> I'm not a techwiz, but wouldn't full 64-bit memory addressing affect the Tflops, > and ultimately, the cluster's performance?
No, it wouldn't. See ARStechnica.com's articles about 64bit and what it will really do for you, for example.
Re:What about those of us
on
CNet on WinFS
·
· Score: 1
delete it from what? If there's no hierarchical directory structure, all you can do is delete it from one particular index (or from all of them, if that's supported). I can imagine a few possible designs, such as a "normal" fs with directories and files, but with indexes as well, or just having indexes (like one big directory).
Re:Isn't this what Hans Reiser is doing?
on
CNet on WinFS
·
· Score: 1
MS can't legally distribute GPLed code linked into their kernel without giving away source to the whole thing. RedHat has only the GPL's restrictions on people copying what's on the CDs (i.e. you can copy and redistribute, but you have to make source available), and they give away source, thus fulfilling the requirements of the GPL.
So, in summary, the difference is that RH gives away the source, and I implicitly assumed MS won't.
> That hurts a bit, but my reaction is to say that AOL doesn't need my mail. But > what happens when ISPs start to enforce no-server limitations? What happens when > governments start to enforce them?!
Here's what John Walker thinks will happen if these trends continue. I don't like the trend toward a consumer Internet either, since I run a mail and DNS server on my cable modem, not to mention an ssh server so I can get at my stuff from other computers. My ISP has/had? a policy against _file_ servers, which is how I plan to justify running the servers I do if they ever give me any trouble.
> Eventually, this will lead to healthy competition between the "subculture nets" > and "The Internet".
That kind of competition isn't healthy. DNS inconsistencies suck. It's extra work for anyone who has to make sure they're DNS setup is ok.
> (we all know there's no such thing as The Internet, right? that it's just a > generic term that we use to refer to consumers of IPV4 address space).
The Internet is an IP internetwork. It is the biggest one around, and is really the only global one that most people are on, so we call it the Internet.
-Carbon in the atmosphere is a mix of C12 and C14. C14 is unstable, so it's always decaying, but is replaced by nuclear processes caused by cosmic rays, etc.
-Anything that gets new carbon from the atmosphere all the time, and gets rid of old carbon (e.g. living things) will have a C12:C14 ratio that depends on the atmospheric ratio. (You're right that living things slightly prefer one of isotopes over the other, and that the ratio varied over time).
-If the thing stops refreshing its carbon from the atmosphere (e.g. by dieing), the C14 will decay. You can measure the C12:C14 ratio, and given the decay rate of C14, you can calculate how long ago the carbon in your sample was cut off from the atmosphere.
If you had a way to figure out what the initial C12:C14 ratio in the water was, you could figure out when the sample had been cut off. Depending on how well aerated the water was, the C12:C14 might not be the same as the atmosphere. You wouldn't be able to tell if there had been short intervals where your puddle (or whatever) was exposed to the air, but they would throw off your calculations.
There is a radioactive decay dating process for sea water, IIRC. Oxygen, I think. There's also potassium-argon dating, which I think is useful for volcanic rocks with pockets of air to trap the argon.
Re:Not neccessarily bad
on
CNet on WinFS
·
· Score: 1
> directory filename from/home/joe where filesize > 1 MB and mtime 1 day ago > > would list all filenames for recent, larger files.
And you think this would be faster than what we have now? find/home/joe -type d -size +1024k -mtime -1 (and you can combine find's arguments with and or or logic, like with SQL. Obviously you don't have the same power as SQL with just find by itself. Traversing directories doesn't take that long, esp. if you've already done it recently so everything's in the cache.)
Re:What about those of us
on
CNet on WinFS
·
· Score: 1
> If you create multiple indexes, that comes down to having files in different > directories at the same time; keeping multiple indexes consistent (when files > are deleted) will cost overhead
Unix filesystems allow files to be linked to multiple directories at once, and they seem to do ok. (inode-based filesystems have a refcount in the inode, which is incremented and decremented as it is linked and unlinked from directories. Now you know why the system call rm(1) uses is unlink(2).). See ln(1) and link(2), and see the second column of the output of ls -l for the number of links a file has.
Unix filesystems don't provide a way to find all links to a file, which is, as you say, what you would need if you wanted to be able to delete a file entirely instead of just unlinking it from its current location. That would cost overhead. Otherwise, I wouldn't worry.
Re:Isn't this what Hans Reiser is doing?
on
CNet on WinFS
·
· Score: 1
> No, no:) They could change it slightly, in > undocumented ways, like they did with Kerberos and > Ldap for Active Directory.
Reiserfs is GPLed, so they couldn't do that to the Free version without rewriting from scratch (with the requisite incompatibilities). They could buy a license from namesys and get their own copy of the code that they could gratuitously change, though.
And BTW, the first time I saw something about WinFS, I thought it sounded like reiserfs.
Duke 3D: "It's time to kick ass and chew bubble gum... And I'm all outta guhhhm." (Big drawl on the last "gum":)
When you meet the Lunar Assault overlord at the end of mission 2 (Lunar Apocalypse), Duke says "I'm gonna rip off your head and shit down your neck". And you don't even need to imagine it, if you beat the ugly mofo (he reads the paper and whistles while shitting down the thing's neck...). A little bit too weird for regular usage, unlike the gum line, which is good for getting psyched up for a variety of sports.
The increase in virtual address space is very welcome. Even on a machine with 1GB of RAM, Linux needs highmem support to use it all. (Linux gives user space processes 3GB of virtual address space, and less than 1GB is left over for the kernel to map memory, because of PCI and AGP memory spaces). With a decent 64bit system, stupid bounce buffers won't be needed.
>... it cannot use mmap()ed access to those files in a 32 bit processor...
If you look at it another way, once 64bit platforms become widespread, programs will start assuming they can just mmap anything they want. That will be a Good Thing, since sometimes mmap()ing is a more efficient or nicer way to do things.
I'm not sure whether mmap() is a higher performance way to do things for file/web servers, though. The TLB misses from mapping that much stuff might hurt... For Linux-optimized stuff, you can use sendfile(2).
It could be even wiser to run mostly 32 bit applications on top of that to preserve some L1 instruction cache (64 bit instructions take more space)
Data size increase is probably more significant, since storing 64 pointers in trees, linked lists, and so on is what will really get you. The instruction size increase is pretty small, I think.
I wasn't trolling, just posting an idea I had, but haven't thoroughly thought through. I don't climb often, and I've only climbed in gyms recently. I'm not trying to pretend to be a climbing expert. I don't live or work in any tall buildings, either.
Obviously you have to buy a rope that's long enough. They are available, right? I guess if you get a rope not designed for climbing, you have to make sure your rappelling device will work with it, or use the leather gloves. As for getting out of the window, you might need to keep a hammer nearby to break it if it doesn't open.
Why don't you see if anything occurs to you for how to make it work, rather than dismissing it out of hand.
> Remember, intelligence agencies have a very bad track record at predicting future > events. That may be because the incentives affecting an individual employee > reward political conformity rather than correctness.
If intelligence people were part of the system, the terrorists would know which of their schemes had been discovered by the going price for futures in it. You don't want people at the CIA tipping off the terrorists to what they know.
> I'm 35 floors up, 40 blocks uptown from where the WTC was, > where lots of people I know and love got killed.
I'm very sorry to hear that. I've got an idea that might help in future situations like that.
Get a nice long rope, and find somewhere near a window to tie it to. If shit happens (including an ordinary accidental fire), go down the rope instead of freaking out and jumping. If you're serious, or if you want to get into climbing anyway once you have the rope, buy a harness and a descender (so you can rappel down). If not, buy a thick-enough pair of leather gloves so you can slide down the rope with your hands still covered by skin. If you just go with the gloves and the rope, there are much cheaper ropes than climbing-quality ones. Climbing ropes have to deal with abrasion without losing their strength, plus they need to handle shock loads. A rope that sits on your floor or in a drawer doesn't need that.
Unless some mofos blow up the whole building and it falls over before you have time to save yourself (and hopefully others, esp. if there's more than one glove available! Make sure you get a good strong rope so it won't fail if more than one person is on it), you shouldn't have to worry too much.
One problem: A big fire could burn through the rope, or make part of the way down uncomfortable.
Anyway, as my scout leader said, "Shit happens. Don't let shit happen to you." Don't forget to learn to tie a decent knot that will hold, or make some kind of arrangement so you don't screw up on that part.
> If the terrorist cell has been infiltrated, or have their phones tapped it won't turn up on the system.
Yeah, I guess gov't employees would have to be excluded from trading in the system, because the less idealistic ones would tend to reveal clues about what the gov't spies and agencies knew.
That's probably because of self-discharge. NiMH batteries suck for low-drain long-time applications. Rechargeable alkalines are supposed to be good for that, but their ideal situation is to be recharged back to full after every little bit of use. If you're going to be recharging them all the time anyway (for things you use regularly, not like a flashlight that you just need to be ready, without often using it), you might as well be using NiMH. I've seen at least one poster on this story claim that rechargeable alkalines worked fine for him even after regularly draining them almost completely, but I don't know about that.
A mouse with a charger cradle would be perfect, like this.
I second that. I've got a pair of C batteries for my bike light. They're made by GP, aka the "Gold Peak Group". 1.2V, 2200mAh, but they're a couple years old, not really used very often, and they're far worse than somewhat newer Rayovac batteries (also 1.2V, 2200mAh). (The Rayovacs have only ever been charged in a good Rayovac charger that knows about NiMH, while the GPs were charged many times in a Dynacharge quick-charger for NiCd batteries. Hmm, that's probably it, given what other people have said about chargers.) The (damaged?) GP batteries seem to have less capacity than the Rayovacs, and I think their shelf life is shorter, like a week instead of a few weeks for the Rayovacs.
My Rayovac charger is a model PS3, not a "quick charger", but it can charge lots of batteries at once, up to D cells, not just AAs. It kicks ass.
I got all my Rayovac gear at Mountain Equipment Co-op. Unfortunately, they don't seem to have rechargeable alkalines.
So give them a general anaesthetic, and remove their vital organs for transplant. No pain, much gain.
However, I have serious reservations about the death penalty being applied unless the standards of proof are significantly higher than they are now. (Fortunately for me, I live in Canada, where we don't have executions. Even our military hardly ever kills anyone.):/
> The word "seminal" means "containing seeds of later development."
In the case of the steady-state universe theory, that implies early, since the theory's now widely considered to be disproved, so there isn't any "later development" going on, at least not any widely accepted work. I don't think non-proponenets would call recent work on the theory "seminal".
However, keep in mind that it is possible (but unlikely) that the Universe is in some sort of steady state. We can never be 100% sure about anything (other than logically necessary or provably true things like 1+1=2, or Fermat's last theorem, respectively).
As for Dr. Gold, Dr. Robert Ehrlich's book "Nine Crazy Ideas in Science" discusses Dr. Gold's theory that so-called fossil fuels actually originate deep within the earth, and well up to the surface in some places. There is some interesting evidence supporting this theory, and Ehrlich rates it as zero cuckoos, or "why not?", whereas the theory that there was no big bang earns three, along with the theory that AIDS is not caused by HIV, and the theory that more guns means less crime. Ehrlich mentions in a footnote that Dr. Gold has "been shown to be right in such diverse areas as a theory of hearing, the nature of pulsars, and a theory of the Earth's axis of rotation." It's a good book, BTW. I recommend picking up a copy, or at least see if the library has it.
It seems that Dr. Gold is prepared to take unpopulare ideas seriously, just in case he comes up with something important. The world needs scientists like that to keep everyone from getting complacent.:)
> If you invested millions into a technology nobody else has wouldn't you like > some guarantee that you have the ability to return the investment?
Governments shouldn't just give business everything they want, just because they want it. (They seem to be moving in that direction these days, though.:( ) There are good arguments for some sort of patent system, but that isn't one of them.
BTW, you can't patent solar sails in general. It's already been though of, and published in papers, textbooks, and science fiction. Patents on ways to make very light reflective material and designs for stretching it out as a sail could be obtained, though.
Hmm, that does suck. Any countries thinking about making a law like that should read the comments on this/. story, and fix problems like that. (people of Europe, email the appropriate government members.)
As others have said, this isn't a problem on the web. They put their rebuttal on their server, you link to it. Unless they go out of their way to get a ton of people to visit your site (which would be stupid, because they're probably rather nobody saw your allegations in the first place), nothing they do is going to increase your bandwidth usage.
mod parent up, or do some research and see if there is a way to change timeouts.
ipchains (kernel 2.2) used to have a variable timeout for NAT, but I think they changed that for netfilter/iptables (kernel 2.4). because they improved something so "normal" connections got dropped when they were over, so the kernel's tables didn't fill up, even with a long timeout. With ipchains, I used to set the timeout to > 2 hours, so TCP keepalives would keep SSH sessions from being dropped. I don't have to do that with iptables (and I guess I actually can't!)
BTW, you're right about the 5 days. Running iptstate on my NAT firewall shows that quite clearly. I had been thinking it was two hours, but it's actually 120 hours! I've got a couple connections kicking around with 60 hours left to time out.
I don't know about tools to do it without breaking the connection (because I think the two sides would get confused about sequence numbers if you just stuck an extra packet in), but I have come across the nemesis project. Haven't actually used it, though. For more general packet-generating, sendip can generate arbitrary ip packets.
Luckily, ambitious hackers have already thought of this. (minus the web interface...)
iptstate does that (not the traffic volume, though) on a Linux NAT firewall. It actually shows you everything the netfilter connection tracking stuff is keeping track of, so it's useful on any stateful firewall, not just NAT.
> One of you guys who thinks they're so goddamn smart write a better algorithm for keeping the troublemakers out of the US.
Stop trying. It won't work. Tons of potential terrorists are already in the US. If the spooks do their job and keep serious terrorism down to much less than other violent crime, what's the problem. Treat terrorism (including support/funding, esp. for suicide attacks) as the violent crime it is. Catch the bastards after they try to do their thing. Stop being so afraid of stuff all the time. Just make sure the CIA gets their act together so they can keep an eye out for terrorist boot camps, and see who's going to them, and stuff like that, and stop harrassing so many innocent people.
This fascination with airplanes is really ridiculous. It's not like hijacking airplanes is the only thing people can do. Car bombs are _so_ much easier, since an angry mob isn't going to let anyone hijack their plane with a metal airline fork. (I know I'd be willing to take on someone with a fork to save my life, esp. if I had a plane full of people to throw stuff at him and help out.) OTOH, for causing terror and upsetting "the economy", getting people freaked about airplanes seems to have worked.
Slashdot Mirror
> I'm not a techwiz, but wouldn't full 64-bit memory addressing affect the Tflops,
> and ultimately, the cluster's performance?
No, it wouldn't. See ARStechnica.com's articles about 64bit and what it will really do for you, for example.
delete it from what? If there's no hierarchical directory structure, all you can do is delete it from one particular index (or from all of them, if that's supported). I can imagine a few possible designs, such as a "normal" fs with directories and files, but with indexes as well, or just having indexes (like one big directory).
MS can't legally distribute GPLed code linked into their kernel without giving away source to the whole thing. RedHat has only the GPL's restrictions on people copying what's on the CDs (i.e. you can copy and redistribute, but you have to make source available), and they give away source, thus fulfilling the requirements of the GPL.
So, in summary, the difference is that RH gives away the source, and I implicitly assumed MS won't.
> That hurts a bit, but my reaction is to say that AOL doesn't need my mail. But
> what happens when ISPs start to enforce no-server limitations? What happens when
> governments start to enforce them?!
Here's what John Walker thinks will happen if these trends continue. I don't like the trend toward a consumer Internet either, since I run a mail and DNS server on my cable modem, not to mention an ssh server so I can get at my stuff from other computers. My ISP has/had? a policy against _file_ servers, which is how I plan to justify running the servers I do if they ever give me any trouble.
> Eventually, this will lead to healthy competition between the "subculture nets" > and "The Internet".
That kind of competition isn't healthy. DNS inconsistencies suck. It's extra work for anyone who has to make sure they're DNS setup is ok.
> (we all know there's no such thing as The Internet, right? that it's just a
> generic term that we use to refer to consumers of IPV4 address space).
The Internet is an IP internetwork. It is the biggest one around, and is really the only global one that most people are on, so we call it the Internet.
Here's the theory behind carbon dating:
-Carbon in the atmosphere is a mix of C12 and C14. C14 is unstable, so it's always decaying, but is replaced by nuclear processes caused by cosmic rays, etc.
-Anything that gets new carbon from the atmosphere all the time, and gets rid of old carbon (e.g. living things) will have a C12:C14 ratio that depends on the atmospheric ratio. (You're right that living things slightly prefer one of isotopes over the other, and that the ratio varied over time).
-If the thing stops refreshing its carbon from the atmosphere (e.g. by dieing), the C14 will decay. You can measure the C12:C14 ratio, and given the decay rate of C14, you can calculate how long ago the carbon in your sample was cut off from the atmosphere.
If you had a way to figure out what the initial C12:C14 ratio in the water was, you could figure out when the sample had been cut off. Depending on how well aerated the water was, the C12:C14 might not be the same as the atmosphere. You wouldn't be able to tell if there had been short intervals where your puddle (or whatever) was exposed to the air, but they would throw off your calculations.
There is a radioactive decay dating process for sea water, IIRC. Oxygen, I think. There's also potassium-argon dating, which I think is useful for volcanic rocks with pockets of air to trap the argon.
> directory filename from /home/joe where filesize > 1 MB and mtime 1 day ago
/home/joe -type d -size +1024k -mtime -1
>
> would list all filenames for recent, larger files.
And you think this would be faster than what we have now?
find
(and you can combine find's arguments with and or or logic, like with SQL. Obviously you don't have the same power as SQL with just find by itself. Traversing directories doesn't take that long, esp. if you've already done it recently so everything's in the cache.)
> If you create multiple indexes, that comes down to having files in different
> directories at the same time; keeping multiple indexes consistent (when files
> are deleted) will cost overhead
Unix filesystems allow files to be linked to multiple directories at once, and they seem to do ok. (inode-based filesystems have a refcount in the inode, which is incremented and decremented as it is linked and unlinked from directories. Now you know why the system call rm(1) uses is unlink(2).). See ln(1) and link(2), and see the second column of the output of ls -l for the number of links a file has.
Unix filesystems don't provide a way to find all links to a file, which is, as you say, what you would need if you wanted to be able to delete a file entirely instead of just unlinking it from its current location. That would cost overhead.
Otherwise, I wouldn't worry.
> No, no :) They could change it slightly, in
> undocumented ways, like they did with Kerberos and
> Ldap for Active Directory.
Reiserfs is GPLed, so they couldn't do that to the Free version without rewriting from scratch (with the requisite incompatibilities). They could buy a license from namesys and get their own copy of the code that they could gratuitously change, though.
And BTW, the first time I saw something about WinFS, I thought it sounded like reiserfs.
Duke 3D: :)
"It's time to kick ass and chew bubble gum... And I'm all outta guhhhm." (Big drawl on the last "gum"
When you meet the Lunar Assault overlord at the end of mission 2 (Lunar Apocalypse), Duke says "I'm gonna rip off your head and shit down your neck". And you don't even need to imagine it, if you beat the ugly mofo (he reads the paper and whistles while shitting down the thing's neck...). A little bit too weird for regular usage, unlike the gum line, which is good for getting psyched up for a variety of sports.
>
If you look at it another way, once 64bit platforms become widespread, programs will start assuming they can just mmap anything they want. That will be a Good Thing, since sometimes mmap()ing is a more efficient or nicer way to do things.
I'm not sure whether mmap() is a higher performance way to do things for file/web servers, though. The TLB misses from mapping that much stuff might hurt... For Linux-optimized stuff, you can use sendfile(2).
Data size increase is probably more significant, since storing 64 pointers in trees, linked lists, and so on is what will really get you. The instruction size increase is pretty small, I think.
I wasn't trolling, just posting an idea I had, but haven't thoroughly thought through. I don't climb often, and I've only climbed in gyms recently. I'm not trying to pretend to be a climbing expert. I don't live or work in any tall buildings, either.
Obviously you have to buy a rope that's long enough. They are available, right? I guess if you get a rope not designed for climbing, you have to make sure your rappelling device will work with it, or use the leather gloves. As for getting out of the window, you might need to keep a hammer nearby to break it if it doesn't open.
Why don't you see if anything occurs to you for how to make it work, rather than dismissing it out of hand.
> Remember, intelligence agencies have a very bad track record at predicting future
> events. That may be because the incentives affecting an individual employee
> reward political conformity rather than correctness.
If intelligence people were part of the system, the terrorists would know which of their schemes had been discovered by the going price for futures in it. You don't want people at the CIA tipping off the terrorists to what they know.
> I'm 35 floors up, 40 blocks uptown from where the WTC was,
> where lots of people I know and love got killed.
I'm very sorry to hear that. I've got an idea that might help in future situations like that.
Get a nice long rope, and find somewhere near a window to tie it to. If shit happens (including an ordinary accidental fire), go down the rope instead of freaking out and jumping. If you're serious, or if you want to get into climbing anyway once you have the rope, buy a harness and a descender (so you can rappel down). If not, buy a thick-enough pair of leather gloves so you can slide down the rope with your hands still covered by skin. If you just go with the gloves and the rope, there are much cheaper ropes than climbing-quality ones. Climbing ropes have to deal with abrasion without losing their strength, plus they need to handle shock loads. A rope that sits on your floor or in a drawer doesn't need that.
Unless some mofos blow up the whole building and it falls over before you have time to save yourself (and hopefully others, esp. if there's more than one glove available! Make sure you get a good strong rope so it won't fail if more than one person is on it), you shouldn't have to worry too much.
One problem: A big fire could burn through the rope, or make part of the way down uncomfortable.
Anyway, as my scout leader said, "Shit happens. Don't let shit happen to you." Don't forget to learn to tie a decent knot that will hold, or make some kind of arrangement so you don't screw up on that part.
> If the terrorist cell has been infiltrated, or have their phones tapped it won't turn up on the system.
Yeah, I guess gov't employees would have to be excluded from trading in the system, because the less idealistic ones would tend to reveal clues about what the gov't spies and agencies knew.
> but the NiMH (1900mAH) lasted two weeks.
That's probably because of self-discharge. NiMH batteries suck for low-drain long-time applications. Rechargeable alkalines are supposed to be good for that, but their ideal situation is to be recharged back to full after every little bit of use. If you're going to be recharging them all the time anyway (for things you use regularly, not like a flashlight that you just need to be ready, without often using it), you might as well be using NiMH. I've seen at least one poster on this story claim that rechargeable alkalines worked fine for him even after regularly draining them almost completely, but I don't know about that.
A mouse with a charger cradle would be perfect, like this.
I second that. I've got a pair of C batteries for my bike light. They're made by GP, aka the "Gold Peak Group". 1.2V, 2200mAh, but they're a couple years old, not really used very often, and they're far worse than somewhat newer Rayovac batteries (also 1.2V, 2200mAh). (The Rayovacs have only ever been charged in a good Rayovac charger that knows about NiMH, while the GPs were charged many times in a Dynacharge quick-charger for NiCd batteries. Hmm, that's probably it, given what other people have said about chargers.) The (damaged?) GP batteries seem to have less capacity than the Rayovacs, and I think their shelf life is shorter, like a week instead of a few weeks for the Rayovacs.
My Rayovac charger is a model PS3, not a "quick charger", but it can charge lots of batteries at once, up to D cells, not just AAs. It kicks ass.
I got all my Rayovac gear at Mountain Equipment Co-op. Unfortunately, they don't seem to have rechargeable alkalines.
So give them a general anaesthetic, and remove their vital organs for transplant. No pain, much gain.
:/
However, I have serious reservations about the death penalty being applied unless the standards of proof are significantly higher than they are now. (Fortunately for me, I live in Canada, where we don't have executions. Even our military hardly ever kills anyone.)
> The word "seminal" means "containing seeds of later development."
:)
In the case of the steady-state universe theory, that implies early, since the theory's now widely considered to be disproved, so there isn't any "later development" going on, at least not any widely accepted work. I don't think non-proponenets would call recent work on the theory "seminal".
However, keep in mind that it is possible (but unlikely) that the Universe is in some sort of steady state. We can never be 100% sure about anything (other than logically necessary or provably true things like 1+1=2, or Fermat's last theorem, respectively).
As for Dr. Gold, Dr. Robert Ehrlich's book "Nine Crazy Ideas in Science" discusses Dr. Gold's theory that so-called fossil fuels actually originate deep within the earth, and well up to the surface in some places. There is some interesting evidence supporting this theory, and Ehrlich rates it as zero cuckoos, or "why not?", whereas the theory that there was no big bang earns three, along with the theory that AIDS is not caused by HIV, and the theory that more guns means less crime. Ehrlich mentions in a footnote that Dr. Gold has "been shown to be right in such diverse areas as a theory of hearing, the nature of pulsars, and a theory of the Earth's axis of rotation." It's a good book, BTW. I recommend picking up a copy, or at least see if the library has it.
It seems that Dr. Gold is prepared to take unpopulare ideas seriously, just in case he comes up with something important. The world needs scientists like that to keep everyone from getting complacent.
> If you invested millions into a technology nobody else has wouldn't you like
:( ) There are good arguments for some sort of patent system, but that isn't one of them.
> some guarantee that you have the ability to return the investment?
Governments shouldn't just give business everything they want, just because they want it. (They seem to be moving in that direction these days, though.
BTW, you can't patent solar sails in general. It's already been though of, and published in papers, textbooks, and science fiction. Patents on ways to make very light reflective material and designs for stretching it out as a sail could be obtained, though.
Hmm, that does suck. Any countries thinking about making a law like that should read the comments on this /. story, and fix problems like that. (people of Europe, email the appropriate government members.)
As others have said, this isn't a problem on the web. They put their rebuttal on their server, you link to it. Unless they go out of their way to get a ton of people to visit your site (which would be stupid, because they're probably rather nobody saw your allegations in the first place), nothing they do is going to increase your bandwidth usage.
mod parent up, or do some research and see if there is a way to change timeouts.
ipchains (kernel 2.2) used to have a variable timeout for NAT, but I think they changed that for netfilter/iptables (kernel 2.4). because they improved something so "normal" connections got dropped when they were over, so the kernel's tables didn't fill up, even with a long timeout. With ipchains, I used to set the timeout to > 2 hours, so TCP keepalives would keep SSH sessions from being dropped. I don't have to do that with iptables (and I guess I actually can't!)
BTW, you're right about the 5 days. Running iptstate on my NAT firewall shows that quite clearly. I had been thinking it was two hours, but it's actually 120 hours! I've got a couple connections kicking around with 60 hours left to time out.
I don't know about tools to do it without breaking the connection (because I think the two sides would get confused about sequence numbers if you just stuck an extra packet in), but I have come across the nemesis project. Haven't actually used it, though. For more general packet-generating, sendip can generate arbitrary ip packets.
Luckily, ambitious hackers have already thought of this. (minus the web interface...)
iptstate does that (not the traffic volume, though) on a Linux NAT firewall. It actually shows you everything the netfilter connection tracking stuff is keeping track of, so it's useful on any stateful firewall, not just NAT.
Also, try ipband, iftop, or iptraf.
> One of you guys who thinks they're so goddamn smart write a better algorithm for keeping the troublemakers out of the US.
Stop trying. It won't work. Tons of potential terrorists are already in the US. If the spooks do their job and keep serious terrorism down to much less than other violent crime, what's the problem. Treat terrorism (including support/funding, esp. for suicide attacks) as the violent crime it is. Catch the bastards after they try to do their thing. Stop being so afraid of stuff all the time. Just make sure the CIA gets their act together so they can keep an eye out for terrorist boot camps, and see who's going to them, and stuff like that, and stop harrassing so many innocent people.
This fascination with airplanes is really ridiculous. It's not like hijacking airplanes is the only thing people can do. Car bombs are _so_ much easier, since an angry mob isn't going to let anyone hijack their plane with a metal airline fork. (I know I'd be willing to take on someone with a fork to save my life, esp. if I had a plane full of people to throw stuff at him and help out.) OTOH, for causing terror and upsetting "the economy", getting people freaked about airplanes seems to have worked.