What you say would be true if stocks were valued according to expected long-term return, but they aren't. They're primarily valued circularly: based on expected value in the next 3-12 months.
AMD's double-point floating point performance is already great. What they lack is the rest of it. The programming model is pretty bad compared to CUDA (nobody is using Brook+), and they seem to be basically waiting for OpenCL to fix that. The bottlenecks in most attempts to use AMD chips for GPGPU code are also not really the floating-point units themselves, but the rest of the architecture; it's hard to keep the ALUs fed with your data without a magic compiler, a better programming model, a better architecture, or some combination of those.
I think there's a potential niche for Facebook games, but there really has to be some reason for them to be on Facebook--- not just using Facebook as a generic delivery platform. The ones that attract interest usually do something with the social network; things like Parking Wars, among a number of examples. These are usually fairly lightweight games that try out some idea about how to turn social networks into game mechanics, not heavyweight games that just use the social network as an advertising tool.
Of course, I'd say the number of interesting ways to turn social networks into game mechanics that have appeared so far is fairly small, and most of them are good mainly for novelty value and get old after you've figured out the dynamics of how they work and spread. But you can make some money at it.
The people who take a "humanity is defined by mental personhood" position consistently do tend to argue that certain levels of mental incapacity suffice to make someone no longer a human in the morally relevant sense. Not a very popular position, so no politicians that I know of argue that, but I do tend to respect the philosophers who do (Peter Singer being the canonical example).
Interest in "diversity" is semi-frequently used by left-leaning college types as a reason to oppose conservative speakers whose views they find offensive.
(I'm left-leaning myself, FWIW, but I find this particular strain of lefties to be as much opposed to my views as right-wingers are, since indeed "free speech was intended to protect offensive speech".)
I don't run any extensions, and I regularly see that sort of thing (currently 1.2 GB after 3 days open and about 20 tabs). However, it seems worst under 64-bit Linux; my 32-bit PPC OS X and 32-bit Windows installs are much better, and 32-bit Linux install is marginally better.
The 64-bit part isn't too hard to believe; lots of stuff doubles in size right off the bat. I'm not sure if the Linux-being-worse aspects are something to do with Firefox, or some issue with the Linux/X graphics stack (which is notoriously broken).
I'm no fan of Mormonism, or Catholicism or Islam for that matter, all of which are backwards, right-wing religions, so I agree with most of your post.
But I don't really see what this particular dispute over trademarks has to do with Mormonism. Whether non-owners of a trademark paying for search results under those terms as keywords is, or ought to be, a violation of trademark law has been argued over in a number of states, and I don't see particularly clear religious faultlines in that debate. If supporting that position is somehow related to Mormonism, does that mean that eBay is run by Mormons?
PhD students in the sciences and engineering do not pay for their education. Nearly 100% of them are funded in one way or another, whether it's fellowships, research assistantships, or teaching assistantships.
I'm sure most of those H1-B's would apply for regular green cards, without the disadvantages you cite, if it were possible to get one without winning a long-odds lottery.
In most industrial settings, if something's built to a specification, and it's later discovered to have failed to meet the specification, the vendor's still at least partly liable, even if the customer failed to discover the defect in initial validation.
Ok, minor nit in my own post, I suppose strcmp() as C defines it actually wouldn't be faster, since you'd still have to scan through to decide whether to return +1 or -1 for non-equal strings. A function that only looks for string equality/nonequality is considerably faster with many Pascal-style strings, though.
Also, strcmp() is much faster for strings that have non-equal length but long common prefixes, and strcat() onto a long string is much faster since you can jump right to the place to begin appending at.
"The actual value of a company or an asset based on an underlying perception of its true value including all aspects of the business, in terms of both tangible and intangible factors."
A site with millions of daily visitors has a non-zero intangible value.
I've yet to see a really top-notch person in EE or CS who wasn't actually interested in it, often from a fairly young age. The people who decide at 18 that they're going to do EE, and treat it as if it were a pre-med-style major, may get good grades, but aren't good at it.
If it weren't a state, it wouldn't be able to seat Senators, for example, since the U.S. Constitution says there are "two Senators from each state", and does not apportion any to Commonwealths.
I'm not sure if it's worth the $500m price being thrown around, but if Twitter's only goal was to make some money without totally alienating its users, it doesn't seem that hard. Put a few ads on the website, and even if only 10% of users go to the website, it's still quite a few. Offer a few minor premium services for $5/mo. Etc.
It's not really that hard to make money at Web 2.0 (or even Web 1.0). The companies that fail are the ones that try to exploit it too greedily and drive off their users. I mean, Craigslist is probably a poster boy for the totally minimalist style of monetization (i.e. do it as little as possible), and they still pull in a good $200m a year or so.
A website with millions of daily visitors has some non-zero intrinsic value, even without considering any potential revenue streams, just due to the fact that there are a lot of people out there who like soapboxes, and something with millions of visitors has value as a good soapbox.
I would personally pay at least $50 for Twitter if nobody else outbid me, just so I could own it and fuck around with it. Probably other wealthy people would pay more for similar vanity or having-fun-with-it reasons.
I think there was a (mostly popped) bubble of speculative investment in Web 2.0 type companies, where VCs would throw money at anything with "social media" in the title. Around 2006-2008 it didn't take a lot of effort or a very plausible-sounding business plan to get backing in that space.
It's less visible in valuations than the late-90s bubble was, because this time around most of the companies never IPOd, and there were fewer to begin with. But I guess we could speculate as to their current values. Bebo was acquired for $850m; what's it worth today? What about Skype; is it still worth the $2.5b that it was acquired for? Many of the companies that did IPO in the past few years are down much more than the stock-market average; e.g. VMWare is down 80%.
I personally would, but I suspect many people wouldn't. For me though, when something is mainly the application of a complex set of rules, I trust a computer to apply them faithfully much more than I trust a doctor to recall them (often from memory) out of some book he may or may not have read recently. In particular, if a diagnosis requires a bunch of factors to go into it, humans (including doctors) are notoriously bad at estimating conditional probabilities, whereas computers are pretty good.
Computers have already been better at diagnosis than most doctors in some limited areas for decades: the early-70s MYCIN system matched experts and did better than the average doctor when diagnosing blood infections.
Now if for various superstitious reasons people want a person in the loop, fine, I don't care that much. But I want to know that the person is just operating the computer and giving me it output, not applying their own hokey judgment.
Well Mr. Smith we'd like to run a series of tests to see what illness you have, but Obama and Congress passed a healthcare bill that limits us to run just one test. We ran it for lupis, but it came back negative, and since we ran one test, we cannot run another.
Comparative-effectiveness research doesn't have much to do with that scenario. Instead, the purpose of this bill is to fund research into, well, comparative effectiveness of treatments. Say you have some particular condition, and there are three possible treatments in reasonably common use. Which one should your doctor prefer? Is there any scientific reason to think one of them is the better treatment to try, or at least to try first? That's what CER aims to tell you.
The current, relatively poor, practice is that doctors get bombarded with information from pharmaceutical companies about how their patented drug is the treatment to prefer, which they can cite various studies backing up, which you probably don't have time to track down. A center for CER would, hopefully, track them down and summarize the results for you.
Slashdot Mirror
What you say would be true if stocks were valued according to expected long-term return, but they aren't. They're primarily valued circularly: based on expected value in the next 3-12 months.
AMD's double-point floating point performance is already great. What they lack is the rest of it. The programming model is pretty bad compared to CUDA (nobody is using Brook+), and they seem to be basically waiting for OpenCL to fix that. The bottlenecks in most attempts to use AMD chips for GPGPU code are also not really the floating-point units themselves, but the rest of the architecture; it's hard to keep the ALUs fed with your data without a magic compiler, a better programming model, a better architecture, or some combination of those.
I think there's a potential niche for Facebook games, but there really has to be some reason for them to be on Facebook--- not just using Facebook as a generic delivery platform. The ones that attract interest usually do something with the social network; things like Parking Wars, among a number of examples. These are usually fairly lightweight games that try out some idea about how to turn social networks into game mechanics, not heavyweight games that just use the social network as an advertising tool.
Of course, I'd say the number of interesting ways to turn social networks into game mechanics that have appeared so far is fairly small, and most of them are good mainly for novelty value and get old after you've figured out the dynamics of how they work and spread. But you can make some money at it.
The people who take a "humanity is defined by mental personhood" position consistently do tend to argue that certain levels of mental incapacity suffice to make someone no longer a human in the morally relevant sense. Not a very popular position, so no politicians that I know of argue that, but I do tend to respect the philosophers who do (Peter Singer being the canonical example).
Interest in "diversity" is semi-frequently used by left-leaning college types as a reason to oppose conservative speakers whose views they find offensive.
(I'm left-leaning myself, FWIW, but I find this particular strain of lefties to be as much opposed to my views as right-wingers are, since indeed "free speech was intended to protect offensive speech".)
I don't run any extensions, and I regularly see that sort of thing (currently 1.2 GB after 3 days open and about 20 tabs). However, it seems worst under 64-bit Linux; my 32-bit PPC OS X and 32-bit Windows installs are much better, and 32-bit Linux install is marginally better.
The 64-bit part isn't too hard to believe; lots of stuff doubles in size right off the bat. I'm not sure if the Linux-being-worse aspects are something to do with Firefox, or some issue with the Linux/X graphics stack (which is notoriously broken).
And it's also on there; stuff, like Nirvana and Smashing Pumpkins that basically everyone who grew up in the early/mid 90s at least recognizes.
I'm no fan of Mormonism, or Catholicism or Islam for that matter, all of which are backwards, right-wing religions, so I agree with most of your post.
But I don't really see what this particular dispute over trademarks has to do with Mormonism. Whether non-owners of a trademark paying for search results under those terms as keywords is, or ought to be, a violation of trademark law has been argued over in a number of states, and I don't see particularly clear religious faultlines in that debate. If supporting that position is somehow related to Mormonism, does that mean that eBay is run by Mormons?
You must live in a weird location if Jew-bashing, for example, is socially acceptable. In some countries, it's not even legal!
Or at least I was told that it's no longer appropriate to refer to Catholics as Papists. :(
PhD students in the sciences and engineering do not pay for their education. Nearly 100% of them are funded in one way or another, whether it's fellowships, research assistantships, or teaching assistantships.
I'm sure most of those H1-B's would apply for regular green cards, without the disadvantages you cite, if it were possible to get one without winning a long-odds lottery.
In most industrial settings, if something's built to a specification, and it's later discovered to have failed to meet the specification, the vendor's still at least partly liable, even if the customer failed to discover the defect in initial validation.
Ok, minor nit in my own post, I suppose strcmp() as C defines it actually wouldn't be faster, since you'd still have to scan through to decide whether to return +1 or -1 for non-equal strings. A function that only looks for string equality/nonequality is considerably faster with many Pascal-style strings, though.
Also, strcmp() is much faster for strings that have non-equal length but long common prefixes, and strcat() onto a long string is much faster since you can jump right to the place to begin appending at.
for providing a defective product?
"The actual value of a company or an asset based on an underlying perception of its true value including all aspects of the business, in terms of both tangible and intangible factors."
A site with millions of daily visitors has a non-zero intangible value.
I've yet to see a really top-notch person in EE or CS who wasn't actually interested in it, often from a fairly young age. The people who decide at 18 that they're going to do EE, and treat it as if it were a pre-med-style major, may get good grades, but aren't good at it.
If it weren't a state, it wouldn't be able to seat Senators, for example, since the U.S. Constitution says there are "two Senators from each state", and does not apportion any to Commonwealths.
I'm not sure if it's worth the $500m price being thrown around, but if Twitter's only goal was to make some money without totally alienating its users, it doesn't seem that hard. Put a few ads on the website, and even if only 10% of users go to the website, it's still quite a few. Offer a few minor premium services for $5/mo. Etc.
It's not really that hard to make money at Web 2.0 (or even Web 1.0). The companies that fail are the ones that try to exploit it too greedily and drive off their users. I mean, Craigslist is probably a poster boy for the totally minimalist style of monetization (i.e. do it as little as possible), and they still pull in a good $200m a year or so.
A website with millions of daily visitors has some non-zero intrinsic value, even without considering any potential revenue streams, just due to the fact that there are a lot of people out there who like soapboxes, and something with millions of visitors has value as a good soapbox.
I would personally pay at least $50 for Twitter if nobody else outbid me, just so I could own it and fuck around with it. Probably other wealthy people would pay more for similar vanity or having-fun-with-it reasons.
I think there was a (mostly popped) bubble of speculative investment in Web 2.0 type companies, where VCs would throw money at anything with "social media" in the title. Around 2006-2008 it didn't take a lot of effort or a very plausible-sounding business plan to get backing in that space.
It's less visible in valuations than the late-90s bubble was, because this time around most of the companies never IPOd, and there were fewer to begin with. But I guess we could speculate as to their current values. Bebo was acquired for $850m; what's it worth today? What about Skype; is it still worth the $2.5b that it was acquired for? Many of the companies that did IPO in the past few years are down much more than the stock-market average; e.g. VMWare is down 80%.
Is that Peter "there's absolutely no bubble in technology" Thiel? Good luck with that...
I personally would, but I suspect many people wouldn't. For me though, when something is mainly the application of a complex set of rules, I trust a computer to apply them faithfully much more than I trust a doctor to recall them (often from memory) out of some book he may or may not have read recently. In particular, if a diagnosis requires a bunch of factors to go into it, humans (including doctors) are notoriously bad at estimating conditional probabilities, whereas computers are pretty good.
Computers have already been better at diagnosis than most doctors in some limited areas for decades: the early-70s MYCIN system matched experts and did better than the average doctor when diagnosing blood infections.
Now if for various superstitious reasons people want a person in the loop, fine, I don't care that much. But I want to know that the person is just operating the computer and giving me it output, not applying their own hokey judgment.
Comparative-effectiveness research doesn't have much to do with that scenario. Instead, the purpose of this bill is to fund research into, well, comparative effectiveness of treatments. Say you have some particular condition, and there are three possible treatments in reasonably common use. Which one should your doctor prefer? Is there any scientific reason to think one of them is the better treatment to try, or at least to try first? That's what CER aims to tell you.
The current, relatively poor, practice is that doctors get bombarded with information from pharmaceutical companies about how their patented drug is the treatment to prefer, which they can cite various studies backing up, which you probably don't have time to track down. A center for CER would, hopefully, track them down and summarize the results for you.