Is there a market for used or stolen Tesla cars or parts?
It wouldn't 100% shock me (though it'd have to be an export job, 'notable and uncommon', 'aggressively interacts with the vendor', and 'stolen' are not attributes that work well together); but it's probably not on the top of the list of cars that flip or chop easily.
On the other hand, its materials/recycle value is probably above average for vehicles of its size.
'Frozen' was referring to the grandparent post's expression of disappointment that they aren't using glycoproteins and sub-freezing temperatures. My point was that ethically doing those experiments, or even getting IRB approval for those experiments, on this patient population would (and should) be essentially impossible.
This study aims to improve outcomes for severe tissue damage. The grandparent poster wanted research into long-term hibernation. Aside from long-term work simply being riskier and more speculative, it is likely that at least some of what works best there is directly contrary to what works best for short term treatment of tissue damage.
Radically better thrust/weight ratios than chemical fuels? Potentially better behaved than 'Project Orion' style nuclear propulsion?
Be that as it may, I'm pretty sure that no sane IRB would sign off on using cryonics and experimental nonhuman proteins on gunshot victims just because Space is Awesome, man! The scope of the study is techniques to provide team trauma surgeon more time to stitch them back up before they bleed out, a short timeframe, and likely one where working on frozen tissue would not make matters easier.
Perhaps on a certain, small, specific, set of hardware targets; but the "Extract Proprietary Blobs" step is part of the Cyanogenmod build process for a reason...
I think that it's mostly AOSP at higher levels; but when 'the details' are kernel-level drivers that can do whatever the hell they want without you noticing, or firmware that has its own CPU and memory space in which to hide and do god-knows-what, you can be pretty sure that if there are devils in the details, you are fucked. Gratuitously.
Could we please stop with the endless pro-drug commentary? Not all of us are drug addicts, and it's tiresome when people like yourself have to bring up drug issues in nearly every thread. I know it's important to you, but normal people don't care about such matters.
You don't have to give a damn about drugs, pro or con, to admit that they are a useful example to bring up in the context of the effect of prohibition on the available margins in a given market.
If you prefer a classier example, from a (not actually more civilized; but definitely better dressed) age; we can talk about booze. Same effects were seen under the Volstead act.
I'm not an HR flack (Thank the nonexistent deities of a thousand dead pantheons); but I'd guess that, if HR (or anybody else who felt like quietly dropping the dime on you for some reason at any point during your tenure...) was feeling nasty, testing for degree mills would be easier than testing for fake or overstated degrees.
Thanks to some combination of FERPA and the Office of the Bursar's desire to extract fees, actually getting the details of a student's stay at a given school is a pain. By contrast, googling the exact name of the school the applicant claims (or even running an automated query against info from accreditation bodies, if you do this in bulk) is easy.
They may or may not care; but trusting the difficulty of that database problem is like turning your back on a dude with a knife and a nasty gleam in his eye.
Not all options actually hug the tradeoff curve(well, in three variables, I suppose it's some sort of surface; but same idea) all that closely (if at all), so it's still a partially legitimate question... (Which state school is basically north of 10k/yr for beer pong and date rape and which one is an affordable and decent college? Is that ad for SOMEBODY TECHNICAL INSTITUTE a total scam? How much of the expense of a traditional campus can I skip without ending up in a 'MOOC' that might as well just be watching a couple of youtube videos, only more expensive?).
That said, I'd be...a trifle nervous... about anyone who "eh, just wants to get a fast, cheap, CS degree, y'know?". Unless you have purely mercenary motives(and a fairly solid estimate of how much more you could be earning if you had one from a school of a given caliber, in which case crunch the cost of going to school, opportunity and direct, compare to expected increased future earnings, and go on your value-rational-homo-economicus way...), you don't get a CS degree, definitely not a CS degree that you wouldn't be ashamed of, just for the CV.
If you are already a programmer, you know enough about CS-like things that 'CS for enrounding you as a person and enrichment' will be irrelevant, so you have two choices: Do you want to take actual, big-kid, CS for people who want a better grasp of a deeply hairy area of mathematics? Or would you be better off skipping that and focusing on software engineering/development related skills that will make your practical-applied-programming more solid, more maintainable, generally better on the logistics side?
Hard math just seems like a bad place for dabblers: If you just skate, you'll be wasted on anybody who just wants a coder, now(since they don't care about your fancy theoretical education, just your work experience); and equally wasted on anybody who wants to pick your brain, see how you think; because faking knowledge of hard subjects is hard.
This really seems like a " Do, or do not." affair.
... everyone will start printing at home, things will be cheaper, more available, better, faster, stronger...
When in reality everyone will just start using pretentious newbie phrases such as "steep learning curve".
I think that the good Mr. Brians might actually have stumbled across the correct answer, possibly not recognized it, then dismissed it in his enthusiasm to accuse users of being mathematically unsophisticated idiots.
If you think of using a tool as being a mixture of doing whatever it is you wanted to do and 'learning' (ie. swearing at the tool itself, and having things that seem like they should work not work for reasons you don't understand rather than getting what you wanted done, done), the 'learning curve' that people draw for various things ends up being (fairly consistently) the curve whose slope, at time X, represents the amount of 'learning' that the user is suffering at time X, and whose X axis is time.
The Y value doesn't directly represent mastery, difficulty, or any other task-related parameter, it's just that people always seem (quite possibly because of the hill-climbing analogy he proposes, and the fact that humans find pain to be a salient experience) to draw the graph such that its slope at each point correctly represents the intensity of the 'learning' at that point. The derivative of the learning curve is the one that actually has a nice task-related variable on each axis, and a relationship between them. The learning curve's Y axis isn't directly task-related at all (and the constant of integration depends on how thick the learner is).
Given that even people who haven't taken calc do it this way, I don't think that it's conscious in most cases; but it squares with what I've always observed people to mean about a task based on the "learning curve" they draw for it: The curve they draw does differ between tracking 'time' on the X axis and tracking 'degree of mastery' (if they implicitly assume that learning will occur with experience, it is time, if they acknowledge the possibility of the user quitting or being inadequate to further learning, it's degree of mastery, but no specific learning curve requires equivocating here, there just isn't a standard); but it seems to be consistently the case that the Y value means nothing directly; but emerges because the slope at each point is chosen to be meaningful, and subsequent Y values are what they are just to keep the line continuous.
The best tasks would have asymptotic learning curves: starts utterly trivial and increasing mastery begets only increasing difficulty to an unbounded degree. Sounds a lot like math, actually.
It's not just a 'killer app', it's a killer app that, for some reason, (illegal? a hideously embarassing sex toy? time-critical?) the user wouldn't be better off just uploading to one of the 3d printing services and having fedexed to them, maybe even picking up in-store if they live in an area with sufficient demand.
2d printers, which are anywhere from 'better' to 'technology indistinguishable from magic' in terms of maturity, ease-of-use, cost, consumer acceptance, etc. already seem to be suffering pretty heavily from this effect. They aren't extinct or anything; but Having Your Very Own Home Printer! is a chore that people hate, and really only do for stuff that can't be sent off to the photo printing service or dashed off on the office's high volume laser. Even as assorted futuristic 'paperless' scenarios fail to pan out year after year, the printer as something you want to have personally is looking rather sickly. It'll be a cold day in hell when 'printing' goes; but nasty little home printers appear to have peaked and gone into retreat, despite their low cost and maturity. 3d printing, as something you actually do, rather than something you order or something that is important to assorted background steps, may never 'peak' at any noticeable level(obviously, there will have to be a 'peak' in there somewhere; but it needn't be very visible or relevant.)
Especially with 3d printing putting so much of the emphasis on materials (rather than mere pigments/dyes), and with most of the really cool materials either coming last/never to low-end gear (barring a radical discontinuity in the cost of high power lasers and optics to suit, laser sintered metal probably isn't coming home to you) or requiring additional processing steps that aren't particularly user friendly(ceramic powder/slurry processes aren't too bad; but the parts aren't much use until you put them through the kiln... modelling waxes for 'lost wax' casting are downright friendly; but the molten bronze steps that follow really aren't, and so forth), there is a very, very, hard sell to be made for having an in-home unit.
Obviously, there already are in-home units in homes, so no theoretical proof is needed of the fact that some people want them; but the 'just pay for a timeslice on somebody's $100,000 printer...' factor makes it much more plausible that the in-home population is not a precursor of a boom just waiting for a little more maturity to take off; but a much closer to stable enthusiast population that will have better printers in 10 years than they do today; but may not be all that much larger.
Plus OLE support. Quite a powerful capability; but one of those powerful capabilities best handled carefully, kept away from direct sunlight, protected from shocks, and otherwise treated as though it is just waiting to ruin your day.
The one trick (comparatively rare; but it happens at times) is that if you take an RTF document and give it a.doc suffix, Word will interact with it happily enough and I think even save it in the RTF format if you modify-and-save.
This means that if you block by suffix, a remotely clueful attacker will just fix their suffix and carry on; but if you block by format a small and fairly unpredictable subset of '.doc' files will be weeded out for reasons users will be unlikely to grasp.
This would hardly make it the most painful thing routinely inflicted on users in the name of security; but it isn't a plus.
Privilege escalation is always worse than 'execute with same privileges as user'; but for primarily-end-user software the distinction seems a great deal less helpful (unlike, say, on the server, where attacks isolated to one service account or daemon are legitimately less dangerous). Joe User's security context has access to more or less his entire life in documents and ill-secured website passwords, and enough permission to plant something that will start when he next logs in in a zillion different places that he isn't likely to notice(details will vary by OS; but the only real exception would be the control-freakier mobile ones). So Joe User is screwed at either privilege level, and, from the perspective of fixing the system, conclusively proving that only user-level access was gained and the system is still secure (much less attempting to fix it if it isn't) is so much more time consuming than just nuking it and applying a fresh image that you'd only try in order to get samples of the attacker, not because it's worth the trouble on its own.
Certainly, banks are a major point of weakness at best (and actively malicious at worst) regardless of currency. That's why I find it so astonishing to watch people with supplies of their preferred cryptocoins line up to hand them over to one. Especially because real banks, long recognized as being both dangerous and potentially unstable, usually come with rather more legal cover than 'exchanges' do. (The exchanges don't have zero, in most jurisdictions they presumably still fall under whatever generic 'please minimize fraud during the course of business' and 'ordered list of creditors to pay in the event of your liquidation' regulations exist; and banks are sufficiently dangerous that they sometimes achieve regulatory capture, rather than regulations designed to reduce their risk.)
If there are actually nontrivial numbers of people who are using cryptocurrencies and living in locations where the local banking system is so fucked that 'banking' in bitcoins is actually safer, they have my sympathy and their behavior makes more sense; but the remainder confuse me: If you want the properties of a cryptocurrency you (with the present designs, at any rate) cannot 'bank' with it. If you do, you are essentially going from enjoying the properties of a cryptocurrency to enjoying the properties of the really, really, really lousy end of the money market pool (about the same security as commercial paper issued by god-knows-who; but generally not fixed term or with the same compensation for risk) mere denominated in your cryptocurrency of choice, something entirely different and notably worse.
If you want the properties of a bank or other financial institution, you have to either be in a seriously shitty location or moving hot assets to not be safer doing it in some accepted currency at whatever available institution seems closer to the 'lawful evil' side of the spectrum. Still plenty of good ways to lose money (anything from the assorted nickle-and-diming of retail banking, up to the sophisticated chicanery that requires a superb grasp of finance to dangerously misunderstand); but nobody even pretends that any fundamental structural properties will save them, because they won't, and so some safeguards, or at least signage about how deep the water is and where the sharks are, have been constructed.
That's the bit that confuses me: if you play by the rules of a cryptocurrency, certain structural guarantees hold relatively strongly; but (best case) you've got about as much legal cover as if Gamestop went out of business before the game you pre-ordered arrived. Frequently less.
If you play by the rules of some set of financial institutions, you have no structural guarantees; but you've got a wide variety of legal covers built to deal with the fact that there are no structural guarantees.
If you attempt to do both at the same time, with the same stuff, you get the worst of both. (Obviously, doing both at the same time with different assets is totally doable, but unrelated). Doing that, I just don't understand.
It's a trifle astonishing to watch how persistently people line up to make the same mistake with their crytopcurrency-of-the-moment again and again.
In theory, cryptocurrencies are secure-through-math and don't rely on flyblown banking institutions and so on, (and, in fairness, they have a decent track record as software goes); but their properties only apply if you use them correctly.
If you give the actual crypto keys that correspond to your cryptocurrency units to me, and I give you an account at First Bank of Fungus with 'X cryptocoins', guess what? From the perspective of all the neat math, I own the coins, and enjoy whatever properties they possess, and you own a not-particularly-distinguished private-label IOU, which offers absolutely no security by design, and probably quite limited security-by-legal-force.
Basically none of the special properties of cryptocurrencies extend beyond your personal grasp on them, and the surrounding institutions are... dubiously stable.
Instead of trying for massive multiplayer, Maybe they should of concentrated on the people that got the series there in the first place - the ones not playing multiplayer?
Thoughts?
Anyone willing to endure the ISO Standard JRPG levelling mechanics ("Wander around an apparently empty landscape until a random encounter occurs, fight it out with some NPCs, repeat A Lot because even if you are now massively overpowered, you know that the actual major boss will fry you into a grease spot with just a nasty look unless you do.") is a perfect candidate for MMORPGs...
The electrical behavior of neurons may be discrete; but synaptic behavior is... profoundly unhelpful... to attempts to treat the whole arrangement as not having a pretty substantial analog component.
Thanks, I'll have to give that a look. We currently have 6.9sp5 bodged into pretty-much-working in the post-XP environment; but it doesn't exactly require oracular powers to suspect that the longer term outlook for Altiris is pretty grim.
Your mention of shipping people 'to various countries' gives me an idea...
Since all the 'extraordinary rendition' bag, drag, and torture kids at the CIA are still running around in arrogant impunity, going so far as to just yoink inconvenient documents from the Senate Intelligence Committee(seriously, most of the members of that are appeasnik fuckwits who basically worship the clandestine services, so it must be really, really bad if the CIA is embarrassed in front of them. Also, if there are things the clandestine services do that even that part of the senate isn't allowed to know about, can we really maintain the pretense that civilian government is actually in anything resembling control?) how about pitting two problems against one another?
It'll be an exciting contest, like a reality TV show; but with higher stakes, rules as follows:
The NSA will be the intelligence-spooks team: their job is to dig up as much dirt on the CIA as possible, by whatever l33t haxx0ring necessary, and try to have the CIA neutralized by political and/or public outrage, at least to the point of organizational collapse, to the point of wholesale hangings-from-the-lampposts for bonus points.
The CIA will be the wet-ops creeps team: they will have to 'disappear' key NSA personnel to our worldwide network of extralegal torture dungeons fast enough to keep the lid on their dirty laundry, and try to drive the NSA to the point of institutional paralysis or collapse, with extra points awarded for any actually-true facts obtained during the 'enhanced interrogation' sessions.
Gentlemen, to the starting line, and may you both lose!
Horrid little vermin, and the thing likely has a couple of brutally well optimized high speed analog PID controllers, all within its (very tight) payload limits, and all since before we were grunting and hitting one another with rocks. Thanks nature...
Slashdot Mirror
Is there a market for used or stolen Tesla cars or parts?
It wouldn't 100% shock me (though it'd have to be an export job, 'notable and uncommon', 'aggressively interacts with the vendor', and 'stolen' are not attributes that work well together); but it's probably not on the top of the list of cars that flip or chop easily.
On the other hand, its materials/recycle value is probably above average for vehicles of its size.
'Frozen' was referring to the grandparent post's expression of disappointment that they aren't using glycoproteins and sub-freezing temperatures. My point was that ethically doing those experiments, or even getting IRB approval for those experiments, on this patient population would (and should) be essentially impossible.
This study aims to improve outcomes for severe tissue damage. The grandparent poster wanted research into long-term hibernation. Aside from long-term work simply being riskier and more speculative, it is likely that at least some of what works best there is directly contrary to what works best for short term treatment of tissue damage.
Radically better thrust/weight ratios than chemical fuels? Potentially better behaved than 'Project Orion' style nuclear propulsion?
Be that as it may, I'm pretty sure that no sane IRB would sign off on using cryonics and experimental nonhuman proteins on gunshot victims just because Space is Awesome, man! The scope of the study is techniques to provide team trauma surgeon more time to stitch them back up before they bleed out, a short timeframe, and likely one where working on frozen tissue would not make matters easier.
"Yes, he's alive, and in perfect hibernation."
Perhaps on a certain, small, specific, set of hardware targets; but the "Extract Proprietary Blobs" step is part of the Cyanogenmod build process for a reason...
I think that it's mostly AOSP at higher levels; but when 'the details' are kernel-level drivers that can do whatever the hell they want without you noticing, or firmware that has its own CPU and memory space in which to hide and do god-knows-what, you can be pretty sure that if there are devils in the details, you are fucked. Gratuitously.
Could we please stop with the endless pro-drug commentary? Not all of us are drug addicts, and it's tiresome when people like yourself have to bring up drug issues in nearly every thread. I know it's important to you, but normal people don't care about such matters.
You don't have to give a damn about drugs, pro or con, to admit that they are a useful example to bring up in the context of the effect of prohibition on the available margins in a given market.
If you prefer a classier example, from a (not actually more civilized; but definitely better dressed) age; we can talk about booze. Same effects were seen under the Volstead act.
I'm not an HR flack (Thank the nonexistent deities of a thousand dead pantheons); but I'd guess that, if HR (or anybody else who felt like quietly dropping the dime on you for some reason at any point during your tenure...) was feeling nasty, testing for degree mills would be easier than testing for fake or overstated degrees.
Thanks to some combination of FERPA and the Office of the Bursar's desire to extract fees, actually getting the details of a student's stay at a given school is a pain. By contrast, googling the exact name of the school the applicant claims (or even running an automated query against info from accreditation bodies, if you do this in bulk) is easy.
They may or may not care; but trusting the difficulty of that database problem is like turning your back on a dude with a knife and a nasty gleam in his eye.
As with everything else, Pick two.
Not all options actually hug the tradeoff curve(well, in three variables, I suppose it's some sort of surface; but same idea) all that closely (if at all), so it's still a partially legitimate question... (Which state school is basically north of 10k/yr for beer pong and date rape and which one is an affordable and decent college? Is that ad for SOMEBODY TECHNICAL INSTITUTE a total scam? How much of the expense of a traditional campus can I skip without ending up in a 'MOOC' that might as well just be watching a couple of youtube videos, only more expensive?).
That said, I'd be...a trifle nervous... about anyone who "eh, just wants to get a fast, cheap, CS degree, y'know?". Unless you have purely mercenary motives(and a fairly solid estimate of how much more you could be earning if you had one from a school of a given caliber, in which case crunch the cost of going to school, opportunity and direct, compare to expected increased future earnings, and go on your value-rational-homo-economicus way...), you don't get a CS degree, definitely not a CS degree that you wouldn't be ashamed of, just for the CV.
If you are already a programmer, you know enough about CS-like things that 'CS for enrounding you as a person and enrichment' will be irrelevant, so you have two choices: Do you want to take actual, big-kid, CS for people who want a better grasp of a deeply hairy area of mathematics? Or would you be better off skipping that and focusing on software engineering/development related skills that will make your practical-applied-programming more solid, more maintainable, generally better on the logistics side?
Hard math just seems like a bad place for dabblers: If you just skate, you'll be wasted on anybody who just wants a coder, now(since they don't care about your fancy theoretical education, just your work experience); and equally wasted on anybody who wants to pick your brain, see how you think; because faking knowledge of hard subjects is hard.
This really seems like a " Do, or do not." affair.
Well, if I had something that I wanted to sell; but didn't know if people wanted to buy, they sound like ideal company...
Did you, perchance, get the glass fabricated by the same entity that probed your eyes?
It is...remarkable... what this does to the price.
... everyone will start printing at home, things will be cheaper, more available, better, faster, stronger ...
When in reality everyone will just start using pretentious newbie phrases such as "steep learning curve".
I think that the good Mr. Brians might actually have stumbled across the correct answer, possibly not recognized it, then dismissed it in his enthusiasm to accuse users of being mathematically unsophisticated idiots.
If you think of using a tool as being a mixture of doing whatever it is you wanted to do and 'learning' (ie. swearing at the tool itself, and having things that seem like they should work not work for reasons you don't understand rather than getting what you wanted done, done), the 'learning curve' that people draw for various things ends up being (fairly consistently) the curve whose slope, at time X, represents the amount of 'learning' that the user is suffering at time X, and whose X axis is time.
The Y value doesn't directly represent mastery, difficulty, or any other task-related parameter, it's just that people always seem (quite possibly because of the hill-climbing analogy he proposes, and the fact that humans find pain to be a salient experience) to draw the graph such that its slope at each point correctly represents the intensity of the 'learning' at that point. The derivative of the learning curve is the one that actually has a nice task-related variable on each axis, and a relationship between them. The learning curve's Y axis isn't directly task-related at all (and the constant of integration depends on how thick the learner is).
Given that even people who haven't taken calc do it this way, I don't think that it's conscious in most cases; but it squares with what I've always observed people to mean about a task based on the "learning curve" they draw for it: The curve they draw does differ between tracking 'time' on the X axis and tracking 'degree of mastery' (if they implicitly assume that learning will occur with experience, it is time, if they acknowledge the possibility of the user quitting or being inadequate to further learning, it's degree of mastery, but no specific learning curve requires equivocating here, there just isn't a standard); but it seems to be consistently the case that the Y value means nothing directly; but emerges because the slope at each point is chosen to be meaningful, and subsequent Y values are what they are just to keep the line continuous.
The best tasks would have asymptotic learning curves: starts utterly trivial and increasing mastery begets only increasing difficulty to an unbounded degree. Sounds a lot like math, actually.
It's not just a 'killer app', it's a killer app that, for some reason, (illegal? a hideously embarassing sex toy? time-critical?) the user wouldn't be better off just uploading to one of the 3d printing services and having fedexed to them, maybe even picking up in-store if they live in an area with sufficient demand.
2d printers, which are anywhere from 'better' to 'technology indistinguishable from magic' in terms of maturity, ease-of-use, cost, consumer acceptance, etc. already seem to be suffering pretty heavily from this effect. They aren't extinct or anything; but Having Your Very Own Home Printer! is a chore that people hate, and really only do for stuff that can't be sent off to the photo printing service or dashed off on the office's high volume laser. Even as assorted futuristic 'paperless' scenarios fail to pan out year after year, the printer as something you want to have personally is looking rather sickly. It'll be a cold day in hell when 'printing' goes; but nasty little home printers appear to have peaked and gone into retreat, despite their low cost and maturity. 3d printing, as something you actually do, rather than something you order or something that is important to assorted background steps, may never 'peak' at any noticeable level(obviously, there will have to be a 'peak' in there somewhere; but it needn't be very visible or relevant.)
Especially with 3d printing putting so much of the emphasis on materials (rather than mere pigments/dyes), and with most of the really cool materials either coming last/never to low-end gear (barring a radical discontinuity in the cost of high power lasers and optics to suit, laser sintered metal probably isn't coming home to you) or requiring additional processing steps that aren't particularly user friendly(ceramic powder/slurry processes aren't too bad; but the parts aren't much use until you put them through the kiln... modelling waxes for 'lost wax' casting are downright friendly; but the molten bronze steps that follow really aren't, and so forth), there is a very, very, hard sell to be made for having an in-home unit.
Obviously, there already are in-home units in homes, so no theoretical proof is needed of the fact that some people want them; but the 'just pay for a timeslice on somebody's $100,000 printer...' factor makes it much more plausible that the in-home population is not a precursor of a boom just waiting for a little more maturity to take off; but a much closer to stable enthusiast population that will have better printers in 10 years than they do today; but may not be all that much larger.
Plus OLE support. Quite a powerful capability; but one of those powerful capabilities best handled carefully, kept away from direct sunlight, protected from shocks, and otherwise treated as though it is just waiting to ruin your day.
I'm pretty sure nobody would notice or care.
The one trick (comparatively rare; but it happens at times) is that if you take an RTF document and give it a .doc suffix, Word will interact with it happily enough and I think even save it in the RTF format if you modify-and-save.
This means that if you block by suffix, a remotely clueful attacker will just fix their suffix and carry on; but if you block by format a small and fairly unpredictable subset of '.doc' files will be weeded out for reasons users will be unlikely to grasp.
This would hardly make it the most painful thing routinely inflicted on users in the name of security; but it isn't a plus.
Privilege escalation is always worse than 'execute with same privileges as user'; but for primarily-end-user software the distinction seems a great deal less helpful (unlike, say, on the server, where attacks isolated to one service account or daemon are legitimately less dangerous). Joe User's security context has access to more or less his entire life in documents and ill-secured website passwords, and enough permission to plant something that will start when he next logs in in a zillion different places that he isn't likely to notice(details will vary by OS; but the only real exception would be the control-freakier mobile ones). So Joe User is screwed at either privilege level, and, from the perspective of fixing the system, conclusively proving that only user-level access was gained and the system is still secure (much less attempting to fix it if it isn't) is so much more time consuming than just nuking it and applying a fresh image that you'd only try in order to get samples of the attacker, not because it's worth the trouble on its own.
Certainly, banks are a major point of weakness at best (and actively malicious at worst) regardless of currency. That's why I find it so astonishing to watch people with supplies of their preferred cryptocoins line up to hand them over to one. Especially because real banks, long recognized as being both dangerous and potentially unstable, usually come with rather more legal cover than 'exchanges' do. (The exchanges don't have zero, in most jurisdictions they presumably still fall under whatever generic 'please minimize fraud during the course of business' and 'ordered list of creditors to pay in the event of your liquidation' regulations exist; and banks are sufficiently dangerous that they sometimes achieve regulatory capture, rather than regulations designed to reduce their risk.)
If there are actually nontrivial numbers of people who are using cryptocurrencies and living in locations where the local banking system is so fucked that 'banking' in bitcoins is actually safer, they have my sympathy and their behavior makes more sense; but the remainder confuse me: If you want the properties of a cryptocurrency you (with the present designs, at any rate) cannot 'bank' with it. If you do, you are essentially going from enjoying the properties of a cryptocurrency to enjoying the properties of the really, really, really lousy end of the money market pool (about the same security as commercial paper issued by god-knows-who; but generally not fixed term or with the same compensation for risk) mere denominated in your cryptocurrency of choice, something entirely different and notably worse.
If you want the properties of a bank or other financial institution, you have to either be in a seriously shitty location or moving hot assets to not be safer doing it in some accepted currency at whatever available institution seems closer to the 'lawful evil' side of the spectrum. Still plenty of good ways to lose money (anything from the assorted nickle-and-diming of retail banking, up to the sophisticated chicanery that requires a superb grasp of finance to dangerously misunderstand); but nobody even pretends that any fundamental structural properties will save them, because they won't, and so some safeguards, or at least signage about how deep the water is and where the sharks are, have been constructed.
That's the bit that confuses me: if you play by the rules of a cryptocurrency, certain structural guarantees hold relatively strongly; but (best case) you've got about as much legal cover as if Gamestop went out of business before the game you pre-ordered arrived. Frequently less.
If you play by the rules of some set of financial institutions, you have no structural guarantees; but you've got a wide variety of legal covers built to deal with the fact that there are no structural guarantees.
If you attempt to do both at the same time, with the same stuff, you get the worst of both. (Obviously, doing both at the same time with different assets is totally doable, but unrelated). Doing that, I just don't understand.
It's a trifle astonishing to watch how persistently people line up to make the same mistake with their crytopcurrency-of-the-moment again and again.
In theory, cryptocurrencies are secure-through-math and don't rely on flyblown banking institutions and so on, (and, in fairness, they have a decent track record as software goes); but their properties only apply if you use them correctly.
If you give the actual crypto keys that correspond to your cryptocurrency units to me, and I give you an account at First Bank of Fungus with 'X cryptocoins', guess what? From the perspective of all the neat math, I own the coins, and enjoy whatever properties they possess, and you own a not-particularly-distinguished private-label IOU, which offers absolutely no security by design, and probably quite limited security-by-legal-force.
Basically none of the special properties of cryptocurrencies extend beyond your personal grasp on them, and the surrounding institutions are... dubiously stable.
Have you been asleep for the whole 'NSA' thing? I'm not heartened by this; but if that isn't efficiency, I'm not sure what efficiency looks like...
Instead of trying for massive multiplayer, Maybe they should of concentrated on the people that got the series there in the first place - the ones not playing multiplayer?
Thoughts?
Anyone willing to endure the ISO Standard JRPG levelling mechanics ("Wander around an apparently empty landscape until a random encounter occurs, fight it out with some NPCs, repeat A Lot because even if you are now massively overpowered, you know that the actual major boss will fry you into a grease spot with just a nasty look unless you do.") is a perfect candidate for MMORPGs...
The electrical behavior of neurons may be discrete; but synaptic behavior is... profoundly unhelpful... to attempts to treat the whole arrangement as not having a pretty substantial analog component.
Thanks, I'll have to give that a look. We currently have 6.9sp5 bodged into pretty-much-working in the post-XP environment; but it doesn't exactly require oracular powers to suspect that the longer term outlook for Altiris is pretty grim.
Your mention of shipping people 'to various countries' gives me an idea...
Since all the 'extraordinary rendition' bag, drag, and torture kids at the CIA are still running around in arrogant impunity, going so far as to just yoink inconvenient documents from the Senate Intelligence Committee(seriously, most of the members of that are appeasnik fuckwits who basically worship the clandestine services, so it must be really, really bad if the CIA is embarrassed in front of them. Also, if there are things the clandestine services do that even that part of the senate isn't allowed to know about, can we really maintain the pretense that civilian government is actually in anything resembling control?) how about pitting two problems against one another?
It'll be an exciting contest, like a reality TV show; but with higher stakes, rules as follows:
The NSA will be the intelligence-spooks team: their job is to dig up as much dirt on the CIA as possible, by whatever l33t haxx0ring necessary, and try to have the CIA neutralized by political and/or public outrage, at least to the point of organizational collapse, to the point of wholesale hangings-from-the-lampposts for bonus points.
The CIA will be the wet-ops creeps team: they will have to 'disappear' key NSA personnel to our worldwide network of extralegal torture dungeons fast enough to keep the lid on their dirty laundry, and try to drive the NSA to the point of institutional paralysis or collapse, with extra points awarded for any actually-true facts obtained during the 'enhanced interrogation' sessions.
Gentlemen, to the starting line, and may you both lose!
Horrid little vermin, and the thing likely has a couple of brutally well optimized high speed analog PID controllers, all within its (very tight) payload limits, and all since before we were grunting and hitting one another with rocks. Thanks nature...
Do you think Symantec is capable of building antivirus software that the NSA would need to have weakened for them?