usenet was a store-and-forward, intermittently-connected hub-and-spoke hack, not really a net. IP, DNS and TCP marked the real beginnings of internet as internet.
isn't this a very conventional piece of hardware? is there some reason a particular distro would matter?
porting the kernel to a new, unrelated CPU would be interesting. porting a distro is never interesting, since, after all, distros are just some bozo typing "make" on a big pile of opensource packages.
I'd be happy with all GMO food being labeled, including all domesticated species. oh, you mean only "bad" GMO? please define. no genes from outside the species? good by nectarines! or are you going to define how "close" the source of the gene can be? what about crossover and other "natural" events that produce novel genes? what about transmission of genes by viruses?
it's simple: define the actual bad properties that make you afraid of GMO food, and regulate those, not the GMOness. does BT-GMO corn produce some sort of residue you can detect, and can make a case is harmful? regulate it. if someone put nut proteins in corn, allergic people would notice and be justified in complaining, or at least wanting useful labels.
otherwise, your anti-GMO labeling should consist of "NOT safe for paranoid luddites". maybe a custom symbol, like a gothic L struck through. for those who aren't down with this newfangled "language" stuff...
or at least one that's been asked a million times before.
the question is whether you want to use an appliance or a general-purpose device. an appliance is relatively fixed-format, and congruent with the concept of a walled garden, as well as revenue plans that make your vendors mbaciles happy. an appliance normally does not have user-serviceable parts, so the vendor is in control of the UX. appliances are fundamentally fixed-function devices, even if the vendor is able to update and even extend it, since they define what the fixed functions are.
being general-purpose is the opposite: it means that the owner really does own (control) the device, and can change its function, install software without regard to what the device vendor provides, approves or even knows about. PCs are fully general-purpose, since everything, from the roms to the OS to add-in cards can be replaced by the device owner.
so the question is really: to what extent is the vendor trying to draw a line across which the device owner cannot cross? no device is truely fixed-function, and even control-freak vendors like Apple provide _some_ affordances through which the device may be extended (hardware connectors, software app-stores). this has always been controversial, since any vendor restriction is at odds with our natural understanding of what "ownership" means (and even companies like Apple tend to show some variance in how locked-down and fixed-function their devices are - I can install Linux on an Apple laptop/desktop without much trouble, but they put a lot of effort into making it hard to root any of the smaller devices.)
I think it's time we get back to basics: when I buy a device, I should completely control it. any anti-rooting mechanisms should be illegal - the same way it would be illegal for a car vendor to specifically detect and sabotage my car if I put on third-party wheels. sure, make me click through a license-revoking agreement. but if you sell me something, and then take control of it out of my hands, you've committed fraud.
we should not allow this issue to become an opportunity for vendors to segment their market by selling a version for tinkerers and another for grandma. mostly, vendors have this impulse because their mbaciles want to lock in customers. instead of just selling devices, the popularity of which is subject to whim, the mbacilic approach is to sell service contracts as well, preferably multi-year, to ensure that customers can't get away without paying, even if the vendor's quality degrades. fixed-function devices are inherently like long-term contracts, since customers want upgrades and new apps, and since they're locked in, you can shove profitable advertising down their digital throats, or at least mine their usage/search behaviors.
oh, come on. buses have been dead for years (sata and pcie are great examples of the prevalence of point-to-point links). no reason we can't think of cachelines as packets (bigger than ATM packets were!). how about hypertransport and QPI?
so will homeowners start installing their own meter on the premises, to verify what the hackable powerco meter is recording? powercos are exactly the kind of non-competitive relics that believe in security-by-obscurity - that is, fiddle with the design until the level of fraud->outcry is low enough to ignore. it's not as if we don't have cheap, secure tech for exactly this kind of application.
why is it that arable land counts for so little in western (at least north american) societies? isn't it a bit of a shame we devote so much land to lawns, rather than something productive? yes, I know: the crops that could be grown are not worth the cost of maintaining them. but why is that? is food too cheap, or labor too expensive? is it a distortion caused by exchange rates?
I wouldn't mind a part-cave house, especially since a cave would presumably be near some sort of elevation (hillside, escarpment). I think everyone values a bit of a view, some sunlit rooms, etc. but one-story houses on flat plots of land are pretty boring once they scale past a cottage.
I love it when some suit mouths off to customers about how profitable they are. guys - you're great... for my wallet!
I wonder how much Sprint makes selling the customer list - after all, someone who chokes down Apple's margins is likely to buy other stuff that's well-marketed.
I'm astonished and disappointed how many of these comments are along the lines of "no way, because of hackers".
yes, online voting would be exposed to online attempts to subvert, but it's absurd to claim that this is a fatal flaw. what are you guys, technophobes? using RoR as an analogy is just asinine, since RoR is a huge framework intended to ease programming effort. if you want a secure site, you won't start with something that's vastly general, and assume it can be made secure, but rather first minimize your attack surface.
online voting is just one way we need to improve the whole electoral process. having polling stations online is another, since even if a vote is cast in person, using paper, we still need a realtime mechanism to prevent multiple voting under the same ID. paper trails are great. crypto receipts are great. to what degree can we trust votes cast from insecure terminals? depends on the process: for instance, suppose I can register my vote at any time up to the deadline, and can check to make sure my vote is still the same as when I first cast it. or suppose casting an online vote results in a verification by a second mechanism (phone, probably). none of this is hard, though it does need careful forethought and vetting.
transparency requires use of open source. I hope we're past the point where people claim that source code inspectability is a security risk...
the public is being ripped off by the incompetence and corruption of the PTO. instead of _promoting_ progress, patents are now just hunting licenses for bloodthirsty lawyers. (which is why you don't "win" a patent - it's merely something you use for extortion until someone calls your bluff and you have to defend it in court.)
the internet seems to be just an excuse for law enforcement to be lazy: what's wrong with limiting them to physical surveillance? yeah, it's harder, but from the citizen's side, that's a _good_ thing.
-- IP protection is not a natural right, but a socially-granted monopoly. -- the purpose of IP protection is to promote progress. -- the IP must be specific, novel and non-obvious.
IP is not about protecting profits, put actually about ensuring that your IP is known to your competitors, enabling them to make it better. thus encouraging them with profits for having improved on your idea!
Apple is an amoral, profit-driven company. who's surprised? it's what western (especially American, MBA-infested) business culture is all about.
what's surprising is that confronting this fact creates a surprising amount of cognitive dissonance among those who hold a romanticised image of Apple (free-thinking, genius designer/artistes who are the underdog of the computer and CE worlds. wrong in so many ways...)
what makes all this coverage more piquant is the fact that the harsh labor practices constitute a remarkably small portion of the BOM for Apple products. for instance, ~10% of the cost of an iPad2 is the enclosure (ie, foxcon aluminum polishing): http://www.isuppli.com/PublishingImages/Press%20Releases/2011-03-12_iPad2_BOM.png
these numbers show apple's markup is about 120% over cost; if they doubled what they pay for the enclosure, their markup would drop to a mere 100%.
have you actually looked at the numbers? AWS, for instance, has a phenomenally high profit margin (and PhB's have the mistaken impression that running your own farm is expensive.) it's not some kind of close thing (oh, they get power 10% cheaper, or their machinerooms run at a PUE of 1.15 whereas ours are 1.4) - cloud providers are charging O(10x) more per compute-unit than cost.
this doesn't really seem worth noting, since many companies have many licensing agreements. it's just a way for business people to do some risk mitigation, since the patent system is so broken as to make actually producing a product a game of russian roulette. this kind of licensing is just a way to trade a little money for the guarantee that there are fewer bullets in the revolver. sure, it's extortion, but so is most risk mitigation.
unfortunately, the idiotic MBA-driven approach that gives rise to this is amoral, so such companies face the moral hazard of having engaged in risk mitigation. not only is it not their money, but can be spun to the public as good stewardship. when in reality, it's just cowardice and unwillingness to stand up and make IP reform happen NOW.
overclocking was always a niche thing - it had some support from game-the-system types, but was mostly tweakers (yes, akin to meth.) it had an aspect of public self-abuse, sort of like getting a gang tatoo on your neck or face. and now, it's totally mainstream and vapid - the elements of display like leds on your fans and clear side panels are no more than a sort of meth-trailer-trash-chic.
though it's interesting that Intel hasn't done much in the way of SoC until recently - they've traditionally kept the CPU quite separate from support chips. from a fab perspective, this makes a certain amount of sense, since the cpu needs transistors that perform quite differently (voltage, drive, frequency, etc) and the limited SoC integration present in, say, sandybridge, is mixed in performance (good memory and pcie controllers, mediocre gpu).
the real question with intel mid SoCs is whether they can drive down power to ARM levels. my smartphone has a 6.11 W-hour battery, and it better last (idle) for at least two days, or one day with reasonable activity. idle-on-net needs to be 1W...
unfortunately, DC studies like this tend to argue their case against a strawman. namely, it's now quite easy to find servers with 94% efficient PSUs. and many datacenters can operate without dual-conversion UPSs (consider Goggle's design of simply sticking a small battery onto each server.)
nah, batteries do not add significant cost. the real issue is that it adds failure modes and thus support costs, which is ugly in a high-volume consumer product.
Slashdot Mirror
usenet was a store-and-forward, intermittently-connected hub-and-spoke hack, not really a net. IP, DNS and TCP marked the real beginnings of internet as internet.
hah. and since we're in a pedantic mood, they're actually *forward* slashes.
sorry, why aren't all applications published in full, as a matter of course?
isn't this a very conventional piece of hardware? is there some reason a particular distro would matter?
porting the kernel to a new, unrelated CPU would be interesting. porting a distro is never interesting, since, after all, distros are just some bozo typing "make" on a big pile of opensource packages.
I'd be happy with all GMO food being labeled, including all domesticated species. oh, you mean only "bad" GMO? please define. no genes from outside the species? good by nectarines! or are you going to define how "close" the source of the gene can be? what about crossover and other "natural" events that produce novel genes? what about transmission of genes by viruses?
it's simple: define the actual bad properties that make you afraid of GMO food, and regulate those, not the GMOness. does BT-GMO corn produce some sort of residue you can detect, and can make a case is harmful? regulate it. if someone put nut proteins in corn, allergic people would notice and be justified in complaining, or at least wanting useful labels.
otherwise, your anti-GMO labeling should consist of "NOT safe for paranoid luddites". maybe a custom symbol, like a gothic L struck through. for those who aren't down with this newfangled "language" stuff...
or at least one that's been asked a million times before.
the question is whether you want to use an appliance or a general-purpose device. an appliance is relatively fixed-format, and congruent with the concept of a walled garden, as well as revenue plans that make your vendors mbaciles happy. an appliance normally does not have user-serviceable parts, so the vendor is in control of the UX. appliances are fundamentally fixed-function devices, even if the vendor is able to update and even extend it, since they define what the fixed functions are.
being general-purpose is the opposite: it means that the owner really does own (control) the device, and can change its function, install software without regard to what the device vendor provides, approves or even knows about. PCs are fully general-purpose, since everything, from the roms to the OS to add-in cards can be replaced by the device owner.
so the question is really: to what extent is the vendor trying to draw a line across which the device owner cannot cross? no device is truely fixed-function, and even control-freak vendors like Apple provide _some_ affordances through which the device may be extended (hardware connectors, software app-stores). this has always been controversial, since any vendor restriction is at odds with our natural understanding of what "ownership" means (and even companies like Apple tend to show some variance in how locked-down and fixed-function their devices are - I can install Linux on an Apple laptop/desktop without much trouble, but they put a lot of effort into making it hard to root any of the smaller devices.)
I think it's time we get back to basics: when I buy a device, I should completely control it. any anti-rooting mechanisms should be illegal - the same way it would be illegal for a car vendor to specifically detect and sabotage my car if I put on third-party wheels. sure, make me click through a license-revoking agreement. but if you sell me something, and then take control of it out of my hands, you've committed fraud.
we should not allow this issue to become an opportunity for vendors to segment their market by selling a version for tinkerers and another for grandma. mostly, vendors have this impulse because their mbaciles want to lock in customers. instead of just selling devices, the popularity of which is subject to whim, the mbacilic approach is to sell service contracts as well, preferably multi-year, to ensure that customers can't get away without paying, even if the vendor's quality degrades. fixed-function devices are inherently like long-term contracts, since customers want upgrades and new apps, and since they're locked in, you can shove profitable advertising down their digital throats, or at least mine their usage/search behaviors.
oh, come on. buses have been dead for years (sata and pcie are great examples of the prevalence of point-to-point links). no reason we can't think of cachelines as packets (bigger than ATM packets were!). how about hypertransport and QPI?
so will homeowners start installing their own meter on the premises, to verify what the hackable powerco meter is recording? powercos are exactly the kind of non-competitive relics that believe in security-by-obscurity - that is, fiddle with the design until the level of fraud->outcry is low enough to ignore. it's not as if we don't have cheap, secure tech for exactly this kind of application.
why is it that arable land counts for so little in western (at least north american) societies? isn't it a bit of a shame we devote so much land to lawns, rather than something productive? yes, I know: the crops that could be grown are not worth the cost of maintaining them. but why is that? is food too cheap, or labor too expensive? is it a distortion caused by exchange rates?
I wouldn't mind a part-cave house, especially since a cave would presumably be near some sort of elevation (hillside, escarpment). I think everyone values a bit of a view, some sunlit rooms, etc. but one-story houses on flat plots of land are pretty boring once they scale past a cottage.
I love it when some suit mouths off to customers about how profitable they are. guys - you're great... for my wallet!
I wonder how much Sprint makes selling the customer list - after all, someone who chokes down Apple's margins is likely to buy other stuff that's well-marketed.
I'm astonished and disappointed how many of these comments are along the lines of "no way, because of hackers".
yes, online voting would be exposed to online attempts to subvert, but it's absurd to claim that this is a fatal flaw. what are you guys, technophobes? using RoR as an analogy is just asinine, since RoR is a huge framework intended to ease programming effort. if you want a secure site, you won't start with something that's vastly general, and assume it can be made secure, but rather first minimize your attack surface.
online voting is just one way we need to improve the whole electoral process. having polling stations online is another, since even if a vote is cast in person, using paper, we still need a realtime mechanism to prevent multiple voting under the same ID. paper trails are great. crypto receipts are great. to what degree can we trust votes cast from insecure terminals? depends on the process: for instance, suppose I can register my vote at any time up to the deadline, and can check to make sure my vote is still the same as when I first cast it. or suppose casting an online vote results in a verification by a second mechanism (phone, probably). none of this is hard, though it does need careful forethought and vetting.
transparency requires use of open source. I hope we're past the point where people claim that source code inspectability is a security risk...
the public is being ripped off by the incompetence and corruption of the PTO. instead of _promoting_ progress, patents are now just hunting licenses for bloodthirsty lawyers. (which is why you don't "win" a patent - it's merely something you use for extortion until someone calls your bluff and you have to defend it in court.)
the only interesting thing about it is that apple has cut their profit margin to compete with android.
what could possibly go wrong?
the internet seems to be just an excuse for law enforcement to be lazy: what's wrong with limiting them to physical surveillance? yeah, it's harder, but from the citizen's side, that's a _good_ thing.
-- IP protection is not a natural right, but a socially-granted monopoly.
-- the purpose of IP protection is to promote progress.
-- the IP must be specific, novel and non-obvious.
IP is not about protecting profits, put actually about ensuring that your IP is known to your competitors, enabling them to make it better. thus encouraging them with profits for having improved on your idea!
Apple is an amoral, profit-driven company. who's surprised? it's what western (especially American, MBA-infested) business culture is all about.
what's surprising is that confronting this fact creates a surprising amount of cognitive dissonance among those who hold a romanticised image of Apple (free-thinking, genius designer/artistes who are the underdog of the computer and CE worlds. wrong in so many ways...)
what makes all this coverage more piquant is the fact that the harsh labor practices constitute a remarkably small portion of the BOM for Apple products. for instance, ~10% of the cost of an iPad2 is the enclosure (ie, foxcon aluminum polishing):
http://www.isuppli.com/PublishingImages/Press%20Releases/2011-03-12_iPad2_BOM.png
these numbers show apple's markup is about 120% over cost; if they doubled what they pay for the enclosure, their markup would drop to a mere 100%.
how can that be? you really only have one project going at a time, and the whole project becomes renderable at once?
have you actually looked at the numbers? AWS, for instance, has a phenomenally high profit margin (and PhB's have the mistaken impression that running your own farm is expensive.) it's not some kind of close thing (oh, they get power 10% cheaper, or their machinerooms run at a PUE of 1.15 whereas ours are 1.4) - cloud providers are charging O(10x) more per compute-unit than cost.
this doesn't really seem worth noting, since many companies have many licensing agreements. it's just a way for business people to do some risk mitigation, since the patent system is so broken as to make actually producing a product a game of russian roulette. this kind of licensing is just a way to trade a little money for the guarantee that there are fewer bullets in the revolver. sure, it's extortion, but so is most risk mitigation.
unfortunately, the idiotic MBA-driven approach that gives rise to this is amoral, so such companies face the moral hazard of having engaged in risk mitigation. not only is it not their money, but can be spun to the public as good stewardship. when in reality, it's just cowardice and unwillingness to stand up and make IP reform happen NOW.
overclocking was always a niche thing - it had some support from game-the-system types, but was mostly tweakers (yes, akin to meth.) it had an aspect of public self-abuse, sort of like getting a gang tatoo on your neck or face. and now, it's totally mainstream and vapid - the elements of display like leds on your fans and clear side panels are no more than a sort of meth-trailer-trash-chic.
though it's interesting that Intel hasn't done much in the way of SoC until recently - they've traditionally kept the CPU quite separate from support chips. from a fab perspective, this makes a certain amount of sense, since the cpu needs transistors that perform quite differently (voltage, drive, frequency, etc) and the limited SoC integration present in, say, sandybridge, is mixed in performance (good memory and pcie controllers, mediocre gpu).
the real question with intel mid SoCs is whether they can drive down power to ARM levels. my smartphone has a 6.11 W-hour battery, and it better last (idle) for at least two days, or one day with reasonable activity. idle-on-net needs to be 1W...
unfortunately, DC studies like this tend to argue their case against a strawman. namely, it's now quite easy to find servers with 94% efficient PSUs. and many datacenters can operate without dual-conversion UPSs (consider Goggle's design of simply sticking a small battery onto each server.)
alas, PSUs have been at 94% for several years now.
also, what kind of batteries do you think are _in_ a UPS?
anyone can do low-current DC. you want a fairly specialized, fully-qualified electrician doing high-current DC, which is what we're talking about.
nah, batteries do not add significant cost. the real issue is that it adds failure modes and thus support costs, which is ugly in a high-volume consumer product.
so far, all bobcat-based chips have been made at TSMC, haven't they? so is this really news?