Since CherryPal seems to be dodgy at worst, and minimally talented rebadge guys at best, what is the preferred direct source for the WM8505 or VT8500 based devices?
Is this an Alibaba, lots of 100+ thing?, Ebay?, Ebay and a bit of luck?, Dealextreme?
Is if any of these attacks are persistent/capable of lurking onboard waiting for some predefined trigger, without a device remaining connected to the diagnostics port.
While corporatist DRM apologists might disagree, the ability to do all sorts of crazy stuff by connecting to your local diagnostics port is what we call a "feature". If anything, we don't have enough control here, and much of the control we do have is inadequately documented "Oh, sure, it's ODBC, in that it is more or less electrically compatible. Good luck with those proprietary codes, and please see your dealer for regularly scheduled service!"
On the other hand, something that allows anybody with 30 seconds of physical access to flash crash_at_60.haxxx permanently into the ECU is what we would call a "major design flaw".
There are both economies and diseconomies to centralization. The real issue(in many "cloud" cases), is that some of the things that could be economies of centralization are being skipped in the mad rush for low costs, and since everything is hidden under a shiny layer of web APIs, people don't notice in time.
In this case, for instance, the cost per server, or per unit work done, to have Real Serious Redundant Power(batteries, generators, multiple utility links, etc.) plummets as the number of servers in a given location increases. As long as people keep in mind that "cloud" equals "buzzword for a set of methods of reducing the transaction costs of outsourcing a variety of IT functions" rather than "magic place where computations are done by happy computrons and packets are carried by unicorns" and ask the appropriate questions, the average reliability of power, bandwidth and other useful stuff should go up. If they fail to keep that in mind and start falling into the stupid(but seductive, and not exactly discouraged by vendors) assumption that "the web API makes it look super easy, so it must be super reliable", market forces will quickly drive the lowest common denominator down to a pretty grim level of service.
A lot of their stuff is freshwater, in any case. As long as the levels of assorting organic carcinogens evaporating and floating around and coming down in rainwater don't get too high, that isn't going to be their problem.
From the gulf of Mexico(in the atlantic), all the way around South America or Africa, into the Pacific, and thence to the high-density freshwater and estuarial operations on the Pacific rim?
Long term, quite possibly; but we are talking decades of pure profit, during which time the gulf fishermen will be catching nothing but tarballs.
"Extinction" is a very high bar to clear, except for losers like panda bears that are large enough to shoot and barely capable of reproducing without assistance.
However, "Ecological and social shifts leading to grinding, nigh-unendurable; but nowhere near fatal enough to kill you quickly and be done with it" is very much more common and plausible.
Unless we start fucking around with self-replicating strangelets, or largish black holes, or other really exotic stuff, "extinction" is not a serious risk. Even nukes would require some real doing. Unfortunately, though, pushing yourself into "and the living shall envy the dead" territory is typically easier than killing yourself off. Even fairly modest ecological disruption could do the bottom billion or so in(and one can hardly expect that they'll go quietly), and make things pretty unpleasant for the remainder.
Could the work that has been done in "slowing" light be used to interfere with this?(or, more practically, since the speed of light varies based on the medium, would you need a completely accurate characterization of the contents of the light paths that the signal travelled over for your certainty to be valid?
On the plus side, this will finally provide a way for Bob to prove to Alice's satisfaction that he isn't with Eve, and Alice will be able to demonstrate the same about Mallory. Bliss through superior quantum physics!
Somebody has, in seriousness, suggested that the military, or NASA, should step in and neutralize a private operator's errant satellite so that our precious, precious, TV service need not fear interruption. I have yet to see a single person mention that this is, perhaps, not a terribly important use of public funds. What gives?
The PS3 launched at $500, if I recall accurately, and may have drifted down a little after that, so the price depends on exactly when they were buying; but that is a decent ballpark figure.
Specs are 1 CPU, 256MB of XDR RAM(another 256 of video RAM; but I believe that that is largely useless for the OtherOS). 1 GigE port.
The QS21 is 7k for 2CPUs and 2 GB of DDR, along with 2 GigE and optional infiniband. For every 14 of those, you'll need a Bladecenter H(6K, plus options) or put 12 in a Bladecenter HT(14K plus options).
If your application requires more RAM, or tighter coupling than 1xGigE, you don't really have much of a choice. If you can get away with it, though, the price difference is nontrivial.
My understanding is that the GPU is neutered to hell and back, you basically just get a dumb framebuffer; but that the GPU is completely separate from the Cell, its basically just near-stock Nvidia silicon from a few generations back. The Cell, though, is more or less as functional as it ever is(which is to say that you only get 7 SPEs rather than the full 8 in the expensive compute gear).
Unless you program your application specifically to use the SPEs, PS3 Linux is basically just not-especially-fast PPC Linux with not much RAM; but the SPEs are available.
Not really a joke. Post civil war, that's basically what they did to get the newly emancipated back in their place, where possible. All sorts of crimes ("vagrancy") and the like, heavy enforcement against the undesirables, and then lease the resulting convicts out as cheap slave labor to various upstanding local businesses.
All perfectly legal and above board.
These days, of course, we have the private, for-profit prison, a truly brilliant institution. The outfits that run these are very reliable "law-and-order" lobbyists, and there was even a case a while back where they were paying a judge a per-inmate kickback for, shall we say, "referrals"...
Apparently, you can't send a unit in for service, even if under warranty or a pay-for-service deal, and not get a complementary upgrade to the latest firmware. How much of this is sinister, and how much is just because "service" probably often means "throw away/send for rework the one they sent, grab another from the pile-o-refurbs", and the service drones simply can't downgrade the firmware on the one from the pile-o-refurbs, is unclear.
Frankly, the more "enterprise" the software is, the less pity I have for the customer(and yes, I have dealt with some pretty ghastly "enterprise" software, this isn't just theoretical sniping from the outside).
If Joe Consumer buys a shiny box-o-shrinkwrap, he basically has zero power. Most stores won't do anything more than replacement-in-kind, for physical defects in the medium or accompanying accessories, no returns. The inevitable nasty EULA will be hiding inside the shrinkwrap that confirms your acceptance of it. His software may (typically) be fairly cheap; but he has absolutely fuck-all power of redress if it doesn't work(or if the online component changes without notice in 3 months, or a mandatory update breaks things, or whatever).
If you are buying "enterprise" software, though, you are probably talking an individual contract(and a real contract, with suits and lawyers on both sides, not some contract of adhesion crap), contact with the vendor, and maybe even a stack of customizations. Why, pray tell, did you let the vendor get away with delivering whatever they wanted, without any assurance that it will do what they said it would? Why didn't the contract specify scary penalties if they fuck up?
Consumer protection law basically exists as a form of aggregate power to protect those who have basically none during each of the transactions they make day to day(and, secondarily, to assist the sellers, by providing a more uniform commercial environment, where certain standards are universal.) However, the closer in power to one another the buyer and the seller are, the less obviously necessary or desirable it is. Enterprises are big boys, why do they need the state to write their contracts for them? In the case of software, there is a lot of excellent "as-is, don't sue me" stuff floating around. It would hardly be helpful to open everybody who, as a service to others, gives anon read access to their CVS, to product defect lawsuits. If you want to sell it in shiny boxes at Gamestop or Office Depot, sure, the customer had better be able to at least return it. If it is "enterprise" software, the state should certainly hold both parties to their sides of the deal, per contract law; but surely they can figure it out between themselves.
Apparently, for the particular bit of number-crunching that they were doing, Cell curb-stomped x86. They then concluded that, if you want Cell, your only options are a few absurdly expensive specialist compute servers(IBM makes one, Sony at least has a model number, I think that there are one or two others) or cheap PS3s. Since, when the started, OtherOS was a standard, supported, option, the only "hacking" involved was the inconvenience of having to touch each machine to kick off the install.
Had their algorithm not suited Cell, the PS3 would have been an absurd choice. Since it did, though, it was actually pretty sensible(barring Sony's hard-to-predict action).
Does loading those twenty modules hurt you if, by virtue of their being in common use, there is some trivial automated way of doing it?
With the exception of people writing bare-metal assembly for microcontrollers or something, pretty much everyone who sits down to write some code has huge swaths of pre-written stuff loaded for them. The only difference is how much of that happens automatically, by default, and how much you see and do yourself.
If the science types happen to like python for some syntactic or structural or design reasons; but need a bunch of modules to make it do what they want, it isn't exactly rocket surgery to bundle them all together so a single "import science" does the job, or even hack together a slight variant of the python environment where that particular import is simply done silently by default.
Oh, obviously family history isn't 100% useful in all situations.
My point was simply that, when evaluating a "predictive" genetic test, "more predictive than nothing" is, in most cases, not actually the standard you should be holding it to. "More predictive than family history" is, except when that is unavailable, the much more appropriate(and much tougher) standard. Particularly for conditions where the genetics are not straightforward, watching your relatives run similar genetic code often tells you more than getting your own raw sequence.
The question, of course, is to what degree lifestyle is influenced by genetic factors...
Unless we are going to cling to the (intuitively satisfying; but rather silly) theory that humans have some sort of extra-material "free will" floating around in the aether, we pretty much have to concede that behavior has a biological basis. And, if something has a biological basis, the odds of it having a genetic and/or epigenetic component are pretty decent.
They gypsy certainly has better applied psych skills; but, for a great many genetic(or suspected but not yet fully elucidated) conditions, there is a way that is cheaper and more effective.
Family history.
With the exception of (not-nonexistent; but quite rare) conditions caused by a mutation or mutations that originated with you, not earlier in the line, or a fairly small number of well developed genetic tests, most of which you aren't going to get over the counter at CVS, you'll have a better chance of learning about the likely phenotypic consequences of your genes by looking at mommy and daddy, keeping their environment in mind(daddy's lung cancer probably doesn't count as "family history" if he was a chain-smoking asbestos miner, it probably does if he wasn't).
DNA sequencing has, certainly, gotten cheap enough that you might actually get a fairly accurate reading of a subset of your genome for a hundred bucks through the mail. However, I'd be quite surprised if, when it comes to predicting the consequences, which are what people actually care about, the method is going to outperform just looking at family history. In a lot of cases, the science simply isn't settled, at any price. Even where it is, you are going to be getting some mail merge algorithm, not a geneticist, or even a genetic counselor, for your hundred bucks.
Slashdot Mirror
Since CherryPal seems to be dodgy at worst, and minimally talented rebadge guys at best, what is the preferred direct source for the WM8505 or VT8500 based devices?
Is this an Alibaba, lots of 100+ thing?, Ebay?, Ebay and a bit of luck?, Dealextreme?
Is if any of these attacks are persistent/capable of lurking onboard waiting for some predefined trigger, without a device remaining connected to the diagnostics port.
While corporatist DRM apologists might disagree, the ability to do all sorts of crazy stuff by connecting to your local diagnostics port is what we call a "feature". If anything, we don't have enough control here, and much of the control we do have is inadequately documented "Oh, sure, it's ODBC, in that it is more or less electrically compatible. Good luck with those proprietary codes, and please see your dealer for regularly scheduled service!"
On the other hand, something that allows anybody with 30 seconds of physical access to flash crash_at_60.haxxx permanently into the ECU is what we would call a "major design flaw".
There are both economies and diseconomies to centralization. The real issue(in many "cloud" cases), is that some of the things that could be economies of centralization are being skipped in the mad rush for low costs, and since everything is hidden under a shiny layer of web APIs, people don't notice in time.
In this case, for instance, the cost per server, or per unit work done, to have Real Serious Redundant Power(batteries, generators, multiple utility links, etc.) plummets as the number of servers in a given location increases. As long as people keep in mind that "cloud" equals "buzzword for a set of methods of reducing the transaction costs of outsourcing a variety of IT functions" rather than "magic place where computations are done by happy computrons and packets are carried by unicorns" and ask the appropriate questions, the average reliability of power, bandwidth and other useful stuff should go up. If they fail to keep that in mind and start falling into the stupid(but seductive, and not exactly discouraged by vendors) assumption that "the web API makes it look super easy, so it must be super reliable", market forces will quickly drive the lowest common denominator down to a pretty grim level of service.
A lot of their stuff is freshwater, in any case. As long as the levels of assorting organic carcinogens evaporating and floating around and coming down in rainwater don't get too high, that isn't going to be their problem.
From the gulf of Mexico(in the atlantic), all the way around South America or Africa, into the Pacific, and thence to the high-density freshwater and estuarial operations on the Pacific rim?
Long term, quite possibly; but we are talking decades of pure profit, during which time the gulf fishermen will be catching nothing but tarballs.
I'm assuming that southeast asia's fish farmers are positively joyful at the prospect...
More's the pity.
"Extinction" is a very high bar to clear, except for losers like panda bears that are large enough to shoot and barely capable of reproducing without assistance.
However, "Ecological and social shifts leading to grinding, nigh-unendurable; but nowhere near fatal enough to kill you quickly and be done with it" is very much more common and plausible.
Unless we start fucking around with self-replicating strangelets, or largish black holes, or other really exotic stuff, "extinction" is not a serious risk. Even nukes would require some real doing. Unfortunately, though, pushing yourself into "and the living shall envy the dead" territory is typically easier than killing yourself off. Even fairly modest ecological disruption could do the bottom billion or so in(and one can hardly expect that they'll go quietly), and make things pretty unpleasant for the remainder.
Could the work that has been done in "slowing" light be used to interfere with this?(or, more practically, since the speed of light varies based on the medium, would you need a completely accurate characterization of the contents of the light paths that the signal travelled over for your certainty to be valid?
On the plus side, this will finally provide a way for Bob to prove to Alice's satisfaction that he isn't with Eve, and Alice will be able to demonstrate the same about Mallory. Bliss through superior quantum physics!
They could have released their own clean version; but they decided that it would be more fun to pirate the pirates' efforts...
Don't you understand? "Open" means "Able to Run Flash as God intended" not some piffle about "does what its owner wants it to"...
Somebody has, in seriousness, suggested that the military, or NASA, should step in and neutralize a private operator's errant satellite so that our precious, precious, TV service need not fear interruption. I have yet to see a single person mention that this is, perhaps, not a terribly important use of public funds. What gives?
The PS3 launched at $500, if I recall accurately, and may have drifted down a little after that, so the price depends on exactly when they were buying; but that is a decent ballpark figure.
Specs are 1 CPU, 256MB of XDR RAM(another 256 of video RAM; but I believe that that is largely useless for the OtherOS). 1 GigE port.
The QS21 is 7k for 2CPUs and 2 GB of DDR, along with 2 GigE and optional infiniband. For every 14 of those, you'll need a Bladecenter H(6K, plus options) or put 12 in a Bladecenter HT(14K plus options).
If your application requires more RAM, or tighter coupling than 1xGigE, you don't really have much of a choice. If you can get away with it, though, the price difference is nontrivial.
Because putting things in space is profitable, while taking them back down is just a cost center, and generally not a legally obligatory one, at that?
My understanding is that the GPU is neutered to hell and back, you basically just get a dumb framebuffer; but that the GPU is completely separate from the Cell, its basically just near-stock Nvidia silicon from a few generations back. The Cell, though, is more or less as functional as it ever is(which is to say that you only get 7 SPEs rather than the full 8 in the expensive compute gear).
Unless you program your application specifically to use the SPEs, PS3 Linux is basically just not-especially-fast PPC Linux with not much RAM; but the SPEs are available.
Not really a joke. Post civil war, that's basically what they did to get the newly emancipated back in their place, where possible. All sorts of crimes ("vagrancy") and the like, heavy enforcement against the undesirables, and then lease the resulting convicts out as cheap slave labor to various upstanding local businesses.
All perfectly legal and above board.
These days, of course, we have the private, for-profit prison, a truly brilliant institution. The outfits that run these are very reliable "law-and-order" lobbyists, and there was even a case a while back where they were paying a judge a per-inmate kickback for, shall we say, "referrals"...
Apparently, you can't send a unit in for service, even if under warranty or a pay-for-service deal, and not get a complementary upgrade to the latest firmware. How much of this is sinister, and how much is just because "service" probably often means "throw away/send for rework the one they sent, grab another from the pile-o-refurbs", and the service drones simply can't downgrade the firmware on the one from the pile-o-refurbs, is unclear.
How long before "Non-Practicing Entity" goes from harmless-sounding euphemism to sinister dysphemism, the way terms like "Ethnic Cleansing" have?
Frankly, the more "enterprise" the software is, the less pity I have for the customer(and yes, I have dealt with some pretty ghastly "enterprise" software, this isn't just theoretical sniping from the outside).
If Joe Consumer buys a shiny box-o-shrinkwrap, he basically has zero power. Most stores won't do anything more than replacement-in-kind, for physical defects in the medium or accompanying accessories, no returns. The inevitable nasty EULA will be hiding inside the shrinkwrap that confirms your acceptance of it. His software may (typically) be fairly cheap; but he has absolutely fuck-all power of redress if it doesn't work(or if the online component changes without notice in 3 months, or a mandatory update breaks things, or whatever).
If you are buying "enterprise" software, though, you are probably talking an individual contract(and a real contract, with suits and lawyers on both sides, not some contract of adhesion crap), contact with the vendor, and maybe even a stack of customizations. Why, pray tell, did you let the vendor get away with delivering whatever they wanted, without any assurance that it will do what they said it would? Why didn't the contract specify scary penalties if they fuck up?
Consumer protection law basically exists as a form of aggregate power to protect those who have basically none during each of the transactions they make day to day(and, secondarily, to assist the sellers, by providing a more uniform commercial environment, where certain standards are universal.) However, the closer in power to one another the buyer and the seller are, the less obviously necessary or desirable it is. Enterprises are big boys, why do they need the state to write their contracts for them? In the case of software, there is a lot of excellent "as-is, don't sue me" stuff floating around. It would hardly be helpful to open everybody who, as a service to others, gives anon read access to their CVS, to product defect lawsuits. If you want to sell it in shiny boxes at Gamestop or Office Depot, sure, the customer had better be able to at least return it. If it is "enterprise" software, the state should certainly hold both parties to their sides of the deal, per contract law; but surely they can figure it out between themselves.
Apparently, for the particular bit of number-crunching that they were doing, Cell curb-stomped x86. They then concluded that, if you want Cell, your only options are a few absurdly expensive specialist compute servers(IBM makes one, Sony at least has a model number, I think that there are one or two others) or cheap PS3s. Since, when the started, OtherOS was a standard, supported, option, the only "hacking" involved was the inconvenience of having to touch each machine to kick off the install.
Had their algorithm not suited Cell, the PS3 would have been an absurd choice. Since it did, though, it was actually pretty sensible(barring Sony's hard-to-predict action).
Does loading those twenty modules hurt you if, by virtue of their being in common use, there is some trivial automated way of doing it?
With the exception of people writing bare-metal assembly for microcontrollers or something, pretty much everyone who sits down to write some code has huge swaths of pre-written stuff loaded for them. The only difference is how much of that happens automatically, by default, and how much you see and do yourself.
If the science types happen to like python for some syntactic or structural or design reasons; but need a bunch of modules to make it do what they want, it isn't exactly rocket surgery to bundle them all together so a single "import science" does the job, or even hack together a slight variant of the python environment where that particular import is simply done silently by default.
Oh, obviously family history isn't 100% useful in all situations.
My point was simply that, when evaluating a "predictive" genetic test, "more predictive than nothing" is, in most cases, not actually the standard you should be holding it to. "More predictive than family history" is, except when that is unavailable, the much more appropriate(and much tougher) standard. Particularly for conditions where the genetics are not straightforward, watching your relatives run similar genetic code often tells you more than getting your own raw sequence.
The question, of course, is to what degree lifestyle is influenced by genetic factors...
Unless we are going to cling to the (intuitively satisfying; but rather silly) theory that humans have some sort of extra-material "free will" floating around in the aether, we pretty much have to concede that behavior has a biological basis. And, if something has a biological basis, the odds of it having a genetic and/or epigenetic component are pretty decent.
They gypsy certainly has better applied psych skills; but, for a great many genetic(or suspected but not yet fully elucidated) conditions, there is a way that is cheaper and more effective.
Family history.
With the exception of (not-nonexistent; but quite rare) conditions caused by a mutation or mutations that originated with you, not earlier in the line, or a fairly small number of well developed genetic tests, most of which you aren't going to get over the counter at CVS, you'll have a better chance of learning about the likely phenotypic consequences of your genes by looking at mommy and daddy, keeping their environment in mind(daddy's lung cancer probably doesn't count as "family history" if he was a chain-smoking asbestos miner, it probably does if he wasn't).
DNA sequencing has, certainly, gotten cheap enough that you might actually get a fairly accurate reading of a subset of your genome for a hundred bucks through the mail. However, I'd be quite surprised if, when it comes to predicting the consequences, which are what people actually care about, the method is going to outperform just looking at family history. In a lot of cases, the science simply isn't settled, at any price. Even where it is, you are going to be getting some mail merge algorithm, not a geneticist, or even a genetic counselor, for your hundred bucks.
I wonder what "other companies and individuals" are affiliated with Pathway Genomics?
Software industry has lowest per unit cost/fixed cost ratio in the world...