit the patch then you have the "focus group" and coder in one person. That's the basic idea with OSS.
You as a maintainer don't have to include it, but most project maintainers have too little time already so I recon they'd appreciate some help.
If you want to move it between two systems a lot then I'd agree with you. But the original poster used the "external drive" as an argument for making it easier to upgrade your system.
If I'm going to spend a few hours to put a new system together I might as well spend 5 minutes moving the drive as well.
But just as he points out in the introduction this is the type of work a compiler should do. And a compiler optimized for a given architecture should be able to maximize the loop unrolling benefit. (W.r.t. caches and such.)
Yeah, I have begun to regularly take a nap during the afternoon. (Preferably after eating.) About an hour or so is quite fine. I can then stay up quite a lot longer and still be functional the next day. (This isn't a "I have to cram before the exam." technique, I use it regularly.)
If you can half an hour after lunch at work is often a good way to stay productive during the afternoon as well.
It's a scaled down version of the Power4 CPU which IBM use in their mainframes. It's also been designed to be compatible with Motorolas AltiVec vector processing.
It's very likely that Apple intend to use it in future high end boxes. (They would be rather stupid not to.) Unless Motorola can one-up IBM by then.
I don't think it has all that much to do with Apples future though. Apple boxes have not really been about high performance lately, and people buy them anyways. Mostly becuase they like the OS/apps and such I assume.
And the specs are pretty much solid now I assume. Creating hardware is/very/ different from software were you can update and patch it to the last minute. They might be able to tweak the clocks a bit, but don't expect a lot more.
True. But it should be noted that the keys are hashes, so there is no way for you to know what other people are searching for. Descriptions of the protocol can be found in the paper on Freenet. (That is not the original paper, but a revised version.)
They also adress the "I don't want to have kiddyporn on my computer" in the FAQ:
The true test of someone who claims to believe in Freedom of Speech is whether they tolerate speech which they disagree with, or even find disgusting. If this is not acceptable to you, you should not run a Freenet node.
There is another thing you can do. Since content in Freenet is available as long as its popular, you can help limit the popularity of whatever information you do not like. For example, if you do not want a file to spread you should not request it and tell everyone you know not to request that specific key.
There has been attacks suggested though. E.g. using the "time to live" variable in order to probe a specific node for what data it stores. The same technique could be done to probe your local store. I'm not sure if these issues have been adressed yet.
The point is that the data on your disc in encrypted. Neither you nor the authorities are going to be able to actually find out which specific files (or parts of files) Freenet has stored on your hdd.
They're not the only ones making high quality handwriting recognition software. I know that a smallish company around here (Decuma) has done some major progress in the area as well. Last I heard they got a lot of good press in eg Japan.
Now this is not free software, and it's made for PocketPC as of now. But apparently there's contenders out there who are on the same level as Microsoft.
What I'd like to see is software like this ported to Linux. Even if I have to pay for it at least it would then give me an alternative. I no longer would have to chose between Windows and good hand writing or Linux and crappy hand writing. A tablet PC is one of the things I/really/ want, but I don't want one that is crippled.
With asynch logic, each stage can operate independently. I see a MAJOR boon in ALU performance - Adds/subtracts/etc. take up FAR less propagation time than multiplies/divides - but in synch logic the ALU has to operate at the speed of the slowest instruction.
Yes, that's why modern CPU architectures are super-scalar. I.e. they have multiple ALU, often two for add, sub (and other fast ops) one for mul, one for div and yet more for floating point ops. Sometimes (Like with the P4.) the ALU units are clocked faster than the rest of the chip to ensure that they finish in time. (The P4 clocks at least the add sub ALU's at double clock rate.) And some units, like FP, mul and div typically take multiple clock cycles. This is legal for a syncronous chip. But all states have to be updated before next clock tick, otherwise you're screwed.
For more info I'd recommend you to browse the Ars Technica articles on CPU tech. Or to get some Patterson&Hennesey books. (Or just take courses in computer architecture.)
And just to point out one thing. There is/no/ way that you're going to create an entirely async CPU. Not with todays tools.
Awww what a cute little attempt to flame. If you try really hard or a few years you might get good enough for me to actually spend some time flaming back./Condescending pat on head.
Have you bought new hardware the last few years? The majority I have bought used this style. A small "getting started" paper for how to make it work and then the full stuff on CD.
The reason for the name apt-cache is pretty clear. You are searching the locally cached package lists.
And it's only if you want to use the command line that you'll need to use it. (In other words, most beginning Debian users will not use it.)
But one thing I've long missed in Debian is a quick way to search for packages online and just marking those I want and sending it to apt-get. I'm sure I could make a script for it, but it would be a nice thing to have added.
Regarding SMP the interview state that the compressed caching isn't compatible with SMP. (At least not now. No details are given.)
And regarding network protocols: It depends on which level you are building. On the application level a simple and slightly verbose protocol is nice from a debugging and future safe perspective. (Think HTTP.) When designing protocols for the link level it's not as much use though. (Most likely you'll just simulate that in any case. "Hot debugging" isn't all that much fun there.)
How about reading the FAQ before you start giving out "facts"? Slashdot is running on: * 5 load balanced Web servers dedicated to pages * 3 load balanced Web servers dedicated to images * 1 SQL server * 1 NFS Server Either the "little 4 way intel" you mention has a serious case of shizofrenia or your just full of it. (Guess which theory I'm going for.)
Besides the poster mentioned that those sites/are/ bigger than Slashdot. E.g. the mention that "Getting your URL posted during Friends" is nothing like getting it posted on Slashdot.
I know I shouldn't feed the trolls, but someone might actually belive this tripe.
Really? There was an earlier discussion on this topic. (Related to 9/11 or some other day with extremely high traffic.)
From that discussion I got the impression that what happens when you are bumped to the front page is that you have tried to access a story with non-standard setup. (What you get if you are logged in and change your view preferences.) The system is setup so that some servers only serve static content. (Because that's what most users view.)
During high load situations a dynamic request is sometimes sent to a static serving server. This is when you are bumped to the front page. (Unfortunately I couldn't find anything about this in the FAQ/About, so I can't verify it.)
Slashdot Mirror
Regarding cheap gigabit ethernet controllers Intel has a rather new one (I think) "INTEL PRO/1000 MT DESKTOP ADAPTER". Here in Sweden it's only ~$10 more than a comparable 10/100 card.
The reason why it's harder to go fast with parallel connections compared to serial is that you need to ensure that all information arrives at the same time. If you have a 40 pin bus then at the other end you must be sure that all 40 pins are stable before you "grab" the data from the bus. Naturally the problem becomes more and more difficult as the speed of the bus increases. (Because that mainly mean that you can't wait as long for the connections to stabalize.)
This problem is avoided with serial connections.
And as someone already pointed out, this is why Serial ATA is on the rise now.
Yes I'm sure that a lot of people are honestly trying to do their best when it comes to this type of research. But you need to realize that there is no single definition of "intelligence" neither is there any way to really measure it.
;-)
It's not like measuring the time it takes for an apple to fall to the ground. Because in psychology you don't know what the apple and ground is, and you can't measure time.
The same thing is happening in other "soft" sciences. Interesting examples in the field on anthropology can be found in the book "Lila" by Robert M Pirzig. (Same guy who wrote Zen and the Art of Motorcycle Maintainence.)
The scientific method is a powerful tool, but like any good tool you might try to apply it to cases where it doesn't apply.
And it should perhaps be mentioned that I'm not knowledgeable about all that much research in the area. I have read some about cognition and that's really interesting IMHO. The researchers there are generally not stupid/brave enough to say that they now have found "intelligence" they tend to discuss more specific problems. Like what happens in your brain when you see different objects, or read different words. Then the scientific method is valid.
And I have yet to take an IQ test which I afterwords felt actually did a good job of exploring my intelligence. (Yeah, it's just because I do poorly, har har.
Yes! Although I would have wanted to do away with the entire paper handling procedure this is what software voting should be.
/myself/.
Most people seem to be stuck into doing software voting the same as paper voting. It would seem that a lot of examples have proven that this is often not a good choice. If you go software/electronic why not use that to give the voters options they have not had before?
The way I though of it would be that every vote would be hashed so that afterwards I could (more or less) download all votes in my district and check if my vote was in that bundle. I could also verify the result
If you want to do software voting then go all the way. Not some half assed attempt at a "paperless voting".
Good job at not getting his point.
If you start a project "for fun" then focus groups are overkill. Now if someone uses your program and finds a missing feature and submit the patch then you have the "focus group" and coder in one person. That's the basic idea with OSS.
You as a maintainer don't have to include it, but most project maintainers have too little time already so I recon they'd appreciate some help.
To quote Cartman: "You kow, you really should get that sand out of your vagina. It's making you all cranky."
I guess an even better way to spend your time than discussing world events would be to troll on Slashdot. I bet your mom is really proud of you now.
If you want to move it between two systems a lot then I'd agree with you. But the original poster used the "external drive" as an argument for making it easier to upgrade your system.
If I'm going to spend a few hours to put a new system together I might as well spend 5 minutes moving the drive as well.
Yeah, that is pretty cute.
But just as he points out in the introduction this is the type of work a compiler should do. And a compiler optimized for a given architecture should be able to maximize the loop unrolling benefit. (W.r.t. caches and such.)
Yeah, I have begun to regularly take a nap during the afternoon. (Preferably after eating.) About an hour or so is quite fine. I can then stay up quite a lot longer and still be functional the next day. (This isn't a "I have to cram before the exam." technique, I use it regularly.)
If you can half an hour after lunch at work is often a good way to stay productive during the afternoon as well.
It's a scaled down version of the Power4 CPU which IBM use in their mainframes. It's also been designed to be compatible with Motorolas AltiVec vector processing.
/very/ different from software were you can update and patch it to the last minute. They might be able to tweak the clocks a bit, but don't expect a lot more.
It's very likely that Apple intend to use it in future high end boxes. (They would be rather stupid not to.) Unless Motorola can one-up IBM by then.
I don't think it has all that much to do with Apples future though. Apple boxes have not really been about high performance lately, and people buy them anyways. Mostly becuase they like the OS/apps and such I assume.
And the specs are pretty much solid now I assume. Creating hardware is
What, the founders of Paypal have gone to Norway now?
They also adress the "I don't want to have kiddyporn on my computer" in the FAQ:
There has been attacks suggested though. E.g. using the "time to live" variable in order to probe a specific node for what data it stores. The same technique could be done to probe your local store. I'm not sure if these issues have been adressed yet.
That's correct. There are OTOH projects which do aim to preserve data. E.g. Eternity Service use distributed servers to do it.
The point is that the data on your disc in encrypted. Neither you nor the authorities are going to be able to actually find out which specific files (or parts of files) Freenet has stored on your hdd.
They're not the only ones making high quality handwriting recognition software. I know that a smallish company around here (Decuma) has done some major progress in the area as well. Last I heard they got a lot of good press in eg Japan.
/really/ want, but I don't want one that is crippled.
Now this is not free software, and it's made for PocketPC as of now. But apparently there's contenders out there who are on the same level as Microsoft.
What I'd like to see is software like this ported to Linux. Even if I have to pay for it at least it would then give me an alternative. I no longer would have to chose between Windows and good hand writing or Linux and crappy hand writing. A tablet PC is one of the things I
Yes, that's why modern CPU architectures are super-scalar. I.e. they have multiple ALU, often two for add, sub (and other fast ops) one for mul, one for div and yet more for floating point ops. Sometimes (Like with the P4.) the ALU units are clocked faster than the rest of the chip to ensure that they finish in time. (The P4 clocks at least the add sub ALU's at double clock rate.) And some units, like FP, mul and div typically take multiple clock cycles. This is legal for a syncronous chip. But all states have to be updated before next clock tick, otherwise you're screwed.
For more info I'd recommend you to browse the Ars Technica articles on CPU tech. Or to get some Patterson&Hennesey books. (Or just take courses in computer architecture.)
And just to point out one thing. There is
Now this has got to be a troll. At least I hope to never have a machine control my life/death and have it being coded by hand i asm.
;-)
Please just slip a a cyanide capsule. At least then it's over fast.
Awww what a cute little attempt to flame. If you try really hard or a few years you might get good enough for me to actually spend some time flaming back. /Condescending pat on head.
Now go and play with the other little trolls.
Have you bought new hardware the last few years? The majority I have bought used this style. A small "getting started" paper for how to make it work and then the full stuff on CD.
Because it's a composition of several factors. First off 1x is about 3 times faster than 1x CD. (Ie the disc spins 3 times faster.)
Second the data is more densely packed on the disc, this contributes another factor of 3.
Total ~ 9 times faster.
The reason for the name apt-cache is pretty clear. You are searching the locally cached package lists.
And it's only if you want to use the command line that you'll need to use it. (In other words, most beginning Debian users will not use it.)
But one thing I've long missed in Debian is a quick way to search for packages online and just marking those I want and sending it to apt-get. I'm sure I could make a script for it, but it would be a nice thing to have added.
Regarding SMP the interview state that the compressed caching isn't compatible with SMP. (At least not now. No details are given.)
And regarding network protocols: It depends on which level you are building. On the application level a simple and slightly verbose protocol is nice from a debugging and future safe perspective. (Think HTTP.) When designing protocols for the link level it's not as much use though. (Most likely you'll just simulate that in any case. "Hot debugging" isn't all that much fun there.)
How about reading the FAQ before you start giving out "facts"? Slashdot is running on:
/are/ bigger than Slashdot. E.g. the mention that "Getting your URL posted during Friends" is nothing like getting it posted on Slashdot.
* 5 load balanced Web servers dedicated to pages
* 3 load balanced Web servers dedicated to images
* 1 SQL server
* 1 NFS Server
Either the "little 4 way intel" you mention has a serious case of shizofrenia or your just full of it. (Guess which theory I'm going for.)
Besides the poster mentioned that those sites
I know I shouldn't feed the trolls, but someone might actually belive this tripe.
Really? There was an earlier discussion on this topic. (Related to 9/11 or some other day with extremely high traffic.)
From that discussion I got the impression that what happens when you are bumped to the front page is that you have tried to access a story with non-standard setup. (What you get if you are logged in and change your view preferences.) The system is setup so that some servers only serve static content. (Because that's what most users view.)
During high load situations a dynamic request is sometimes sent to a static serving server. This is when you are bumped to the front page. (Unfortunately I couldn't find anything about this in the FAQ/About, so I can't verify it.)
Unfortuantely GCC is really poor at optimizing in any case, so it's pretty much moot.
Just compare GCC at max optimization and SUN cc at minimum or first level. SUN cc beats GCC even there.
So while it's true that more data -> better optimization is correct I don't think it's a very pressing issue for GCC front ends.