Honestly, it largely depends on how intel decides to price them:
If they continue with their recent trend of fairly aggressive pricing on 'real' CoreWhatever ULV chips, of which even the weakest 'celeron' branded ones are superior to the atom, and crazy optimistic pricing and deliberate crippling of Atom parts and boards, it will be hard to get worked up about them.
If, on the other hand, Intel is genuinely getting a bit rattled by some of the fancier ARM gear, and chooses to price the Atom more in line with its die size(Atoms are not fast; but they are tiny compared to Intel's punchier designs, rather than their desire to spare the lower end of the Core line, I'll take several.
Atoms aren't screaming fast, or milliwatt power sippers; but they are excellent in various appliance applications: In network attached storage, for instance, the superiority of an atom board to some horrid little SoC is downright alarming. The performance of a nasty little plastic router vs. an Atom running monowall or similar is equally unfair.
If the price is right, there are plenty of places where a bit more punch, and the ability to use 100% normal PC linux experience, are worth the modest additional power consumption. If they keep pricing them in order to save the Core line, they get a lot less exciting...
Aren't sociopaths supposed to be glib, charming, and expertly manipulative? This guy seems to have the 'narcissistic disregard for others' and 'delusional grandiosity' covered; but the entire situation is an utter clusterfuck of how to absolutely and unbelievably fail a fairly trivial customer-service brush-off task.
I'm certainly all for grinding him into the dirt, scum aren't exactly a limited resource; but he doesn't seem to fit the pop-psych sociopath profile all that well....
As a professional, you're not entitled to having bad days.
This goes double because email isn't exactly an immediate medium. It is utterly trivial to delay responding to a given email(or even fake an out-of-office if you simply must have more time).
In situations where you can't escape and you have somebody physically in your face, right now, some risk of snapping inevitably exists. Some people bear up better than others; but it can happen. Flipping out over email, though, is flipping out even after you've had the benefit of choosing how much time you need to calm down, drafting as many revisions as you need, and knowing for certain that this text is on the record... Everybody has a breaking point, and a sufficiently bad situation can push you to it; but email is far lower pressure than most situations.
Given his reasoning in favor of ethical behavior "must be nice to people, because they might have friends more powerful than me and mine", he's an ethical lost cause. You Just Don't turn your back on somebody who thinks that the only reason not to stab you is because they might be punished.
The best we can hope for is that this unexpected blowup will inspire a degree of caution verging on paranoia, and he'll be rendered relatively innocuous by fear of possible punishment. Ideally, somebody should introduce him to a particularly nasty fire-and-brimstone religion. If somebody is actually so depraved that they act only through fear of power, the notion that power that could crush them like a bug is watching at all times can be quite useful....
I, for one, can only wonder how he managed to get married and spawn.
He's sort of an interesting case: He has a boorish disregard for others, and a sense of grandiosity bordering on full-fledged delusion; but he shows none of the low, animal cunning that one would expect from somebody who has managed to worm their way into an actual position of responsiblity and contact with the public....
The world is full of boors, narcissists, assholes, and general scum; but the ones that make it to positions of visibility usually have some sort of compensatory traits(many of them also vices; but still). This guy doesn't seem to. No glib lying, no superficial charm, just some really hollow name-dropping and chest thumping about unit sales, in an email exchange prompted by the fact that their supply chain is sucking right now...
That is what befuddles me. Does this guy have charisma indistinguishable from magic in person? Was he 'roid raging when he wrote those emails? Are the standards of freelance PR flacks so pitifully low that they can't even afford unemployed English majors who have at least mastered sentence construction?
It doesn't surprise me that he is a bad person, that is quite common, especially among marketing weasels. What confounds me is that he is so utterly bad at being a bad person. This situation seems like it would have been smooth-talk 101 to walk away from at, at most, the cost of a $10 credit to an enthusiastic customer. Instead, he managed to score frontpage mockery on the who's-who of gaming websites, make some n00b mistakes on twitter, sockpuppet from an email address linked to some hilarious posts about his attempts at muscle building, and generally snatch defeat from the jaws of victory, the jaws of the bystanders, and anywhere else it could be found. Where do they get people like that?
Ah, grasshopper, let me educate you in the official party line concerning consumer product safety or product labelling regulations:
Situation #1: The state proposes regulating certain aspects of the health, safety, purity, and/or, potency of some product. The relevant industry's lobbyists, backed by general purpose heavy guns like the USCoC and AEI, howl in protest "Heavy-handed, job-killing regulation, unsupported by Sound Science(tm), will destroy the industry! Consumer Choice! Let the customer decide what they want!"
Situation #2: The state calls their bluff: "Ok, fuckers, let's let the consumer decide, everybody label their product according to what it is, and let the most popular player win!" The relevant industry's lobbyists, backed by general purpose heavy guns like the USCoC and AEI, howl in protest "Your burdensome labelling requirements will cost eleventy billion dollars and 4254535452 american jobs to comply with! They will only confuse consumers, who do not understand what they want. We demand that labelling not only be optional, people who label their products with things that make us look bad, like 'contains no recombinant bovine growth hormone' or 'non-GMO' be legally forced to abandon the practice!"
It makes perfect sense, if you do your absolute best to think in very short bursts...
Not successfully, as yet.... The theory was that the antifreeze proteins used by the arctic flounder to resist cold damage in its rather hostile environment would produce a tomato resistant to frosts and cold storage.
Splicing the gene in worked just fine. However, the product wasn't significantly better, as a tomato, and the PR was bad.
Good old Green Fluorescent Protein, a jellyfish derivative, has been spliced into just about anything and everything somebody in a lab coat has cared to hold still for 10 minutes; but largely as a proof-of-technique or imaging agent, it has no obvious value for food crops.
Our experience to the present suggests that attempting to grab useful animal traits and shove them into plants(I, for one, welcome the tomeato with enthusiasm!) is harder than naive speculation would suggest; but that there is no magic barrier to splicing animal genes into plants, other animals, bacterial, etc.
I think you mean, "I don't see why this is a bigger problem than cracking other people's networks." It's obviously a problem for people to be breaking into others' systems; it's just that you're saying that the USCoC is nothing special. Right?
It sounds, to me, like a combination of "Remember, no matter how hard they work to imply otherwise, the 'US Chamber of Commerce' is just a private lobbying group, not part of the government" and "Awww, so sad, couldn't have happened to a nicer bunch of people."
I'm not a huge fan of internet security being a total free-fire zone; but the USCoC are pretty much slime, have been known to use dirty tricks themselves, so it just isn't feasible to exude too many tears for them.
My sources tell me that evildoers possess advanced ICMP technology that would allow a pirate to verify whether or not a forbidden server is active, among other criminal surveillance, from anywhere in the Homeland!
While Firefox's marketshare has been suffering slightly, I can't imagine that the per-seat value of being the default search engine has changed particularly, and FF is probably the competitor from which Google gains the most: FF reliably agrees with them on most major issues, has no significant strength to threaten Google's actually profitable ventures, and no(well, almost no, you could build FF-only XUL webapps; but nobody does) competing application environment.
Microsoft has a browser, a search engine, win32, and silverlight, so they aren't exactly somebody that Google wants gaining ground, Apple has impressive control of certain high margin markets, and an iron grip on their mobile devices. Firefox has a browser. Unless Google has some aesthetic reason to crush anything it can, and risk the wrath of the antitrust guys, Firefox's existence is somewhere between 'harmless' and 'downright convenient'.
I'd be the first to agree that progress in that area has been surprisingly uneven, with a lot of "this isn't an 'advance' this is just a cheaper version of what we were doing in VMS, damn it" and even outright regressions, as with the apparent death-spiral of OpenSolaris without btrfs being all there yet.
My point was just that, beyond the level of reliability necessary to attract OEM contracts(ie. acceptably low number die in the first, second, and 3rd years, so warranty support doesn't bleed the OEM dry) and the level of reliability necessary to keep RAID arrays and SANs from having disks die faster than they can be rebuilt, there is more or less zero additional willingness to pay for reliability at the level of the single drive.
All the additional willingness to buy reliability(and there is a lot of it) is absorbed by a combination of RAID chipset vendors, SAN appliance guys, various flavors of backup hardware and software(with emphasis on things like dedupe, site failover, and regulatory compliance on the high end, and 'being easy enough that I might actually use it' on the consumer end).
On the speed side, the picture is similar, there is a great deal of money in it for you if you can deliver higher throughput on people's fancy databases and very high traffic websites; but the amount of 'speed premium' money that is paid out at the drive level is probably declining. SSDs make the IOPs difference between a 5400k and a 15000k RPM drive look like a rounding error, and pools of memcache servers or high end storage appliances that will expose a half TB of RAM as a fibre channel device, and the like, are the ones soaking up the real money to be made on speed. Even when HDDs are used, the fancy controller required to not bottleneck a shelf of 15Ks will cost a nontrivial percentage of the drives that populate it.
I don't mean to imply that these markets are necessarily going in the right direction, just that the HDD guys are very poorly placed to grab any of the money on the table. All the solutions either call for no HDDs at all, or are of the "Send us a bunch of dumb block devices. If they are 50% larger and cost half as much as last year, that would be good." There just isn't anything that you can do, within the confines of the single device(with mechanical HDD tech, and the mechanical device companies don't have silicon fab background, so that doesn't help them) that is all that interesting, or worth a premium, except for bumping capacity and lowering price.
You still need a warrant if the surveillance is directed at an individual. And if it's just patrolling, how is that any different than a cop walking his beat?
Surveillance technologies bring two main changes to the table, even when otherwise analogous to some prior method:
1. Economics: There is no legal problem with having cops walking 100% of the legally public beats 100% of the time. Economically, though, there just aren't enough cops to do that. In practice, one of the major protections from the state historically enjoyed by most people is not law; but simple lack of resources. It may be legal to have a cop follow you on a public road, and determine your route; but that cop isn't cheap, so you'll have to attract some suspicion first. Slapping a $100, reusable, magnetic GPS bug on your car, on the other hand, is overwhelmingly cheaper than having a $50,000/yr cop following you. Even if the two are analogous, the level of protection enjoyed in one case is far lower than in the other.
2. Retention: Humans, by necessity, have lousy memories. Our eyes just slide right over mundane happenings and they fall away almost immediately. Storage of electronic surveillance data, on the other hand, is cheap and getting cheaper(and easier to automatically search). Trying to track the routes of all motorists in a city based on data from the beat cops would be essentially impossible. Doing the same from an equivalent number of license-plate cameras? Hard; but tractable.
The crux of the matter is that, as cost decreases and retention increases, 'just patrolling' and 'surveillance directed at an individual' stop being distinct categories: the agents that are 'just patrolling' gather and retain enough data that (proactively or retroactively) turning that patrol into surveillance is essentially just a matter of doing the DB lookup. We haven't reached that point yet; but basically any advance in the cost or capability of automated surveillance technology moves us closer. Patrolling and targeted surveillance aren't fundamentally different, they are different because human agents are really bad at patrolling, and have to be given quite different orders if you want them to get useful data on a specific target. If an agent is good at patrolling, all people that pass within its view are effectively surveilled...
How'd you like to be the patent examiner who has to strike down that application because IBM's own work in tracking people from ghetto to oven constituted prior art?
Surely they could just tap one of the statistics professors at hamburger university. I imagine that describing the origin of a given nugget in the n-dimensional probability space encompassing all the possible distributions of the various animals on the line at production time would be rather like describing the position of an electron in space: You couldn't actually say where an individual nugget falls; but with knowledge of the production process and the input animals you could model the statistical distribution of where each nugget is more or less likely to fall...
We in the agriculture and food processing industry take customer satisfaction extremely seriously. For a small additional packaging fee, we would be delighted to ensure that your food has been locally produced in the location of your choice and certified to whatever standard you desire by whatever certification bodies you trust most. Our graphics department may require 8-10 additional business days for certification logos not already in our library, and 4-colour printing is extra.
To suit the requirements of today's environmentally sensitive customer, we are proud to label all our products as being sustainably derived from non-endangered species, or endangered species whose tissues are indistinguishable by any test likely to be employed by the customs agents of your jurisdiction.
For discrete cuts of meat, the labeling should be simple enough; but some of the more, er, 'waste minimizing' meat products are going to get seriously complex.
The composition of a given hamburger would probably have to be given as a joint probability density function across a set of hundreds or thousands of animals or something similarly messy. That would give label-readers something to ponder...
Apple designs have a rather long history(Apple III-present, somewhat intermittently) of running everything right near thermal redline in order to keep it quiet. On the plus side, it is pretty impressive how quiet some of their hardware is, especially given its chassis size(quiet is easy when you can just have a row of low-speed 120mm fans blowing over everything, much harder when you have a maze of teeny little air channels and speed controlled blowers and stuff); but it gets damn toasty in there...
In such a brutally commodified market as HDDs, I suspect that all the companies are run by bean counters and the visionaries are dead.
Because they are so cheap, per GB, mechanical HDDs aren't going anywhere for a decent chunk of the future; but they've been boxed in such that there isn't any noticeable room for 'visonary' development:
1. Performance? If you want that, you'll be talking to a totally different company with a background in semiconductors, either Flash or DRAM, depending on how much money you are made of. Nothing stopping the HDD people from selling rebadges; but rebadging is not exactly a visionary(or terribly high margin) business.
2. Reliability? Because servicing a warranty request isn't inexpensive(phone drones, fedex, etc.) anybody who can't deliver drives with a low failure rate during standard PC OEM warranty periods is going to find their sales limited; but reliability at the drive level isn't actually worth very much: The value of the world's spinny disks is peanuts compared to the value of the data on them. Most of the reliability money and R&D is going into RAID, advanced filesystems, various automated redundancy and backup solutions, etc. Again, nothing stops the HDD guys from selling rebadge RAID controllers or cloud backup services; but rebadging is not exciting.
3. Features? If it doesn't just drop in and play nice with the SATA/SAS controllers of the world, including the legacy and currently shipping ones, it's a dud. If it has some cool feature that is supported only by your proprietary utility, on controllers that directly pass the necessary nonstandard commands, it isn't going to be wildly useful. If it achieves sufficiently broad adoption that OS and HDD controller support starts coming standard, it is no longer a unique competitive advantage...
Cynically, there is also the fact that even people buying on the basis of desire for mechanical reliability don't have access to very good information: hardware and firmware revs change constantly, sometimes with a change in model designation, often not, some designs turn out to be workhorses, some are deathstars, some batches are bad, some aren't. Everybody has an emotional position on reliability, based largely on which brands failed them in the recent past; but unless they are buying in serious volume and somewhat behind the tech curve, data about the past are largely obsolete.
As a rough heuristic, any state wealth-redistribution program that doesn't have packs of lobbyists and AEI economists shrieking about socialism, communism, and class warfare is very likely converting your tax dollars into somebody's shareholder value more or less by design.
Wealth-redistribution plans that do have such a pack are somewhat less clear. They are less likely to be an overt screwjob; but their efficiency or efficacy may still be miserable.
Fermilab Director Pier Oddone also managed to mention that the Fermilab gift shop is now stocking a variety of novelty plush quarks, 'Experi-mint' brand candies, and similar knicknacks before breaking down and sobbing something about 'Is this why I made it through my postdoc?'
While it does nothing against the possibility that the allegedly US cables are, in fact, forgeries by some third party, Kazakhstan seems like an odd place for the US to risk local relations in order to smear Russia(it's not as though there aren't plenty of other options, of varying degrees of truth, available). Our options for things like supply routes and airbases in the region are tepid to say the least and Kazakhstan is one of the countries we've been trying to butter up for that sort of access, along with their delicious petroleum.
It would seem extraordinarily stupid, even if one ignores any minor hindrances like 'ethics' or 'truth', to run an anti-Russia smear campaign at the expense of relations with a strategically located (and, so far, fairly cooperative in our 'perpetual war on terror' plans) country.
Don't worry. Keratin is durable as hell by biological standards, so Dear Leader's amazing hair should still be in excellent condition decades from now...
I, for one, would like to thank the citizens of Texas for subsidizing my semiconductor purchases!
Yeah, I'm sure Texas just gives those incentives because they're what, stupid? Altruistic? No - they and the rest of the world that gives these incentives understand that in the end it comes out as a big win for their local economy.
I certainly wouldn't accuse them of altruism, and only some of them of stupidity:
'Incentives' are, indeed, a very good way of obtaining a package of immediate and visible benefits in exchange for a package of costs that is generally relatively small up front and packs most of its punch in indirect or time-deferred costs. Politically, it's a great deal. Economically, the suggestions that it is a good idea are equivocal at best, negative at worst.
Dell is, arguably, more of a logistics and integration company than a tech company(which isn't necessarily a bad thing, they are pretty decent at it, and somebody has to do it)...
They are pretty good at providing a one-stop-shop for a variety of Intel and AMD silicon, with supporting chips from a number of other vendors, mounted on a standardized set of boards from a few pacific rim OEM shops, and stuffed into plastic boxes in Mexico according to your order. Juggling that worldwide logistics effort is no mean feat; but they don't mix much in the way of dell technology into the sauce. It's like fedex with driver updates...
Slashdot Mirror
Honestly, it largely depends on how intel decides to price them:
If they continue with their recent trend of fairly aggressive pricing on 'real' CoreWhatever ULV chips, of which even the weakest 'celeron' branded ones are superior to the atom, and crazy optimistic pricing and deliberate crippling of Atom parts and boards, it will be hard to get worked up about them.
If, on the other hand, Intel is genuinely getting a bit rattled by some of the fancier ARM gear, and chooses to price the Atom more in line with its die size(Atoms are not fast; but they are tiny compared to Intel's punchier designs, rather than their desire to spare the lower end of the Core line, I'll take several.
Atoms aren't screaming fast, or milliwatt power sippers; but they are excellent in various appliance applications: In network attached storage, for instance, the superiority of an atom board to some horrid little SoC is downright alarming. The performance of a nasty little plastic router vs. an Atom running monowall or similar is equally unfair.
If the price is right, there are plenty of places where a bit more punch, and the ability to use 100% normal PC linux experience, are worth the modest additional power consumption. If they keep pricing them in order to save the Core line, they get a lot less exciting...
Aren't sociopaths supposed to be glib, charming, and expertly manipulative? This guy seems to have the 'narcissistic disregard for others' and 'delusional grandiosity' covered; but the entire situation is an utter clusterfuck of how to absolutely and unbelievably fail a fairly trivial customer-service brush-off task.
I'm certainly all for grinding him into the dirt, scum aren't exactly a limited resource; but he doesn't seem to fit the pop-psych sociopath profile all that well....
As a professional, you're not entitled to having bad days.
This goes double because email isn't exactly an immediate medium. It is utterly trivial to delay responding to a given email(or even fake an out-of-office if you simply must have more time).
In situations where you can't escape and you have somebody physically in your face, right now, some risk of snapping inevitably exists. Some people bear up better than others; but it can happen. Flipping out over email, though, is flipping out even after you've had the benefit of choosing how much time you need to calm down, drafting as many revisions as you need, and knowing for certain that this text is on the record... Everybody has a breaking point, and a sufficiently bad situation can push you to it; but email is far lower pressure than most situations.
Given his reasoning in favor of ethical behavior "must be nice to people, because they might have friends more powerful than me and mine", he's an ethical lost cause. You Just Don't turn your back on somebody who thinks that the only reason not to stab you is because they might be punished.
The best we can hope for is that this unexpected blowup will inspire a degree of caution verging on paranoia, and he'll be rendered relatively innocuous by fear of possible punishment. Ideally, somebody should introduce him to a particularly nasty fire-and-brimstone religion. If somebody is actually so depraved that they act only through fear of power, the notion that power that could crush them like a bug is watching at all times can be quite useful....
I, for one, can only wonder how he managed to get married and spawn.
He's sort of an interesting case: He has a boorish disregard for others, and a sense of grandiosity bordering on full-fledged delusion; but he shows none of the low, animal cunning that one would expect from somebody who has managed to worm their way into an actual position of responsiblity and contact with the public....
The world is full of boors, narcissists, assholes, and general scum; but the ones that make it to positions of visibility usually have some sort of compensatory traits(many of them also vices; but still). This guy doesn't seem to. No glib lying, no superficial charm, just some really hollow name-dropping and chest thumping about unit sales, in an email exchange prompted by the fact that their supply chain is sucking right now...
That is what befuddles me. Does this guy have charisma indistinguishable from magic in person? Was he 'roid raging when he wrote those emails? Are the standards of freelance PR flacks so pitifully low that they can't even afford unemployed English majors who have at least mastered sentence construction?
It doesn't surprise me that he is a bad person, that is quite common, especially among marketing weasels. What confounds me is that he is so utterly bad at being a bad person. This situation seems like it would have been smooth-talk 101 to walk away from at, at most, the cost of a $10 credit to an enthusiastic customer. Instead, he managed to score frontpage mockery on the who's-who of gaming websites, make some n00b mistakes on twitter, sockpuppet from an email address linked to some hilarious posts about his attempts at muscle building, and generally snatch defeat from the jaws of victory, the jaws of the bystanders, and anywhere else it could be found. Where do they get people like that?
Ah, grasshopper, let me educate you in the official party line concerning consumer product safety or product labelling regulations:
Situation #1: The state proposes regulating certain aspects of the health, safety, purity, and/or, potency of some product. The relevant industry's lobbyists, backed by general purpose heavy guns like the USCoC and AEI, howl in protest "Heavy-handed, job-killing regulation, unsupported by Sound Science(tm), will destroy the industry! Consumer Choice! Let the customer decide what they want!"
Situation #2: The state calls their bluff: "Ok, fuckers, let's let the consumer decide, everybody label their product according to what it is, and let the most popular player win!" The relevant industry's lobbyists, backed by general purpose heavy guns like the USCoC and AEI, howl in protest "Your burdensome labelling requirements will cost eleventy billion dollars and 4254535452 american jobs to comply with! They will only confuse consumers, who do not understand what they want. We demand that labelling not only be optional, people who label their products with things that make us look bad, like 'contains no recombinant bovine growth hormone' or 'non-GMO' be legally forced to abandon the practice!"
It makes perfect sense, if you do your absolute best to think in very short bursts...
Not successfully, as yet.... The theory was that the antifreeze proteins used by the arctic flounder to resist cold damage in its rather hostile environment would produce a tomato resistant to frosts and cold storage.
Splicing the gene in worked just fine. However, the product wasn't significantly better, as a tomato, and the PR was bad.
Good old Green Fluorescent Protein, a jellyfish derivative, has been spliced into just about anything and everything somebody in a lab coat has cared to hold still for 10 minutes; but largely as a proof-of-technique or imaging agent, it has no obvious value for food crops.
Our experience to the present suggests that attempting to grab useful animal traits and shove them into plants(I, for one, welcome the tomeato with enthusiasm!) is harder than naive speculation would suggest; but that there is no magic barrier to splicing animal genes into plants, other animals, bacterial, etc.
I think you mean, "I don't see why this is a bigger problem than cracking other people's networks." It's obviously a problem for people to be breaking into others' systems; it's just that you're saying that the USCoC is nothing special. Right?
It sounds, to me, like a combination of "Remember, no matter how hard they work to imply otherwise, the 'US Chamber of Commerce' is just a private lobbying group, not part of the government" and "Awww, so sad, couldn't have happened to a nicer bunch of people."
I'm not a huge fan of internet security being a total free-fire zone; but the USCoC are pretty much slime, have been known to use dirty tricks themselves, so it just isn't feasible to exude too many tears for them.
My sources tell me that evildoers possess advanced ICMP technology that would allow a pirate to verify whether or not a forbidden server is active, among other criminal surveillance, from anywhere in the Homeland!
While Firefox's marketshare has been suffering slightly, I can't imagine that the per-seat value of being the default search engine has changed particularly, and FF is probably the competitor from which Google gains the most: FF reliably agrees with them on most major issues, has no significant strength to threaten Google's actually profitable ventures, and no(well, almost no, you could build FF-only XUL webapps; but nobody does) competing application environment.
Microsoft has a browser, a search engine, win32, and silverlight, so they aren't exactly somebody that Google wants gaining ground, Apple has impressive control of certain high margin markets, and an iron grip on their mobile devices. Firefox has a browser. Unless Google has some aesthetic reason to crush anything it can, and risk the wrath of the antitrust guys, Firefox's existence is somewhere between 'harmless' and 'downright convenient'.
I'd be the first to agree that progress in that area has been surprisingly uneven, with a lot of "this isn't an 'advance' this is just a cheaper version of what we were doing in VMS, damn it" and even outright regressions, as with the apparent death-spiral of OpenSolaris without btrfs being all there yet.
My point was just that, beyond the level of reliability necessary to attract OEM contracts(ie. acceptably low number die in the first, second, and 3rd years, so warranty support doesn't bleed the OEM dry) and the level of reliability necessary to keep RAID arrays and SANs from having disks die faster than they can be rebuilt, there is more or less zero additional willingness to pay for reliability at the level of the single drive.
All the additional willingness to buy reliability(and there is a lot of it) is absorbed by a combination of RAID chipset vendors, SAN appliance guys, various flavors of backup hardware and software(with emphasis on things like dedupe, site failover, and regulatory compliance on the high end, and 'being easy enough that I might actually use it' on the consumer end).
On the speed side, the picture is similar, there is a great deal of money in it for you if you can deliver higher throughput on people's fancy databases and very high traffic websites; but the amount of 'speed premium' money that is paid out at the drive level is probably declining. SSDs make the IOPs difference between a 5400k and a 15000k RPM drive look like a rounding error, and pools of memcache servers or high end storage appliances that will expose a half TB of RAM as a fibre channel device, and the like, are the ones soaking up the real money to be made on speed. Even when HDDs are used, the fancy controller required to not bottleneck a shelf of 15Ks will cost a nontrivial percentage of the drives that populate it.
I don't mean to imply that these markets are necessarily going in the right direction, just that the HDD guys are very poorly placed to grab any of the money on the table. All the solutions either call for no HDDs at all, or are of the "Send us a bunch of dumb block devices. If they are 50% larger and cost half as much as last year, that would be good." There just isn't anything that you can do, within the confines of the single device(with mechanical HDD tech, and the mechanical device companies don't have silicon fab background, so that doesn't help them) that is all that interesting, or worth a premium, except for bumping capacity and lowering price.
I'm guessing that the Revolutionary Guard doesn't have to worry about the FCC or obstruction of justice charges...
You still need a warrant if the surveillance is directed at an individual. And if it's just patrolling, how is that any different than a cop walking his beat?
Surveillance technologies bring two main changes to the table, even when otherwise analogous to some prior method:
1. Economics: There is no legal problem with having cops walking 100% of the legally public beats 100% of the time. Economically, though, there just aren't enough cops to do that. In practice, one of the major protections from the state historically enjoyed by most people is not law; but simple lack of resources. It may be legal to have a cop follow you on a public road, and determine your route; but that cop isn't cheap, so you'll have to attract some suspicion first. Slapping a $100, reusable, magnetic GPS bug on your car, on the other hand, is overwhelmingly cheaper than having a $50,000/yr cop following you. Even if the two are analogous, the level of protection enjoyed in one case is far lower than in the other.
2. Retention: Humans, by necessity, have lousy memories. Our eyes just slide right over mundane happenings and they fall away almost immediately. Storage of electronic surveillance data, on the other hand, is cheap and getting cheaper(and easier to automatically search). Trying to track the routes of all motorists in a city based on data from the beat cops would be essentially impossible. Doing the same from an equivalent number of license-plate cameras? Hard; but tractable.
The crux of the matter is that, as cost decreases and retention increases, 'just patrolling' and 'surveillance directed at an individual' stop being distinct categories: the agents that are 'just patrolling' gather and retain enough data that (proactively or retroactively) turning that patrol into surveillance is essentially just a matter of doing the DB lookup. We haven't reached that point yet; but basically any advance in the cost or capability of automated surveillance technology moves us closer. Patrolling and targeted surveillance aren't fundamentally different, they are different because human agents are really bad at patrolling, and have to be given quite different orders if you want them to get useful data on a specific target. If an agent is good at patrolling, all people that pass within its view are effectively surveilled...
How'd you like to be the patent examiner who has to strike down that application because IBM's own work in tracking people from ghetto to oven constituted prior art?
Surely they could just tap one of the statistics professors at hamburger university. I imagine that describing the origin of a given nugget in the n-dimensional probability space encompassing all the possible distributions of the various animals on the line at production time would be rather like describing the position of an electron in space: You couldn't actually say where an individual nugget falls; but with knowledge of the production process and the input animals you could model the statistical distribution of where each nugget is more or less likely to fall...
Dear Cayenne8,
We in the agriculture and food processing industry take customer satisfaction extremely seriously. For a small additional packaging fee, we would be delighted to ensure that your food has been locally produced in the location of your choice and certified to whatever standard you desire by whatever certification bodies you trust most. Our graphics department may require 8-10 additional business days for certification logos not already in our library, and 4-colour printing is extra.
To suit the requirements of today's environmentally sensitive customer, we are proud to label all our products as being sustainably derived from non-endangered species, or endangered species whose tissues are indistinguishable by any test likely to be employed by the customs agents of your jurisdiction.
Sincerely Yours,
The Supply Chain.
For discrete cuts of meat, the labeling should be simple enough; but some of the more, er, 'waste minimizing' meat products are going to get seriously complex.
The composition of a given hamburger would probably have to be given as a joint probability density function across a set of hundreds or thousands of animals or something similarly messy. That would give label-readers something to ponder...
Apple designs have a rather long history(Apple III-present, somewhat intermittently) of running everything right near thermal redline in order to keep it quiet. On the plus side, it is pretty impressive how quiet some of their hardware is, especially given its chassis size(quiet is easy when you can just have a row of low-speed 120mm fans blowing over everything, much harder when you have a maze of teeny little air channels and speed controlled blowers and stuff); but it gets damn toasty in there...
In such a brutally commodified market as HDDs, I suspect that all the companies are run by bean counters and the visionaries are dead.
Because they are so cheap, per GB, mechanical HDDs aren't going anywhere for a decent chunk of the future; but they've been boxed in such that there isn't any noticeable room for 'visonary' development:
1. Performance? If you want that, you'll be talking to a totally different company with a background in semiconductors, either Flash or DRAM, depending on how much money you are made of. Nothing stopping the HDD people from selling rebadges; but rebadging is not exactly a visionary(or terribly high margin) business.
2. Reliability? Because servicing a warranty request isn't inexpensive(phone drones, fedex, etc.) anybody who can't deliver drives with a low failure rate during standard PC OEM warranty periods is going to find their sales limited; but reliability at the drive level isn't actually worth very much: The value of the world's spinny disks is peanuts compared to the value of the data on them. Most of the reliability money and R&D is going into RAID, advanced filesystems, various automated redundancy and backup solutions, etc. Again, nothing stops the HDD guys from selling rebadge RAID controllers or cloud backup services; but rebadging is not exciting.
3. Features? If it doesn't just drop in and play nice with the SATA/SAS controllers of the world, including the legacy and currently shipping ones, it's a dud. If it has some cool feature that is supported only by your proprietary utility, on controllers that directly pass the necessary nonstandard commands, it isn't going to be wildly useful. If it achieves sufficiently broad adoption that OS and HDD controller support starts coming standard, it is no longer a unique competitive advantage...
Cynically, there is also the fact that even people buying on the basis of desire for mechanical reliability don't have access to very good information: hardware and firmware revs change constantly, sometimes with a change in model designation, often not, some designs turn out to be workhorses, some are deathstars, some batches are bad, some aren't. Everybody has an emotional position on reliability, based largely on which brands failed them in the recent past; but unless they are buying in serious volume and somewhat behind the tech curve, data about the past are largely obsolete.
As a rough heuristic, any state wealth-redistribution program that doesn't have packs of lobbyists and AEI economists shrieking about socialism, communism, and class warfare is very likely converting your tax dollars into somebody's shareholder value more or less by design.
Wealth-redistribution plans that do have such a pack are somewhat less clear. They are less likely to be an overt screwjob; but their efficiency or efficacy may still be miserable.
Fermilab Director Pier Oddone also managed to mention that the Fermilab gift shop is now stocking a variety of novelty plush quarks, 'Experi-mint' brand candies, and similar knicknacks before breaking down and sobbing something about 'Is this why I made it through my postdoc?'
While it does nothing against the possibility that the allegedly US cables are, in fact, forgeries by some third party, Kazakhstan seems like an odd place for the US to risk local relations in order to smear Russia(it's not as though there aren't plenty of other options, of varying degrees of truth, available). Our options for things like supply routes and airbases in the region are tepid to say the least and Kazakhstan is one of the countries we've been trying to butter up for that sort of access, along with their delicious petroleum.
It would seem extraordinarily stupid, even if one ignores any minor hindrances like 'ethics' or 'truth', to run an anti-Russia smear campaign at the expense of relations with a strategically located (and, so far, fairly cooperative in our 'perpetual war on terror' plans) country.
Don't worry. Keratin is durable as hell by biological standards, so Dear Leader's amazing hair should still be in excellent condition decades from now...
I, for one, would like to thank the citizens of Texas for subsidizing my semiconductor purchases!
Yeah, I'm sure Texas just gives those incentives because they're what, stupid? Altruistic? No - they and the rest of the world that gives these incentives understand that in the end it comes out as a big win for their local economy.
I certainly wouldn't accuse them of altruism, and only some of them of stupidity:
'Incentives' are, indeed, a very good way of obtaining a package of immediate and visible benefits in exchange for a package of costs that is generally relatively small up front and packs most of its punch in indirect or time-deferred costs. Politically, it's a great deal. Economically, the suggestions that it is a good idea are equivocal at best, negative at worst.
Dell is, arguably, more of a logistics and integration company than a tech company(which isn't necessarily a bad thing, they are pretty decent at it, and somebody has to do it)...
They are pretty good at providing a one-stop-shop for a variety of Intel and AMD silicon, with supporting chips from a number of other vendors, mounted on a standardized set of boards from a few pacific rim OEM shops, and stuffed into plastic boxes in Mexico according to your order. Juggling that worldwide logistics effort is no mean feat; but they don't mix much in the way of dell technology into the sauce. It's like fedex with driver updates...