Oooh, wow, making money *launching satellites*. Yeah, that's a real accomplishment. Whoopee. Frankly, I don't give a damn about launching sats into orbit... if it was up to corps, we'd have never have launched the Apollo missions, let alone Pathfinder or any of the other interplanetary scientific missions. Why? Because there's no money in sending missions outside of Earth, hence no commercial impedence to do so. Thus, this kind of work will always need government funding until, as the grandparent poster mentioned, interplanetary space flight is commercially profitable.
Buffy fans in the audience forgive me, but I think the majority of that show's fanbase is composed of...
And I think all people who are interested in computers are geeky, socially inept freaks. Oh, wait, that's wrong too... this is what happens when you try to stereotype a group of people who are interested in something you're not. After all, just because you don't "get it" doesn't mean that there isn't a diverse group of people out there who disagree with you.
Perhaps, but do we really care if that code gets into Linux? Do you really want a bunch of patches in the Kernel that allow it run on a specific set of hardware in a particular DVD player?
My point is that unless these companies are making interesting architectural changes to the kernel, or otherwise improving it in a way that more than just a few people care about, what difference does it make if they release their changes or not?
Presuming, of course, that they're making any substantive changes to the kernel code base that anyone is actually interested in. After all, most of the modifications they do make are probably very specific to the hardware they're using, etc, and hence aren't that useful to anyone outside the company.
Well, according to the paper, during their simulations, even flow-based throttling mechanisms have trouble with this attack (they specifically mention RED-PD).
Too bad this is a *completely different attack*! Jeez, read the friggin' paper, people. The paper you reference talks about a DoS which exploits data structures commonly used in TCP stacks. The DoS in the paper referenced for this article exploits TCP congestion control algorithms to "fool" the TCP stack into thinking the pipe is full when it really isn't by sending carefully timed packet bursts.
That may be the case for the small-scale RTT period, but according to the paper, the large-scale RTO is more predictable (eg, a constant, like 1 second), meaning it's exploitable. The paper does describe some techniques for randoming the RTO, but they aren't fool proof.
"Using a combination of analytical modeling, simulations, and Internet experiments, we show that maliciously chosen low-rate DoS traffic patterns that exploit TCP's retramission time-out mechanism can throttle TCP flows to a small fraction of their ideal rate while eluding detection."
And, in the next section:
"Considering first a single TCP flow, if the total traffic (DoS and TCP traffic) during an RTT-timescale burst is sufficient to induce enough packet losses, the TCP flow will enter a timeout and attempt to send a new packet RTO seconds later. If the period of the DoS flow approximates the RTO of the TCP flow, the TCP flow will continually incur loss as it tries to exit the timeout state, fail to exit timeout, and obtain near zero throughput."
Sounds like they're talking about unicast flows to me. But, perhaps I'm missing something...
When the number of flows in the system is high, a fraction of flows' retransmission timers will expire sufficiently near time (alpha) such that those flows can partially recover and utilize the available bandwidth in the period from time (alpha) to time (beta), when all flows will again experience an outage.
Bah, the paper isn't that bad. Heck, without reading the whole thing and knowing a little bit about what they discuss (based on the first section), I can understand what you've quoted (if I'm correct, this is from their section on mitigating attacks using randomized RTOs).
Really, the basic concepts are *incredibly* simple. Send a burst of traffic which causes drops in the short term. This results in the TCP stack backing off and re-transmitting the packet after the defined RTO. So, if you hit the stack with another burst of packets just as the RTO is expiring, the stack will back off again. Lather, rinse, repeat. This requires a lot less traffic, since your bursts are spaced apart (roughly a second per burst, typically, since that's a pretty standard RTO).
Really, all you need is a basic understanding of TCP flow control to understand the concepts in this paper (which, BTW, they attempt to explain in the first section). The rest of the content (modelling TCP flow rates relative to DoS flow rates, etc) is really just the formal analysis of the basic attack, which certainly isn't important if all you care about is implementing it.
Given that the majority of things these days even are made of ceramics or plastics which have readily abundant quantities,
Hmm... first, plastics and ceramics aren't that abundant... they require mining for minerals or hydrocarbons in order to manufacture. Thus, there is *plenty* of scarcity here (wow, talk about an oxymoron:).
Second, even if these compounds were limitless, last I checked, you can't make food from plastics or ceramics. Basically, these compounds have a limited (if large) range of uses... so there'll still be plenty of scarcity in other areas.
Fusion power is not a limitless power supply. Certainly *very* abundant, though we still have many problems containing and sustaining the reaction, with little to no foreseeable improvements. I may be
Eh, things happen in baby steps (although "no foreseeable improvements" is, IMHO, a vast overstatement... things advance, if slowly). Still, it's really just a matter of getting the engineering right (eg, whether magnetic or inertial confinement is the way to go, etc). Basically, I'm of the opinion that these practical issues will be sorted out within the next century.
wrong, in fact there's a good chance I am... but I *do* expect fine molecular manipulation before commercially viable fusion power.
En masse? I doubt it. We can barely manifacture simple nanoscale devices, and even then, only in *highly* controlled environments. Frankly, we don't even fully understand the engineering problems at these scales (for example, quantum effects are a real problem when dealing in nanotechnology). Besides, even if we *could* manufacture anything we wanted (via molecular manipulation), you still need plenty of raw materials... sounds like scarcity to me.
The fact is, until we can take matter and transmute it into whatever compounds we want, and *then* take those compounds and fabricate whatever materials we want, there will always be scarcity (on the "stuff"-side).
OTOH, once large scale renewable power generation can be done cheaply and efficiently (fusion, solar, etc), we'll have a practically unlimited supply of power (heck, the moon has enough He3 to supply us for a LONG time).
And this would be why trusting science to corporate interests is a *bad* thing. Because they're obsessed with the profit motive, they'll never do work which doesn't have obvious immediate benefit. Hell, with this attitude, we might as well just scrap the entire space program... there's no diamonds in orbit, last I checked.
Woah, wait a sec, you think matter replication/manufacture (and hence unlimited "stuff") will be developed before some limitless power supply (fusion, solar, etc) is developed? You're kidding, right? The technology for matter replication isn't anywhere in our forseeable future, technology-wise... we don't even know *how* to do it. Power production, OTOH, is a well-understood problem. The only hard part is the engineering, which, while admittedly very hard, is at least within our current technological boundaries (fusion power, for example, while not a practical reality, is at least being heavily researched).
And who is AOL competing with that this is unfair to? If anything, this is *damaging* AOLs business, since people will be unsatisfied with their service (due to undelivered emails) and leave.
Umm... OOC, did you try using your ISPs SMTP server as your "default gateway" when sending email? I've done this, and I've had no problems sending emails (AFAICT), and it still gives me all the control I want... my mail server just relays all email through my ISPs email server. I see no problems with this. *shrug*
And you make an interesting point, too... especially considering that trade secret law apparently gives the owner of the secret a form of indefinite monopoly, in that the owner can always bar another from using their secret if it was misappropriated...
Dude, you're mixing up your IP. Trade secrets and patents are pretty well diametrically opposed. A patent on an item gives you the right to a limited monopoly on the production/use/whatever of that item. However, in exchange for those rights, you must publically disclose, in detail, the workings of your item. A trade secret, OTOH, is just that. A secret. It absolutely must not be publically disclosed (hence the use of contracts, NDAs, etc, to prevent exposure of the secret).
Of course, that doesn't change your point that trade secrets are valuable (your Coca-Cola formula example is, actually, a good one... it's a trade secret, not a patent), however, one must not allow corporations to trump the rights of the public in order to protect their bottom line (a disturbing trend these days).
What does that mean? Correct me if I'm wrong, but last I checked, there's no such thing as "trade secret rights". Trade secrets are secret because you keep them secret (via NDA or whatever). Once they escape, they're public knowledge, end of story. I wonder how long it'll take before trade secrets are lumped together with patents, copyrights, and trademarks as "IP". *sigh*
Is it so hard to believe that the caffeine in coffee is more efficiently extracted than caffeine in tea, resulting in a more potent cup? I know, *crazy*...
AFAICT, yes, black tea contains alkaloids... it's called caffeine! However, I can find no reference which supports your claim. In fact, quite the opposite... there are a ton of sites which claim that black tea can help remove nicotine from the body (validation of this claim is left as an exercise for the reader).
For example, 90% of desktop CPU use could get by without floating point math, especially if there were some key fixed point instructions in the integer unit. But every Pentium 4 and Athlon not only includes 80-bit floating point units, but massive FP vector processing units as well. (Not to mention outmoded MMX instructions that are almost completely ignored.)
LOL! You MUST be trolling. Seriously! I'll bite anyway, though. How many people, with their computer:
1) play audio/video 2) edit audio/video 3) heck, edit still images 4) model/render 3D 5) play games
This is a small list PC applications classes, and I will guarantee you, they *all* use MMX or one of the other major vector units, and in many cases, the floating point unit is heavily used as well!
Sounds like a nice idea for the desktop or for certain classes of research, but there will always be a place for massive computational capacity on a single chip since there is a large class of computing problems which are not easily parallelizable, and hence can not take advantage of parallel computing.
Incidentally, there is also a limit to how fast your parallel computer will get... it's call the bus. If you can't build high speed interconnects, or if your software isn't designed well (not as easy as it sounds!), you will inevitably have problems with the system bus becoming overtaxed. Heck, this is already a problem in our primarily single-CPU world.
Slashdot Mirror
Oooh, wow, making money *launching satellites*. Yeah, that's a real accomplishment. Whoopee. Frankly, I don't give a damn about launching sats into orbit... if it was up to corps, we'd have never have launched the Apollo missions, let alone Pathfinder or any of the other interplanetary scientific missions. Why? Because there's no money in sending missions outside of Earth, hence no commercial impedence to do so. Thus, this kind of work will always need government funding until, as the grandparent poster mentioned, interplanetary space flight is commercially profitable.
Buffy fans in the audience forgive me, but I think the majority of that show's fanbase is composed of...
And I think all people who are interested in computers are geeky, socially inept freaks. Oh, wait, that's wrong too... this is what happens when you try to stereotype a group of people who are interested in something you're not. After all, just because you don't "get it" doesn't mean that there isn't a diverse group of people out there who disagree with you.
Perhaps, but do we really care if that code gets into Linux? Do you really want a bunch of patches in the Kernel that allow it run on a specific set of hardware in a particular DVD player?
My point is that unless these companies are making interesting architectural changes to the kernel, or otherwise improving it in a way that more than just a few people care about, what difference does it make if they release their changes or not?
Presuming, of course, that they're making any substantive changes to the kernel code base that anyone is actually interested in. After all, most of the modifications they do make are probably very specific to the hardware they're using, etc, and hence aren't that useful to anyone outside the company.
Well, according to the paper, during their simulations, even flow-based throttling mechanisms have trouble with this attack (they specifically mention RED-PD).
Too bad this is a *completely different attack*! Jeez, read the friggin' paper, people. The paper you reference talks about a DoS which exploits data structures commonly used in TCP stacks. The DoS in the paper referenced for this article exploits TCP congestion control algorithms to "fool" the TCP stack into thinking the pipe is full when it really isn't by sending carefully timed packet bursts.
That may be the case for the small-scale RTT period, but according to the paper, the large-scale RTO is more predictable (eg, a constant, like 1 second), meaning it's exploitable. The paper does describe some techniques for randoming the RTO, but they aren't fool proof.
Did we read the same paper? Quote:
"Using a combination of analytical modeling, simulations, and Internet experiments, we show that maliciously chosen low-rate DoS traffic patterns that exploit TCP's retramission time-out mechanism can throttle TCP flows to a small fraction of their ideal rate while eluding detection."
And, in the next section:
"Considering first a single TCP flow, if the total traffic (DoS and TCP traffic) during an RTT-timescale burst is sufficient to induce enough packet losses, the TCP flow will enter a timeout and attempt to send a new packet RTO seconds later. If the period of the DoS flow approximates the RTO of the TCP flow, the TCP flow will continually incur loss as it tries to exit the timeout state, fail to exit timeout, and obtain near zero throughput."
Sounds like they're talking about unicast flows to me. But, perhaps I'm missing something...
When the number of flows in the system is high, a fraction of flows' retransmission timers will expire sufficiently near time (alpha) such that those flows can partially recover and utilize the available bandwidth in the period from time (alpha) to time (beta), when all flows will again experience an outage.
Bah, the paper isn't that bad. Heck, without reading the whole thing and knowing a little bit about what they discuss (based on the first section), I can understand what you've quoted (if I'm correct, this is from their section on mitigating attacks using randomized RTOs).
Really, the basic concepts are *incredibly* simple. Send a burst of traffic which causes drops in the short term. This results in the TCP stack backing off and re-transmitting the packet after the defined RTO. So, if you hit the stack with another burst of packets just as the RTO is expiring, the stack will back off again. Lather, rinse, repeat. This requires a lot less traffic, since your bursts are spaced apart (roughly a second per burst, typically, since that's a pretty standard RTO).
Really, all you need is a basic understanding of TCP flow control to understand the concepts in this paper (which, BTW, they attempt to explain in the first section). The rest of the content (modelling TCP flow rates relative to DoS flow rates, etc) is really just the formal analysis of the basic attack, which certainly isn't important if all you care about is implementing it.
Given that the majority of things these days even are made of ceramics or plastics which have readily abundant quantities,
:).
Hmm... first, plastics and ceramics aren't that abundant... they require mining for minerals or hydrocarbons in order to manufacture. Thus, there is *plenty* of scarcity here (wow, talk about an oxymoron
Second, even if these compounds were limitless, last I checked, you can't make food from plastics or ceramics. Basically, these compounds have a limited (if large) range of uses... so there'll still be plenty of scarcity in other areas.
Fusion power is not a limitless power supply. Certainly *very* abundant, though we still have many problems containing and sustaining the reaction, with little to no foreseeable improvements. I may be
Eh, things happen in baby steps (although "no foreseeable improvements" is, IMHO, a vast overstatement... things advance, if slowly). Still, it's really just a matter of getting the engineering right (eg, whether magnetic or inertial confinement is the way to go, etc). Basically, I'm of the opinion that these practical issues will be sorted out within the next century.
wrong, in fact there's a good chance I am... but I *do* expect fine molecular manipulation before commercially viable fusion power.
En masse? I doubt it. We can barely manifacture simple nanoscale devices, and even then, only in *highly* controlled environments. Frankly, we don't even fully understand the engineering problems at these scales (for example, quantum effects are a real problem when dealing in nanotechnology). Besides, even if we *could* manufacture anything we wanted (via molecular manipulation), you still need plenty of raw materials... sounds like scarcity to me.
The fact is, until we can take matter and transmute it into whatever compounds we want, and *then* take those compounds and fabricate whatever materials we want, there will always be scarcity (on the "stuff"-side).
OTOH, once large scale renewable power generation can be done cheaply and efficiently (fusion, solar, etc), we'll have a practically unlimited supply of power (heck, the moon has enough He3 to supply us for a LONG time).
Someone should check out Engrish.com. Be careful, though... your co-workers may start to wonder why you're laughing so much. :)
One of my favorites is the joke of Python and Perl merging into a project called Parrot, which inspired this.
And this would be why trusting science to corporate interests is a *bad* thing. Because they're obsessed with the profit motive, they'll never do work which doesn't have obvious immediate benefit. Hell, with this attitude, we might as well just scrap the entire space program... there's no diamonds in orbit, last I checked.
Woah, wait a sec, you think matter replication/manufacture (and hence unlimited "stuff") will be developed before some limitless power supply (fusion, solar, etc) is developed? You're kidding, right? The technology for matter replication isn't anywhere in our forseeable future, technology-wise... we don't even know *how* to do it. Power production, OTOH, is a well-understood problem. The only hard part is the engineering, which, while admittedly very hard, is at least within our current technological boundaries (fusion power, for example, while not a practical reality, is at least being heavily researched).
And who is AOL competing with that this is unfair to? If anything, this is *damaging* AOLs business, since people will be unsatisfied with their service (due to undelivered emails) and leave.
Umm... OOC, did you try using your ISPs SMTP server as your "default gateway" when sending email? I've done this, and I've had no problems sending emails (AFAICT), and it still gives me all the control I want... my mail server just relays all email through my ISPs email server. I see no problems with this. *shrug*
*sigh* No, I just misread.
And you make an interesting point, too... especially considering that trade secret law apparently gives the owner of the secret a form of indefinite monopoly, in that the owner can always bar another from using their secret if it was misappropriated...
Dude, you're mixing up your IP. Trade secrets and patents are pretty well diametrically opposed. A patent on an item gives you the right to a limited monopoly on the production/use/whatever of that item. However, in exchange for those rights, you must publically disclose, in detail, the workings of your item. A trade secret, OTOH, is just that. A secret. It absolutely must not be publically disclosed (hence the use of contracts, NDAs, etc, to prevent exposure of the secret).
Of course, that doesn't change your point that trade secrets are valuable (your Coca-Cola formula example is, actually, a good one... it's a trade secret, not a patent), however, one must not allow corporations to trump the rights of the public in order to protect their bottom line (a disturbing trend these days).
Ahhh, *very* interesting. This I didn't know. Thanks.
What does that mean? Correct me if I'm wrong, but last I checked, there's no such thing as "trade secret rights". Trade secrets are secret because you keep them secret (via NDA or whatever). Once they escape, they're public knowledge, end of story. I wonder how long it'll take before trade secrets are lumped together with patents, copyrights, and trademarks as "IP". *sigh*
Is it so hard to believe that the caffeine in coffee is more efficiently extracted than caffeine in tea, resulting in a more potent cup? I know, *crazy*...
AFAICT, yes, black tea contains alkaloids... it's called caffeine! However, I can find no reference which supports your claim. In fact, quite the opposite... there are a ton of sites which claim that black tea can help remove nicotine from the body (validation of this claim is left as an exercise for the reader).
Teabags? Harrods Earl Grey is a loose tea, and it's priced at 6.65 for a 125g tin.
Which is, incidentally, the only way to have really good tea... loose, that is.
For example, 90% of desktop CPU use could get by without floating point math, especially if there were some key fixed point instructions in the integer unit. But every Pentium 4 and Athlon not only includes 80-bit floating point units, but massive FP vector processing units as well. (Not to mention outmoded MMX instructions that are almost completely ignored.)
LOL! You MUST be trolling. Seriously! I'll bite anyway, though. How many people, with their computer:
1) play audio/video
2) edit audio/video
3) heck, edit still images
4) model/render 3D
5) play games
This is a small list PC applications classes, and I will guarantee you, they *all* use MMX or one of the other major vector units, and in many cases, the floating point unit is heavily used as well!
Seriously, what kind of crack are you on?
Sounds like a nice idea for the desktop or for certain classes of research, but there will always be a place for massive computational capacity on a single chip since there is a large class of computing problems which are not easily parallelizable, and hence can not take advantage of parallel computing.
Incidentally, there is also a limit to how fast your parallel computer will get... it's call the bus. If you can't build high speed interconnects, or if your software isn't designed well (not as easy as it sounds!), you will inevitably have problems with the system bus becoming overtaxed. Heck, this is already a problem in our primarily single-CPU world.