But I assume that a small bonus to an employee every month their machine/isn't/ compromised is perfectly legal, even in a country with sane labor laws? Or perhaps a free lunch?
Of course, this does cost some money, but you'd be surprised how even a small amount of money or food can motivate people to make tiny changes to their routine.
The thing is that the method being trustworthy is necessary, though not sufficient, for trust of the process. With Diebold, a blatantly untrustworthy system mired in problems with bias, incompetence, and lack of security, it doesn't really matter how much I hypothetically trusted the government deploying it.
Make sure to grind the aluminum oxide and iron filings very very finely, this is a mistake that beginners often make and then it doesn't work. I personally recommend a fine file, but this will be SLOW.
It would be a very healthy thing for the freedom of press in America, and do a lot to fight bias. Note to libertarians: I'm not advocating government intervention, just saying I want this to happen. No need to flame me.
This is very different from load balancing as done by Mosix. Mosix was capable of moving threads around as you say, but not of splitting a thread into new ones.
There are some similarities (knowing when a thread will halt is similar) to the problem of making efficient use of multiple cores, but there you have the additional problem of needing to separate out different/branches/ and computations of code, based on an inferred dependency graph, without explicit forks/threads or locking semantics to guide you. Thus you need to know/what/ two different computational threads are, what depends on what, and when, and how long they'll take.
I don't mean to imply approximating it is hopeless, there are a lot of smart researchers working on it, only that doing it well is a long ways away, and that doing it optimally is impossible.
Also keep in mind you're looking at a huge RAM footprint as well, since multiple copies of the same library will be loaded per-application, whereas sometimes you can make optimizations for libraries shared across applications with paging magic.
That said, I thought that OS X apps were statically linked (except with OS libs), and thus tend to be large, but reduce this issue since so much functionality is based on system libs. I could be wrong; I don't really ever work with non-linux systems.
You don't have to lack competition to be a monopoly, you just have to dominate the market. Once you're a "standard", this is generally about when it starts to qualify. Apple is as much a monopoly in MP3 players as Microsoft is in OS, despite both having competition.
My my my don't we jump to assumptions. Stevens was a doddering old man who was let off easy because that's what happens when you commit political crime.
And my knowledge of Stevens' corruption predates my knowledge of the name Barack Obama. Just because a major party says something doesn't make it false, you just have to sift the propaganda mills with a fine sieve.
Could I be wrong? Of course. But I've looked into this guy before, even before he earned the "tubes" name, and was always shocked he had never been brought down.
Keep in mind Alaska doesn't have tax, they have government handouts.
I've had no problem with IG chips, they run all my software just fine and can even play Oblivion. Unless you're gaming or running some eye candy OS, I don't see why you'd need more.
Then again, I suppose I'm judging a discrete graphics card vs integrated as two very different things. If I have a discrete graphics card I expect it to do more than "even play Oblivion".
Yeah turns out this problem is really hard and, depending on how you formalize it, uncomputable. What it boils down to is that the system needs to know how long it will take to run a program on itself and how long it will take to spin it off into another thread.
Sound familiar? It should; that's halt.
Of course, approximating it well is an interesting research problem in language theory, but I wouldn't expect a general purpose system in actual production use any time soon. Even doing it in a pure, strongly typed language will be extremely difficult; implementing one for something like Java/C[++] will be far harder. And of course the ultimate win is doing it to machine code.
MS Research is like a research university for all intents and purposes; they basically have academic latitude. Of course by the time the product reaches market it will be made, um..."better".
I was in awe about it. I had hopes for the ultimate field notebook finally being designed; netbook with crazy in voltage regulation, solar recharge capability, hand crank capability, minimal os almost no power use...The XO was crude but it had incredible potential. The mesh networking alone, if they could get adoption...insane!
If I had more time I'd fork off the project and work on this. Pricepoint $600 or so, a netbook intended for field work. Insane recharge capabilities, crazy networking, sun viewable screen, durable, super light (something the XO fails), etc.
I'd totally buy one. I'm looking forward to when we start seein ghigher end netbooks (perhaps when Apple or Thinkpad* comes out with one), where the focus is on quality and power in a small, light, durable package, rather than the cheapest computer possible.
*Lenovo does make netbooks under ideapad, but as anyone who has bought IBM knows, thinkpad and ideapad are a world apart in quality.
I'm sure that has nothing to do with the auto industry, on which Michigan is pretty much entirely dependent, dying out entirely. No it's clearly the Pinko Commie Nazi taxes.
What about ones which can be either? Interactive ssh versus file transfer.
What I'd love would be if we could let users prioritize their own traffic (with sensible defaults for those who don't know what an internet is), and give them N GB high priority traffic per month and uncapped low priority.
Of course, there are tons of problems with this, but there are times I would willingly prioritize myself down (up/download I intend to leave overnight, 4 GB differential full system backup).
Never mind that with the way current userland and networking is set up this isn't feasible, I think it's a neat idea.
Ah. Not sure if I agree with that law, I can see some good arguments for and against it. I'd be surprised, but not terribly so, if it applied to less serious crimes.
Not only am I not a lawyer, I have no idea what I'm talking about! You should take my legal advice:-)
Slashdot Mirror
But I assume that a small bonus to an employee every month their machine /isn't/ compromised is perfectly legal, even in a country with sane labor laws? Or perhaps a free lunch?
Of course, this does cost some money, but you'd be surprised how even a small amount of money or food can motivate people to make tiny changes to their routine.
The thing is that the method being trustworthy is necessary, though not sufficient, for trust of the process. With Diebold, a blatantly untrustworthy system mired in problems with bias, incompetence, and lack of security, it doesn't really matter how much I hypothetically trusted the government deploying it.
Make sure to grind the aluminum oxide and iron filings very very finely, this is a mistake that beginners often make and then it doesn't work. I personally recommend a fine file, but this will be SLOW.
The same way the Sedition Act wasn't supporting sedition;(
And innovation takes another giant leap backwards.
They'll be be money to be made for those who defect.
It would be a very healthy thing for the freedom of press in America, and do a lot to fight bias. Note to libertarians: I'm not advocating government intervention, just saying I want this to happen. No need to flame me.
This is very different from load balancing as done by Mosix. Mosix was capable of moving threads around as you say, but not of splitting a thread into new ones.
There are some similarities (knowing when a thread will halt is similar) to the problem of making efficient use of multiple cores, but there you have the additional problem of needing to separate out different /branches/ and computations of code, based on an inferred dependency graph, without explicit forks/threads or locking semantics to guide you. Thus you need to know /what/ two different computational threads are, what depends on what, and when, and how long they'll take.
I don't mean to imply approximating it is hopeless, there are a lot of smart researchers working on it, only that doing it well is a long ways away, and that doing it optimally is impossible.
Also keep in mind you're looking at a huge RAM footprint as well, since multiple copies of the same library will be loaded per-application, whereas sometimes you can make optimizations for libraries shared across applications with paging magic.
That said, I thought that OS X apps were statically linked (except with OS libs), and thus tend to be large, but reduce this issue since so much functionality is based on system libs. I could be wrong; I don't really ever work with non-linux systems.
You don't have to lack competition to be a monopoly, you just have to dominate the market. Once you're a "standard", this is generally about when it starts to qualify. Apple is as much a monopoly in MP3 players as Microsoft is in OS, despite both having competition.
My my my don't we jump to assumptions. Stevens was a doddering old man who was let off easy because that's what happens when you commit political crime.
And my knowledge of Stevens' corruption predates my knowledge of the name Barack Obama. Just because a major party says something doesn't make it false, you just have to sift the propaganda mills with a fine sieve.
Could I be wrong? Of course. But I've looked into this guy before, even before he earned the "tubes" name, and was always shocked he had never been brought down.
Keep in mind Alaska doesn't have tax, they have government handouts.
Discrete video costs battery life (if you use it, they can usually be turned off)
Contrast that with Alaska, the king of federal funding. Though with Tubes behind bars, I imagine it will be less so.
I've had no problem with IG chips, they run all my software just fine and can even play Oblivion. Unless you're gaming or running some eye candy OS, I don't see why you'd need more.
Then again, I suppose I'm judging a discrete graphics card vs integrated as two very different things. If I have a discrete graphics card I expect it to do more than "even play Oblivion".
Hell they should just let people donate their spare cycles for cash. I'd do it.
Yeah turns out this problem is really hard and, depending on how you formalize it, uncomputable. What it boils down to is that the system needs to know how long it will take to run a program on itself and how long it will take to spin it off into another thread.
Sound familiar? It should; that's halt.
Of course, approximating it well is an interesting research problem in language theory, but I wouldn't expect a general purpose system in actual production use any time soon. Even doing it in a pure, strongly typed language will be extremely difficult; implementing one for something like Java/C[++] will be far harder. And of course the ultimate win is doing it to machine code.
MS Research is like a research university for all intents and purposes; they basically have academic latitude. Of course by the time the product reaches market it will be made, um..."better".
I was in awe about it. I had hopes for the ultimate field notebook finally being designed; netbook with crazy in voltage regulation, solar recharge capability, hand crank capability, minimal os almost no power use...The XO was crude but it had incredible potential. The mesh networking alone, if they could get adoption...insane!
If I had more time I'd fork off the project and work on this. Pricepoint $600 or so, a netbook intended for field work. Insane recharge capabilities, crazy networking, sun viewable screen, durable, super light (something the XO fails), etc.
I'd totally buy one. I'm looking forward to when we start seein ghigher end netbooks (perhaps when Apple or Thinkpad* comes out with one), where the focus is on quality and power in a small, light, durable package, rather than the cheapest computer possible.
*Lenovo does make netbooks under ideapad, but as anyone who has bought IBM knows, thinkpad and ideapad are a world apart in quality.
Someone has to feed the conspiracy theorists;) http://en.wikipedia.org/wiki/Information_Awareness_Office
Let me know how that 0% tax goes for you. http://dor.wa.gov/content/FindTaxesAndRates/IncomeTax/
I'm sure that has nothing to do with the auto industry, on which Michigan is pretty much entirely dependent, dying out entirely. No it's clearly the Pinko Commie Nazi taxes.
No, they were already there.
What about ones which can be either? Interactive ssh versus file transfer.
What I'd love would be if we could let users prioritize their own traffic (with sensible defaults for those who don't know what an internet is), and give them N GB high priority traffic per month and uncapped low priority.
Of course, there are tons of problems with this, but there are times I would willingly prioritize myself down (up/download I intend to leave overnight, 4 GB differential full system backup).
Never mind that with the way current userland and networking is set up this isn't feasible, I think it's a neat idea.
I use gmailfs you insensitive clod!
I doubt fatcat lobbyists care whether they're using AES256 or ROT13.
Ah. Not sure if I agree with that law, I can see some good arguments for and against it. I'd be surprised, but not terribly so, if it applied to less serious crimes.
Not only am I not a lawyer, I have no idea what I'm talking about! You should take my legal advice:-)