That sounds nice and well, but the fact is that your CD has already been distorted -- down to only two channels, only 44khz, only 16bit.
(if only they used a lossy compression scheme for CDs, instead of just truncating the audio to fit, then the same CDs that you have no could have held the same music but at much higher qualit -- maybe with more channels, or more than 16bit.)
We're STILL paying when we get them "free" through advertising. The money has to come from somewhere, after all. I remember looking at it back in 2000 when (in the US) about $6000 was spent per citizen on advertising. It'll be much higher now. So we're paying in the form of higher costs on every single product we buy.
Except the cost we pay through advertising is higher because it has to go through all those layers of middlemen. It's a kind of grossly inefficient redistributive taxation that takes money away from people who buy products and gives it to those who take advantage of "free" stuff.
Breaking news: the FDA has reclassified nicotine, heroin and alcohol as unrestricted food items in the same category as regular table salt. They justified the move with the assertion that "too much of anything is bad for you, be it salt or drugs or alcohol". When asked about the long history of research that shows that some substances are quantifiably more addictive than others, they dismissed it as "scientific technobabble."
"if a polite ad helps keep some site up and running, it costs me nothing."
So where does the ad money doesn't come, do you think? In 1999, spending in the US on advertising was about $6000 per capita. We ourselves are spending all this money, in the form of inflated product prices, to subject ourselves to the ads. How much of that money trickles into the websites we wish to support? I don't know, but I bet it's only a tiny fraction.
"The high instance of divorce in the US is much more related to materialism, disconnectedness (also called 'independence') and ideas of 'self', attitudes towards relationships and the myth of satisfaction than any scapegoat."
You can put arbitrarily large blocks of code inside the atomic{...} blocks. I indicated as much with ellipses. You can also nest atomic blocks. Also an atomic block can contain the statement "retry()" which rolls it back.
The STM machinery still needs to be able to roll-back out of an atomic block. This rules out certain code from appearing inside the atomic block, especially side-effecting code.
All STM platforms allow code that manipulates "transactional" heap variables, i.e. ones whose side-effects are logged and rollbackable. In these systems the idea is to use STMs mainly for concurrent data-structures, eg. queues and hash tables and so on.
Some platforms also allow side effects on transacted resources, e.g. a transacted filesystem or webserver or database.
Other platforms also allow side-effects due to file I/O: they log the inputs (so as to resignal them in case of rollback) and buffer the outputs (the ouputs only commit to disk after the atomic block has completed).
What are the proposed alternatives that simplify synchronization?
(1) Software Transactional Memory, STMs. You write "int STM x=5;" and then one thread can do "atomic {x+=1;}" and another does "atomic {...;x-=1;...}" and the runtime+compiler magically make the atomic blocks execute atomically. This is different from Java synchronization blocks because these can be executed optimistically in parallel and because they never result in deadlock. People therefore say that STMs are "compositional" in a way that locks are not. The key is that the compiler/runtime know how to roll back an atomic block. This kind of atomic block interfaces very gracefully with SQL transactions.
(2) Message-passing. Academically it's embodied in the "pi calculus". In web terms it's embodied in the W3C "choreography" working draft. In practical terms it's embodied in Microsoft's Biztalk and in Ericcson's Erlang language and in Microsoft's new Robotics SDK. It's also a little reminiscent of "tuple-space" operating systems. The idea is that threads communicate by sending messages to each other. It's still possible to deadlock (e.g. if one thread waits for a message that will never come) but these errors seem more rare in practice. Also it's easier to analyse for this kind of problem at compile-time than it is with synchronization.
(3) Write the code in a functional way, so the compiler infers parallelism automatically and you don't need to. Or, take existing C code and have the compiler infer dataflow in it, then proceed as above. e.g. WaveScaler at UW, I think. Many people think this is the best hope for the coming world of multi-core chips.
Touching the fruit is a bad way of assessing ripeness. Most nice fruits ripen only when they're still attached to the plant. Ripening means becoming sweeter and developing their aromas. Once they're plucked, the don't really ripen in a good way anymore, they just get mushy.
The best way to judge ripeness is with your nose. Sniff out the sweetest and nicest-smelling fruits. If it doesn't smell nice in the supermarket, it will never smell nicer in your kitchen.
(exception: banana will ripen at home as its starches are converted into sugars; and avocado only starts to ripen after it has been plucked).
I suspect it works the other way round... in a public place there is no reasonable expectation of privacy and so you're allowed to record. But inside a house, I reckon there is.
(similarly in 11 US states you're not allowed to record a telephone conversation unless both parties are aware of it, again because of the reasonable expectation of privacy. Your own end of your telephone counts as your "OWN FUCKING PROPERTY" but that's not enough to let you record.)
"MS isn't about to open up Outlook or Exchange" ???
On the contrary, Outlook+Exchange are pretty much entirely open. They're documented thoroughly in the book "Inside MAPI" and at msdn.microsoft.com.
MAPI exposes the generic interface between stores/services (e.g. Exchange, Personal Folders, local calendar) and client (e.g. Outlook). It encourages you to write your own rival stores and your own rival clients. Why has no one done this so far? -- presumably because rival clients would reinforce Exchange's hegemony. although lots of companies write their own rival stores related to their own business processes.
MAPI lets you write any client (as an alternative to Outlook) which interacts with your existing message+calendar+etc stores/services, and it lets you write any store/service (as an alternative to Exchange and Personal Folders) which interacts with Outlook and any other client.
"T-Mobile sends an SMS to my phone when it detects new mail on my POP3 server"...
Don't you hate it like this? My T616 phone polls my IMAP server for email every couple of minutes, for free, and downloads the new emails it finds onto the phone (or at least, a 5k portion of the text part of the email). It's a native TCP-speaking inhabitant of the internet and so email+web works independently of my phone provider.
Why should you pay the fee even if it costs them no extra to provide it? -- because with no fee, then each customer has an incentive to set the "QoS" bit on every packet they send. (First just a few users will buy some third-party "web-accelerator" software which sets the bit, and then more users will buy a cheaper version of the web-accelerator, and then everyone will be setting their QoS bit). The QoS bit will inevitably become meaningless. You need pricing to preserve it.
The french metro system has a service much like this. Some carriages are divided into two identical halves. You can buy an 10-euro to go in half "X", or a 5-euro ticket to go in half "Y". The result is a beautifully self-regulating pricing scheme: the cost is exactly equal to the "utility" function amongst the customers for how worthwhile they find the degree of uncrowdedness.
I attended a lecture today by someone who, using fMRI, had discovered (1) a part of the brain whose activity was proportional to the expected outcome of a bet, and (2) a part whose activity was proportional to the distribution of risk on that bet, and (3) the hint that maybe part 1 was actually responding to an economist's UTILITY function rather than the probabilist's expectancy function. [paper in "Science", Dec 9th 2005, pp. 1680-1683]
When you talk about a blanket "neural response" it doesn't mean anything. I'm sure they've got much more sophisticated statistics. And I'm sure they'll find brain areas that correlate with higher spending in response to ads, similar to the research above.
Then an insurance company will come along that offers phishing insurance against your "though fucking noogies". And most banks will start bundling this insurance as part of their basic deals to attract customers. And they'll attract more customers and get less bad press than the banks that don't. And so we'll end up at exactly where we are now.
I saw a copy of his original lab-notes on the eye-poking experiment. (I attended a talk by someone in the History and Philosophy of Science dept at Cambridge who was doing her research on them).
His actual goal was to prove that there are two kinds of motion: that caused by external forces and that caused by "the will". First he moved his eye using "his will" and made it point to an object, and he could see the object. Next he used an external force (the blunt pin) to move his eye to point to the object, and it looked blurred and out of focus. Therefore there really are these two different kinds of motion, as witnessed by their qualitatively different effects.
The article used the word "ALSO", so I think that the fifteen were charged with conspiracy (5 years) plus the actual infringement (3 years) for a total of 8 years. Sort of like a extra penalty if your crime actually came to fruition. Or, equivalently, you get a reduced sentence if the crime didn't succeed. My guess is that if someone is guilty of infringement then they are necessary guilty of conspiracy to infringe.
(that way it will fit on web pages, and no one will be suspicious of you embedding links, and also your private amazon browsing history won't be embedded in the url.)
Slashdot Mirror
Read the article. It says "The Arctic vault will act as a back-up store for a global network of seed banks financially supported by the trust."
Correlation may not imply causation?
-- true, but it's highly correlated with causation.
That sounds nice and well, but the fact is that your CD has already been distorted -- down to only two channels, only 44khz, only 16bit.
(if only they used a lossy compression scheme for CDs, instead of just truncating the audio to fit, then the same CDs that you have no could have held the same music but at much higher qualit -- maybe with more channels, or more than 16bit.)
We're STILL paying when we get them "free" through advertising. The money has to come from somewhere, after all. I remember looking at it back in 2000 when (in the US) about $6000 was spent per citizen on advertising. It'll be much higher now. So we're paying in the form of higher costs on every single product we buy.
Except the cost we pay through advertising is higher because it has to go through all those layers of middlemen. It's a kind of grossly inefficient redistributive taxation that takes money away from people who buy products and gives it to those who take advantage of "free" stuff.
The search field works like Run used to, for many tasks. e.g.:
Ctrl+Esc notepad ENTER
Ctrl+Esc http://slashdot.org/ ENTER
Ctrl+Esc itunes ENTER
So basically just don't press the "R" key like you used to!
Breaking news: the FDA has reclassified nicotine, heroin and alcohol as unrestricted food items in the same category as regular table salt. They justified the move with the assertion that "too much of anything is bad for you, be it salt or drugs or alcohol". When asked about the long history of research that shows that some substances are quantifiably more addictive than others, they dismissed it as "scientific technobabble."
"if a polite ad helps keep some site up and running, it costs me nothing."
So where does the ad money doesn't come, do you think? In 1999, spending in the US on advertising was about $6000 per capita. We ourselves are spending all this money, in the form of inflated product prices, to subject ourselves to the ads. How much of that money trickles into the websites we wish to support? I don't know, but I bet it's only a tiny fraction.
"Fakes by any standard"? "Since they were not authorized by the artist and are not numbered"?
Then that same standard says that pirated mp3s are fakes compared to the original mp3s. And seems a silly standard.
"The high instance of divorce in the US is much more related to materialism, disconnectedness (also called 'independence') and ideas of 'self', attitudes towards relationships and the myth of satisfaction than any scapegoat."
Quoting for truth. That's pretty darn insightful.
You can put arbitrarily large blocks of code inside the atomic{...} blocks. I indicated as much with ellipses. You can also nest atomic blocks. Also an atomic block can contain the statement "retry()" which rolls it back.
The STM machinery still needs to be able to roll-back out of an atomic block. This rules out certain code from appearing inside the atomic block, especially side-effecting code.
All STM platforms allow code that manipulates "transactional" heap variables, i.e. ones whose side-effects are logged and rollbackable. In these systems the idea is to use STMs mainly for concurrent data-structures, eg. queues and hash tables and so on.
Some platforms also allow side effects on transacted resources, e.g. a transacted filesystem or webserver or database.
Other platforms also allow side-effects due to file I/O: they log the inputs (so as to resignal them in case of rollback) and buffer the outputs (the ouputs only commit to disk after the atomic block has completed).
What are the proposed alternatives that simplify synchronization?
(1) Software Transactional Memory, STMs. You write "int STM x=5;" and then one thread can do "atomic {x+=1;}" and another does "atomic {...;x-=1;...}" and the runtime+compiler magically make the atomic blocks execute atomically. This is different from Java synchronization blocks because these can be executed optimistically in parallel and because they never result in deadlock. People therefore say that STMs are "compositional" in a way that locks are not. The key is that the compiler/runtime know how to roll back an atomic block. This kind of atomic block interfaces very gracefully with SQL transactions.
(2) Message-passing. Academically it's embodied in the "pi calculus". In web terms it's embodied in the W3C "choreography" working draft. In practical terms it's embodied in Microsoft's Biztalk and in Ericcson's Erlang language and in Microsoft's new Robotics SDK. It's also a little reminiscent of "tuple-space" operating systems. The idea is that threads communicate by sending messages to each other. It's still possible to deadlock (e.g. if one thread waits for a message that will never come) but these errors seem more rare in practice. Also it's easier to analyse for this kind of problem at compile-time than it is with synchronization.
(3) Write the code in a functional way, so the compiler infers parallelism automatically and you don't need to. Or, take existing C code and have the compiler infer dataflow in it, then proceed as above. e.g. WaveScaler at UW, I think. Many people think this is the best hope for the coming world of multi-core chips.
The water is more hygienic than toilet paper, and with the water jet you end up cleaner. Also it's a godsend for hemmeroids sufferers.
Touching the fruit is a bad way of assessing ripeness. Most nice fruits ripen only when they're still attached to the plant. Ripening means becoming sweeter and developing their aromas. Once they're plucked, the don't really ripen in a good way anymore, they just get mushy.
The best way to judge ripeness is with your nose. Sniff out the sweetest and nicest-smelling fruits. If it doesn't smell nice in the supermarket, it will never smell nicer in your kitchen.
(exception: banana will ripen at home as its starches are converted into sugars; and avocado only starts to ripen after it has been plucked).
The Starbucks drink sizes are already oversized compared to european servings, so I think the names are fair.
Italian small = Starbucks "short"
Italian medium = Starbucks "tall"
Italian large = Starbucks "grande"
Italian (doesn't exist) = Starbucks "venti"
I suspect it works the other way round... in a public place there is no reasonable expectation of privacy and so you're allowed to record. But inside a house, I reckon there is.
(similarly in 11 US states you're not allowed to record a telephone conversation unless both parties are aware of it, again because of the reasonable expectation of privacy. Your own end of your telephone counts as your "OWN FUCKING PROPERTY" but that's not enough to let you record.)
Site doesn't render correctly in IE...
"MS isn't about to open up Outlook or Exchange" ???
On the contrary, Outlook+Exchange are pretty much entirely open. They're documented thoroughly in the book "Inside MAPI" and at msdn.microsoft.com.
MAPI exposes the generic interface between stores/services (e.g. Exchange, Personal Folders, local calendar) and client (e.g. Outlook). It encourages you to write your own rival stores and your own rival clients. Why has no one done this so far? -- presumably because rival clients would reinforce Exchange's hegemony. although lots of companies write their own rival stores related to their own business processes.
MAPI lets you write any client (as an alternative to Outlook) which interacts with your existing message+calendar+etc stores/services, and it lets you write any store/service (as an alternative to Exchange and Personal Folders) which interacts with Outlook and any other client.
"T-Mobile sends an SMS to my phone when it detects new mail on my POP3 server"...
Don't you hate it like this? My T616 phone polls my IMAP server for email every couple of minutes, for free, and downloads the new emails it finds onto the phone (or at least, a 5k portion of the text part of the email). It's a native TCP-speaking inhabitant of the internet and so email+web works independently of my phone provider.
"i.e." stands for "id est" and means "that is [to say]".
"e.g." stands for "exempli gratia" and means "for example".
The article summary should have used e.g. instead of i.e. I see this mistake all the time and it irritates me.
Why should you pay the fee even if it costs them no extra to provide it? -- because with no fee, then each customer has an incentive to set the "QoS" bit on every packet they send. (First just a few users will buy some third-party "web-accelerator" software which sets the bit, and then more users will buy a cheaper version of the web-accelerator, and then everyone will be setting their QoS bit). The QoS bit will inevitably become meaningless. You need pricing to preserve it.
The french metro system has a service much like this. Some carriages are divided into two identical halves. You can buy an 10-euro to go in half "X", or a 5-euro ticket to go in half "Y". The result is a beautifully self-regulating pricing scheme: the cost is exactly equal to the "utility" function amongst the customers for how worthwhile they find the degree of uncrowdedness.
I attended a lecture today by someone who, using fMRI, had discovered (1) a part of the brain whose activity was proportional to the expected outcome of a bet, and (2) a part whose activity was proportional to the distribution of risk on that bet, and (3) the hint that maybe part 1 was actually responding to an economist's UTILITY function rather than the probabilist's expectancy function. [paper in "Science", Dec 9th 2005, pp. 1680-1683]
When you talk about a blanket "neural response" it doesn't mean anything. I'm sure they've got much more sophisticated statistics. And I'm sure they'll find brain areas that correlate with higher spending in response to ads, similar to the research above.
Then an insurance company will come along that offers phishing insurance against your "though fucking noogies". And most banks will start bundling this insurance as part of their basic deals to attract customers. And they'll attract more customers and get less bad press than the banks that don't. And so we'll end up at exactly where we are now.
I saw a copy of his original lab-notes on the eye-poking experiment. (I attended a talk by someone in the History and Philosophy of Science dept at Cambridge who was doing her research on them).
His actual goal was to prove that there are two kinds of motion: that caused by external forces and that caused by "the will". First he moved his eye using "his will" and made it point to an object, and he could see the object. Next he used an external force (the blunt pin) to move his eye to point to the object, and it looked blurred and out of focus. Therefore there really are these two different kinds of motion, as witnessed by their qualitatively different effects.
(I think he concluded that God exists.)
The article used the word "ALSO", so I think that the fifteen were charged with conspiracy (5 years) plus the actual infringement (3 years) for a total of 8 years. Sort of like a extra penalty if your crime actually came to fruition. Or, equivalently, you get a reduced sentence if the crime didn't succeed. My guess is that if someone is guilty of infringement then they are necessary guilty of conspiracy to infringe.
Do you know you can shorten amazon links? To just the first part:
http://www.amazon.com/gp/product/0471752738
(that way it will fit on web pages, and no one will be suspicious of you embedding links, and also your private amazon browsing history won't be embedded in the url.)