Yeah, I think where they failed was getting: 1) serious, prestigious courses; that are 2) taught by top professors. Intro psych taught by a not-particularly-famous psych professor is not going to draw. But I think the concept could've worked if they had, say, a solid bioinformatics course taught by someone like David Haussler. There are plenty of UC professors across the campuses with similar draw potential. But they need to target higher-prestige than generic intro courses, because nobody is going to pay $1400 for precalc.
Mostly a rumor, but: I heard some discussions about it from some academic colleagues, and afaict part of the problem was that there was no real framework for actually setting up the courses. They were hoping profs would volunteer to do it, without providing any real guidance or resources, and you would teach it in addition to your regular teaching. Unsurprisingly, they didn't get a lot of volunteers.
Indexing on data sets of that size is itself a pretty big challenge. You don't want an index that takes years to build, and it doesn't do much good if it's so huge that it is itself super-slow to access.
There is some research [pdf] on making compressed full-text indexes, but much of it is still research-level.
That's the main problem, yes. In CS it's slightly better, because there's a constant outflux of PhDs to industry: every time you see one of those announcements about Google hiring a professor away from CMU, that's one more academic job freed up. But in physics the imbalance is much worse, and slogging away in a series of postdocs hoping for something to open up is the usual course.
I mostly agree, but I haven't personally met anyone older who's been happy with dictating to a text-to-speech program unless they literally can't type. But most of the people I know who tried tried it a while ago, so maybe the programs are better now?
One of my relatives uses a tablet for most things: web browsing, reading emails, writing short replies, making notes, watching videos, etc. But she still goes up to an ancient desktop to type out longer emails.
If you are looking for a low stress profession you may as well choose the academic career and opt to avoid any stress within it. Chances are you will succeed
No, chances are that you'll fail, because you will never get tenure if you take a low-stress, laid-back approach to the job, unless you're at a community college perhaps. The academic career is completely organized around deadlines and management: the NSF grant application deadlines, conference paper deadlines, hiring grad students and postdocs, etc. In CS at least, if you don't bring in substantial funding, crank out many publications, and support a large-ish lab of students and postdocs (who you have to pay for!), you won't get tenure, and therefore won't have a job very long.
In particular, from the perspective whether a crime was committed, a common mistake is someone underestimating their BAC, so they think they are not driving drunk when in fact they are (this is different from subjective impairment). If you thought you had a BAC of 0.07 and it was actually 0.09, that's an error that takes your behavior from legal to illegal in a state where 0.08 is the legal limit.
One way to reduce that particular kind of error is to carry a breathalyzer with you, but that's not very common.
To most civilized parts of the world U.S. prisons look positively barbaric, though. Most first-world countries have managed to figure out how to run a prison system without high levels of rape, for one thing.
I'm not sure what else you could do for a 128-bit address. The format isn't inherently any more complex, just longer: instead of four 8-bit numbers separated by dots, it's eight 16-bit numbers separated by colons.
If you have some kind of regularity in the addresses, there are also alternate formats you can use, if you find it more convenient, to try to make them shorter and easier to type. For example, you can omit segments that are 0, and collapse consecutive such segments, which is why you can write the loopback address as::1.
Considering that through much of history men have married women with lower levels of educational attainment and income, and been able to be happy in those relationships without considering their wives "low-lives", I'm not sure why the reverse would be impossible.
All Google agreed was that the patents it holds which are essential to the implementation of certain mobile-telephony standards will be licensed under FRAND terms. They didn't agree to let them be used for free or anything.
Why weren't those already the terms? Standards bodies are supposed to, if they're doing their job, approve standards with some kind of FRAND licensing condition, in order for the standards to actually function as standards. The point of a standard is that everyone making a device with a certain kind of functionality is supposed to conform to certain agreed ("standard") behavior. To do that, they have to be able to legally able to implement the standard, which means any patents essential to the implementation need to be generally available to any third party for licensing, on reasonable and nondiscriminatory terms.
This magazine (Reason) is clearly against public-private partnerships, if anything seeing them as even worse than regular purely-government projects, due to the increased potential for this kind of cronyist corruption (but still inferior, in their view, to purely private-sector endeavors). A lot of free-market conservatives see them as an improvement on purely-government projects, though: public-private partnerships are often promoted as a way to bring efficiencies of the private sector into infrastructure and development (e.g. a pubic-private partnership to build a toll road), rather than having the "big government" do it all.
Are those two different camps of the economic right? Perhaps "libertarians" versus "pro-business conservatives" or something along those lines?
Yeah, that kind of procedural generation is also pretty interesting, but I think of it as a bit different. It's sometimes grouped under "procedural content generation" (PCG), i.e. the content of a game-world is generated: names, maps, etc. Even stuff like SpeedTree might go under that, since it procedurally generates, well, trees.
What I was trying to pull out here are systems that generate the rules or mechanics of the game, rather than the content. Admittedly the distinction can get hazy, because there's often some interdependency.
Considering how angry the gun-owners are about the publication, and how not-angry non-gun-owners are about it, empirically not many people seem to be worried about your version of the question.
I can't say I'd mind getting rid of the stacks of paper HR sends me.
In theory I'd agree, but in practice so far these have been replaced, in my experience, with things that are even worse than receiving stacks of paper:
1. Far too many emails.
2. Online systems that are damn near impossible to use. As an example, the former system we used for hiring was that I got a stack of resumes with cover letters, on paper, in my internal mailbox. The paperless system we have moved to, "HR Manager", through some combination of its design and/or our HR department's configuration of it, results in me needing to click through about 6 menus and select a bunch of options just to see the list of people who applied for a position. And then more if I want to actually download PDFs of their resumes and cover letters.
It's catching on a bit later in sports other than baseball due to the difficulty (until recently) of collecting fine-grained statistics in many sports. Baseball is fairly discrete: it operates pitch by pitch, with a lot of down time in between. So a large number of relevant statistics can be tallied by hand, which is why we have piles of statistics dating back decades. For every pitch, you can mark down whether it was a strike or ball, whether the batter swung, where in the field it went to if hit, what the fielder did with it, what the runners did in response, etc.
For football (and even more so, soccer), a lot of the relevant information you'd get from watching replays is more "continuous" and harder to extract manually. Traditional statistics did measured things like passing completion percentage and yards gained by a running back, but they didn't collect data that could be used to quantify things like the quality of an offensive line, or of blockers, except indirectly through overall team performance. Now a lot more of that information is being automatically tallied using computer-vision algorithms churning through digitized camera footage.
Although that's been an effect, and is probably a major motivation these days, the original purpose of the federal government giving states highway grants wasn't to control their drinking laws, but to build the Interstate Highway system. Eisenhower set up that funding mechanism a half-century ago, where the federal government designed and paid for the interstate highway system at an overall level, but individual segments were constructed and maintained by the states they ran through, using funds that transferred to states from the federal government.
Slashdot Mirror
But then how are we going to keep the birth rate up?
Yeah, I think where they failed was getting: 1) serious, prestigious courses; that are 2) taught by top professors. Intro psych taught by a not-particularly-famous psych professor is not going to draw. But I think the concept could've worked if they had, say, a solid bioinformatics course taught by someone like David Haussler. There are plenty of UC professors across the campuses with similar draw potential. But they need to target higher-prestige than generic intro courses, because nobody is going to pay $1400 for precalc.
Mostly a rumor, but: I heard some discussions about it from some academic colleagues, and afaict part of the problem was that there was no real framework for actually setting up the courses. They were hoping profs would volunteer to do it, without providing any real guidance or resources, and you would teach it in addition to your regular teaching. Unsurprisingly, they didn't get a lot of volunteers.
Indexing on data sets of that size is itself a pretty big challenge. You don't want an index that takes years to build, and it doesn't do much good if it's so huge that it is itself super-slow to access.
There is some research [pdf] on making compressed full-text indexes, but much of it is still research-level.
Not to mention that in practice you could probably just target a recent version of Ubuntu and reach most potential customers.
That's the main problem, yes. In CS it's slightly better, because there's a constant outflux of PhDs to industry: every time you see one of those announcements about Google hiring a professor away from CMU, that's one more academic job freed up. But in physics the imbalance is much worse, and slogging away in a series of postdocs hoping for something to open up is the usual course.
Their plans to end the world have failed, so they're retooling with more power to try again!
I mostly agree, but I haven't personally met anyone older who's been happy with dictating to a text-to-speech program unless they literally can't type. But most of the people I know who tried tried it a while ago, so maybe the programs are better now?
One of my relatives uses a tablet for most things: web browsing, reading emails, writing short replies, making notes, watching videos, etc. But she still goes up to an ancient desktop to type out longer emails.
No, chances are that you'll fail, because you will never get tenure if you take a low-stress, laid-back approach to the job, unless you're at a community college perhaps. The academic career is completely organized around deadlines and management: the NSF grant application deadlines, conference paper deadlines, hiring grad students and postdocs, etc. In CS at least, if you don't bring in substantial funding, crank out many publications, and support a large-ish lab of students and postdocs (who you have to pay for!), you won't get tenure, and therefore won't have a job very long.
In particular, from the perspective whether a crime was committed, a common mistake is someone underestimating their BAC, so they think they are not driving drunk when in fact they are (this is different from subjective impairment). If you thought you had a BAC of 0.07 and it was actually 0.09, that's an error that takes your behavior from legal to illegal in a state where 0.08 is the legal limit.
One way to reduce that particular kind of error is to carry a breathalyzer with you, but that's not very common.
There are a reasonable number of subsistence farmers in the US. Visit Appalachia or the Mississippi valley along the MS/AL border sometime.
To most civilized parts of the world U.S. prisons look positively barbaric, though. Most first-world countries have managed to figure out how to run a prison system without high levels of rape, for one thing.
I'm not sure what else you could do for a 128-bit address. The format isn't inherently any more complex, just longer: instead of four 8-bit numbers separated by dots, it's eight 16-bit numbers separated by colons.
If you have some kind of regularity in the addresses, there are also alternate formats you can use, if you find it more convenient, to try to make them shorter and easier to type. For example, you can omit segments that are 0, and collapse consecutive such segments, which is why you can write the loopback address as ::1.
Considering that through much of history men have married women with lower levels of educational attainment and income, and been able to be happy in those relationships without considering their wives "low-lives", I'm not sure why the reverse would be impossible.
Disheveled hipsters making six figures, who can't show up on time, run silicon valley. ;-)
All Google agreed was that the patents it holds which are essential to the implementation of certain mobile-telephony standards will be licensed under FRAND terms. They didn't agree to let them be used for free or anything.
Why weren't those already the terms? Standards bodies are supposed to, if they're doing their job, approve standards with some kind of FRAND licensing condition, in order for the standards to actually function as standards. The point of a standard is that everyone making a device with a certain kind of functionality is supposed to conform to certain agreed ("standard") behavior. To do that, they have to be able to legally able to implement the standard, which means any patents essential to the implementation need to be generally available to any third party for licensing, on reasonable and nondiscriminatory terms.
This magazine (Reason) is clearly against public-private partnerships, if anything seeing them as even worse than regular purely-government projects, due to the increased potential for this kind of cronyist corruption (but still inferior, in their view, to purely private-sector endeavors). A lot of free-market conservatives see them as an improvement on purely-government projects, though: public-private partnerships are often promoted as a way to bring efficiencies of the private sector into infrastructure and development (e.g. a pubic-private partnership to build a toll road), rather than having the "big government" do it all.
Are those two different camps of the economic right? Perhaps "libertarians" versus "pro-business conservatives" or something along those lines?
Yeah, that kind of procedural generation is also pretty interesting, but I think of it as a bit different. It's sometimes grouped under "procedural content generation" (PCG), i.e. the content of a game-world is generated: names, maps, etc. Even stuff like SpeedTree might go under that, since it procedurally generates, well, trees.
What I was trying to pull out here are systems that generate the rules or mechanics of the game, rather than the content. Admittedly the distinction can get hazy, because there's often some interdependency.
Considering how angry the gun-owners are about the publication, and how not-angry non-gun-owners are about it, empirically not many people seem to be worried about your version of the question.
Employees don't want to work for HP anymore, and HP gets closer to its "restructuring targets" without even having to fire them!
AAAI set up a General Gameplaying Competition in 2005, so some interest did belatedly develop.
There was a story last year about how digitizing industrial-plant blueprints in the 1990s "paperless office" push worked out for 'em...
In theory I'd agree, but in practice so far these have been replaced, in my experience, with things that are even worse than receiving stacks of paper:
1. Far too many emails.
2. Online systems that are damn near impossible to use. As an example, the former system we used for hiring was that I got a stack of resumes with cover letters, on paper, in my internal mailbox. The paperless system we have moved to, "HR Manager", through some combination of its design and/or our HR department's configuration of it, results in me needing to click through about 6 menus and select a bunch of options just to see the list of people who applied for a position. And then more if I want to actually download PDFs of their resumes and cover letters.
It's catching on a bit later in sports other than baseball due to the difficulty (until recently) of collecting fine-grained statistics in many sports. Baseball is fairly discrete: it operates pitch by pitch, with a lot of down time in between. So a large number of relevant statistics can be tallied by hand, which is why we have piles of statistics dating back decades. For every pitch, you can mark down whether it was a strike or ball, whether the batter swung, where in the field it went to if hit, what the fielder did with it, what the runners did in response, etc.
For football (and even more so, soccer), a lot of the relevant information you'd get from watching replays is more "continuous" and harder to extract manually. Traditional statistics did measured things like passing completion percentage and yards gained by a running back, but they didn't collect data that could be used to quantify things like the quality of an offensive line, or of blockers, except indirectly through overall team performance. Now a lot more of that information is being automatically tallied using computer-vision algorithms churning through digitized camera footage.
Yes, you can flip a "developer mode" switch to disable the hardware lockdown, and install a "regular" distribution of Linux like Ubuntu.
Although that's been an effect, and is probably a major motivation these days, the original purpose of the federal government giving states highway grants wasn't to control their drinking laws, but to build the Interstate Highway system. Eisenhower set up that funding mechanism a half-century ago, where the federal government designed and paid for the interstate highway system at an overall level, but individual segments were constructed and maintained by the states they ran through, using funds that transferred to states from the federal government.