The poor review assignment at large conferences contributes to that effect as well, I think. I almost always have at least one of three reviewers, and sometimes even two of three, give a noncommittal review along the lines of, "well this isn't really my area, but it seems pretty good". Those reviews basically are non-reviews, so the acceptance decision is then entirely up to the remaining one or two reviewers. So it often comes down to: did the one person who actually provided an opinion on your paper like it or not like it?
In my experience that's often pretty subjective, especially for conferences with tight length limits (standard in AI is six pages). If the reviewer personally found the paper to be on an interesting subject with an interesting approach that he/she felt should be investigated, almost any shortcomings can be excused, and the reviewer will conclude that "Overall, this paper provides a valuable contribution to an important ongoing discussion in this area." But if the reviewer doesn't like it, finds it boring, dislikes the approach, etc., it's easy to find something that had insufficient detail, didn't sufficiently distinguish from related work, didn't sufficiently motivate the problem or investigate/validate the applications, etc., etc., since you really can't fit that much in six pages.
That's the point of "has no effect"--- the disagree mod wouldn't actually reduce a comment's score, it would just waste a mod point of the person who tried to use it, thereby removing mod points from people who shouldn't have them.
As a computer scientist, my impression is that the program committees really are pretty random, or at least based on some sort of preference other than a widely agreed "quality" standard. Try it sometime: resubmit a paper rejected from a top CS conference verbatim to another top CS conference. The correlation between the reviews is usually quite low, both in terms of the numerical scores, and especially in terms of what they liked / complained about.
If you want to go down that route, here's a syllabus from a course that's been taught a few times using Game Maker (also to mainly non-CS students), which might be useful to get ideas.
You don't think a game designer should know something about how interactive systems, procedures, dialog trees, preconditions, etc. work? I mean, you can't design a good interactive experience without at least having a vague idea about interaction and computation, even if it's mainly at a pseudocode level.
Isn't not just small to medium sized businesses; most tech companies, even really huge ones, don't buy this kind of enterprise equipment. You won't find any of it at Google or Amazon, for example, even though they are quite large.
Are they really that superior to the Sun storage products (the ones Sun invented ZFS for) to be worth the big multiple in price? I mean, Sun isn't stuff-you-put-together-at-Frys prices either, but it's a lot cheaper than EMC or NetApp.
That's still a particular type of compression, isn't it? I mean, I can buy giving it a new name, since it has a bunch of infrastructure around it, but it's a perfectly general kind of data-compression algorithm as well, even if not the world's most efficient: break the data into fixed-size blocks, then, for any blocks that appear more than once, replace all occurrences after the first with a pointer to the first. Block-based RLE compression is basically a simpler version of that (where you only deduplicate the blocks when they appear in sequence).
I dunno, at places I've been the low-level sysadmin access is not very closely monitored. "Official" access through the normal APIs is logged and monitored, but when the Unix sysadmin has root on the database machine, he could be grepping through the database for all anybody knows.
Cross-platformness in a radical sense (all hardware, all operating systems) does seem to be quickly falling by the wayside, and not just with IE only running on Windows..
The Apple version of Webkit (Safari) of course only runs on OSX, or OSX+iOS if you count Mobile Safari as the same browser. Chrome runs on the three major OSs, but only x86, x86-64, and ARM architectures, and is hard to port, due to generating machine code in its Javascript engine. Opera runs on x86, x86-64, ARM, and SuperH, and is reportedly somewhat easier to port, but it's closed-source so who knows. Firefox 4 will run only on x86 and x86-64.
So Firefox 3.6.x may be the last modern web browser that runs basically everywhere. You can get binaries for all major platforms, and Debian currently ships it for all 8 of its supported architectures: x86, x86-64, alpha, ARM, MIPS, PowerPC, S390 (!), and SPARC.
Sort of step backwards from the original Unix solution to portability: you write your stuff in C+POSIX, and then it runs everywhere we've ported a C compiler and a POSIX layer. Now apps are sprouting their own architecture-specific virtual machines! Perhaps LLVM will save us? It'd be nice if we managed to agree again on a single point of porting, so instead of saying "Chrome runs on x86, x86-64, and ARM, Firefox runs on x86 and x86-64", you can say "Browser Foo runs on anything with an LLVM port".
They're symmetric, though, which might not matter for many people, but I find nice. The 30 Mbps lowest tier is 30 Mbps each way, whereas Comcast's 30 Mbps service is 30 down, 7 up.
This rule appears to apply only to scientists who directly work for a government agency as employees, though, not to professors who are funded by federal grant money, or even professors who teach at public universities.
I agree that this is a weird/annoying feature, but it's existed in other games too, and doesn't seem to offend people (including me) as much there. For example, Magic: The Gathering has exactly that property where you can spend money to buy better cards. Is it that they've balanced it better, so the proportional effect of card-buying is less (you can't purely buy yourself victory)? I do think they've moved in that direction with the current set of rules, which is probably a good thing, but back in the days I played (mid-90s, mostly Revised/3rd Ed), there was a quite big effect to be had from buying a handful of powerful/rare/expensive cards, and people didn't seem to think that ruined the game.
I have some CFLs and LEDs, but I really hate them. The light is ridiculously non-broad-spectrum, though the CFLs in particular are better than they used to be. They're now good enough for basic house lighting imo, e.g. I have them in my ceiling sockets (incandescents only for reading lamps).
I guess I don't see it as a particularly large usage, especially by the standards here in the US. Maybe my choice of lightbulbs is eccentric and wasteful, but so is my neighbor's choice of a big-screen TV, which uses more power than my lightbulbs. Oh sure, the neighbor will probably claim that a smaller TV just isn't the same, but given that his TV is too big for his living room anyway, I find that claim ludicrous. Now maybe people find my claim that incandescents are much more pleasant than CFLs also ludicrous, so I suppose we each have our oddities. Until his big-screen TV is banned, I don't feel bad with my light bulbs.;-)
In any case, I think we should go after total usage. Your 5 kWh/day is an impressive target, but the U.S. is way off of that: the average residential energy usage in the U.S. is 35 kWh/day. We could even be generous and start going after only the really absurd users. For example, slap a luxury tax on houses using more than 100 kWh/day (they actually exist!).
Indeed, that's the main viewpoint I think needs to be changed. Or, the amount people have to pay for using more than their fair share of energy should be much higher.
For example, I would support a %-of-income penalty for very high users, to make sure even rich people were deterred. If you're in, say, the 99.9th percentile of energy users for your area, you pay 5% of your income as a penalty.
The top 1% of residential energy users are using a very disproportionate share of the total usage, so I think those are the people we should go after, not trying to squeeze a few extra watts out of small users. And commercial/industrial users are even worse, and not nearly stringently regulated enough.
The publisher probably can't be prosecuted (which is one reason they're being paid off instead, probably), but ex-intelligence officials do need permission to publish about their work. When accepting employment, they sign a contract agreeing to run any future publications about their work by the Publication Review Board for prepublication clearance.
The Supreme Court upheld that arrangement in 1980 in Snepp v. US, in a short 6-3 per curiam opinion. It's a strange opinion, because this sort of thing usually isn't permitted, but the Court was probably swayed by the "CIA stuff is important / national security / etc.".
Note that it does only apply to actual CIA employees. A non-CIA-employee author who's acquired information through leaks and interviews doesn't need the CIA's permission to publish his book.
True, incandescents in recessed ceiling sockets disperse their heat pretty inefficiently, and they're also the least interesting to me (for diffuse full-room lighting like that I have no problem with CFLs, or even tube fluorescents). I mostly use incandescents in floor and desk lamps.
I suppose to me, pro-environment means using less energy, however you do that. My preferred way is to cut out some sources of energy usage entirely. Other people might want to use lots of things but try to squeeze an extra 10% efficiency out of each. Why not just measure my usage at the curb and decide how each approach is doing based purely on results? If someone tells me I'm using X more energy than is allowable for a household, then let me decide how to reduce my usage by X.
My suspicion is that we don't take this approach because, at least around here (CA), there are plenty of people in big houses with 100% politically-correct appliances who use huge amounts of energy, and don't want to change that.
FWIW, I think thermostat threshholds aren't actually entirely ridiculous. For residential usage I imagine there'd be a backlash, but there's no reason so many square feet of office space have to be a/c'd to 72 or even 70 constantly.
For home energy usage, though, why micromanage what I do within my energy budget, so long as my total energy usage is quite low? I personally hate CFLs for reading, and I don't think my three total incandescent bulbs (225W total when all on) are really killing the environment. That's why I think just going by total usage is more fair. If my neighbor wants to run a ridiculous thermostat and television, and I don't, why can't I use my energy savings on something I prefer? My whole apartment probably uses less than 50% of the average energy around here, so I'd pass any actually objective threshholds anyone chose to impose.
But with this per-item efficiency thing, I can't run 225W of incandescent bulbs, but my neighbor can run 2000W of home-theater equipment? How is that fair or pro-environment?
Slashdot Mirror
That's sort of what "overrated" is for I guess, though it only changes the comment's score, not the adjective.
The poor review assignment at large conferences contributes to that effect as well, I think. I almost always have at least one of three reviewers, and sometimes even two of three, give a noncommittal review along the lines of, "well this isn't really my area, but it seems pretty good". Those reviews basically are non-reviews, so the acceptance decision is then entirely up to the remaining one or two reviewers. So it often comes down to: did the one person who actually provided an opinion on your paper like it or not like it?
In my experience that's often pretty subjective, especially for conferences with tight length limits (standard in AI is six pages). If the reviewer personally found the paper to be on an interesting subject with an interesting approach that he/she felt should be investigated, almost any shortcomings can be excused, and the reviewer will conclude that "Overall, this paper provides a valuable contribution to an important ongoing discussion in this area." But if the reviewer doesn't like it, finds it boring, dislikes the approach, etc., it's easy to find something that had insufficient detail, didn't sufficiently distinguish from related work, didn't sufficiently motivate the problem or investigate/validate the applications, etc., etc., since you really can't fit that much in six pages.
That's the point of "has no effect"--- the disagree mod wouldn't actually reduce a comment's score, it would just waste a mod point of the person who tried to use it, thereby removing mod points from people who shouldn't have them.
I've long thought there should be a "-1, Disagree" option in the drop-down box that takes a mod point but has no effect.
As a computer scientist, my impression is that the program committees really are pretty random, or at least based on some sort of preference other than a widely agreed "quality" standard. Try it sometime: resubmit a paper rejected from a top CS conference verbatim to another top CS conference. The correlation between the reviews is usually quite low, both in terms of the numerical scores, and especially in terms of what they liked / complained about.
If you want to go down that route, here's a syllabus from a course that's been taught a few times using Game Maker (also to mainly non-CS students), which might be useful to get ideas.
You don't think a game designer should know something about how interactive systems, procedures, dialog trees, preconditions, etc. work? I mean, you can't design a good interactive experience without at least having a vague idea about interaction and computation, even if it's mainly at a pseudocode level.
Besides, what Plan 9 user needs journaling anyway?
Ah yes, the wonders of logic: making vacuously true statements about the empty set. ;-)
Isn't not just small to medium sized businesses; most tech companies, even really huge ones, don't buy this kind of enterprise equipment. You won't find any of it at Google or Amazon, for example, even though they are quite large.
Are they really that superior to the Sun storage products (the ones Sun invented ZFS for) to be worth the big multiple in price? I mean, Sun isn't stuff-you-put-together-at-Frys prices either, but it's a lot cheaper than EMC or NetApp.
That's still a particular type of compression, isn't it? I mean, I can buy giving it a new name, since it has a bunch of infrastructure around it, but it's a perfectly general kind of data-compression algorithm as well, even if not the world's most efficient: break the data into fixed-size blocks, then, for any blocks that appear more than once, replace all occurrences after the first with a pointer to the first. Block-based RLE compression is basically a simpler version of that (where you only deduplicate the blocks when they appear in sequence).
Isn't Firefox's new JS engine, JaegerMonkey, only available for x86 and x86-64?
Oops, somehow I missed that it's been out on Windows for a while. Well, still no Linux version, anyway. =]
I dunno, at places I've been the low-level sysadmin access is not very closely monitored. "Official" access through the normal APIs is logged and monitored, but when the Unix sysadmin has root on the database machine, he could be grepping through the database for all anybody knows.
Cross-platformness in a radical sense (all hardware, all operating systems) does seem to be quickly falling by the wayside, and not just with IE only running on Windows..
The Apple version of Webkit (Safari) of course only runs on OSX, or OSX+iOS if you count Mobile Safari as the same browser. Chrome runs on the three major OSs, but only x86, x86-64, and ARM architectures, and is hard to port, due to generating machine code in its Javascript engine. Opera runs on x86, x86-64, ARM, and SuperH, and is reportedly somewhat easier to port, but it's closed-source so who knows. Firefox 4 will run only on x86 and x86-64.
So Firefox 3.6.x may be the last modern web browser that runs basically everywhere. You can get binaries for all major platforms, and Debian currently ships it for all 8 of its supported architectures: x86, x86-64, alpha, ARM, MIPS, PowerPC, S390 (!), and SPARC.
Sort of step backwards from the original Unix solution to portability: you write your stuff in C+POSIX, and then it runs everywhere we've ported a C compiler and a POSIX layer. Now apps are sprouting their own architecture-specific virtual machines! Perhaps LLVM will save us? It'd be nice if we managed to agree again on a single point of porting, so instead of saying "Chrome runs on x86, x86-64, and ARM, Firefox runs on x86 and x86-64", you can say "Browser Foo runs on anything with an LLVM port".
They're symmetric, though, which might not matter for many people, but I find nice. The 30 Mbps lowest tier is 30 Mbps each way, whereas Comcast's 30 Mbps service is 30 down, 7 up.
This rule appears to apply only to scientists who directly work for a government agency as employees, though, not to professors who are funded by federal grant money, or even professors who teach at public universities.
I agree that this is a weird/annoying feature, but it's existed in other games too, and doesn't seem to offend people (including me) as much there. For example, Magic: The Gathering has exactly that property where you can spend money to buy better cards. Is it that they've balanced it better, so the proportional effect of card-buying is less (you can't purely buy yourself victory)? I do think they've moved in that direction with the current set of rules, which is probably a good thing, but back in the days I played (mid-90s, mostly Revised/3rd Ed), there was a quite big effect to be had from buying a handful of powerful/rare/expensive cards, and people didn't seem to think that ruined the game.
I have some CFLs and LEDs, but I really hate them. The light is ridiculously non-broad-spectrum, though the CFLs in particular are better than they used to be. They're now good enough for basic house lighting imo, e.g. I have them in my ceiling sockets (incandescents only for reading lamps).
I guess I don't see it as a particularly large usage, especially by the standards here in the US. Maybe my choice of lightbulbs is eccentric and wasteful, but so is my neighbor's choice of a big-screen TV, which uses more power than my lightbulbs. Oh sure, the neighbor will probably claim that a smaller TV just isn't the same, but given that his TV is too big for his living room anyway, I find that claim ludicrous. Now maybe people find my claim that incandescents are much more pleasant than CFLs also ludicrous, so I suppose we each have our oddities. Until his big-screen TV is banned, I don't feel bad with my light bulbs. ;-)
In any case, I think we should go after total usage. Your 5 kWh/day is an impressive target, but the U.S. is way off of that: the average residential energy usage in the U.S. is 35 kWh/day. We could even be generous and start going after only the really absurd users. For example, slap a luxury tax on houses using more than 100 kWh/day (they actually exist!).
Indeed, that's the main viewpoint I think needs to be changed. Or, the amount people have to pay for using more than their fair share of energy should be much higher.
For example, I would support a %-of-income penalty for very high users, to make sure even rich people were deterred. If you're in, say, the 99.9th percentile of energy users for your area, you pay 5% of your income as a penalty.
The top 1% of residential energy users are using a very disproportionate share of the total usage, so I think those are the people we should go after, not trying to squeeze a few extra watts out of small users. And commercial/industrial users are even worse, and not nearly stringently regulated enough.
No it doesn't, because I use it for an average of an hour a day. That is 0.225 kWh/d.
Do you really think I use my reading lamps 24 hours a day?
The publisher probably can't be prosecuted (which is one reason they're being paid off instead, probably), but ex-intelligence officials do need permission to publish about their work. When accepting employment, they sign a contract agreeing to run any future publications about their work by the Publication Review Board for prepublication clearance.
The Supreme Court upheld that arrangement in 1980 in Snepp v. US , in a short 6-3 per curiam opinion. It's a strange opinion, because this sort of thing usually isn't permitted, but the Court was probably swayed by the "CIA stuff is important / national security / etc.".
Note that it does only apply to actual CIA employees. A non-CIA-employee author who's acquired information through leaks and interviews doesn't need the CIA's permission to publish his book.
True, incandescents in recessed ceiling sockets disperse their heat pretty inefficiently, and they're also the least interesting to me (for diffuse full-room lighting like that I have no problem with CFLs, or even tube fluorescents). I mostly use incandescents in floor and desk lamps.
I suppose to me, pro-environment means using less energy, however you do that. My preferred way is to cut out some sources of energy usage entirely. Other people might want to use lots of things but try to squeeze an extra 10% efficiency out of each. Why not just measure my usage at the curb and decide how each approach is doing based purely on results? If someone tells me I'm using X more energy than is allowable for a household, then let me decide how to reduce my usage by X.
My suspicion is that we don't take this approach because, at least around here (CA), there are plenty of people in big houses with 100% politically-correct appliances who use huge amounts of energy, and don't want to change that.
FWIW, I think thermostat threshholds aren't actually entirely ridiculous. For residential usage I imagine there'd be a backlash, but there's no reason so many square feet of office space have to be a/c'd to 72 or even 70 constantly.
For home energy usage, though, why micromanage what I do within my energy budget, so long as my total energy usage is quite low? I personally hate CFLs for reading, and I don't think my three total incandescent bulbs (225W total when all on) are really killing the environment. That's why I think just going by total usage is more fair. If my neighbor wants to run a ridiculous thermostat and television, and I don't, why can't I use my energy savings on something I prefer? My whole apartment probably uses less than 50% of the average energy around here, so I'd pass any actually objective threshholds anyone chose to impose.
But with this per-item efficiency thing, I can't run 225W of incandescent bulbs, but my neighbor can run 2000W of home-theater equipment? How is that fair or pro-environment?