If you only allow publication of effects with p less than 0.005, that means that in order to prevent the publication of one false positive, you are discarding ~190 results that had a true difference in outcome. I'd agree that it is better to publish nothing than to publish a wrong result, but this level of certainty seems to me excessive. Maybe 0.05 is a little too high, but surely at 0.02 or 0.01 (a one-in-a-hundred chance that you are wrong), it is time to move on to the next experiment, not keep doing the same work over and over, trying to reach the magic 0.005. Saying this is "doable and easy" is ludicrous. In biological sciences, a twofold difference could be extremely important, but to get p = 0.005 significance for a twofold difference is highly unusual. You'd have to double or triple the number of replicate experiments, with each replicate often taking weeks and many thousands of dollars to perform. Genome analysis is such a special case, you are just comparing sequences, not measuring quantitative variables in finicky cells in a wet lab. This will slow science to a crawl, probably for the sake of a marginal improvement in reproducibility.
If I go into Font, select Arial, click "Set as Default", select "All documents based on the Normal template", then "OK", the next new document has Arial by default. But if I close the program and restart, it's back to Calibri. Can't you change it permanently?
OK, so they save some money by using thinner boreholes, but how? The width of the boreholes is determined by the need to insert a loop of 2 pipes with big enough diameter to handle the coolant. If you decrease the pipe size, the resistance increases dramatically and pretty soon you're so much energy to pump it you're not saving anything. Maybe they are running the refrigerant directly into the loops, instead of water/methanol as is typical, but that's just a guess. In any case, you're not going to save $35,000 on the wells. We put in geothermal 2 years ago, I researched it pretty thoroughly and I've never heard of a system costing $60,000, so that's just a wild exaggeration. The estimates for our house ranged $31,000-36,000 and it's pretty rare for a system to top $45,000. I've never heard of anyone with 1000-ft deep wells, either. We have two 360-ft wells (although the house is small ~1500 sf). In the end, I acted as my own contractor. I paid the driller $14,000 to put in the loops, bought a heat pump on ebay, and paid a plumber to link it to the existing cast iron radiators, so no messing with the ductwork. Total cost was ~$21,000, or ~$15,000 after the (now expired) tax credit. But my point is, the loops aren't the only reason these systems are expensive. The fancy heat pumps they typically use are pretty pricey, especially after a nice markup by the HVAC contractor. If alterations to the ductwork are necessary, that's a lot of expensive labor. If Dandelion can do it cheaper, great, but I remain skeptical of how much they can save just by making the wells thinner.
As free online news sites, and Yahoo in particular, have degenerated to the point of uselessness, I now subscribe to digital New York Times and Washington Post, and don't mind paying for them. Somebody has to pay all those reporters and editors, online advertising clearly isn't going to be sufficient. This, I believe (and hope), is the future of news: a few prominent players, each with tens of millions of subscribers paying a fee small enough that it just isn't worth it to cheat with workarounds. Free sites will be increasingly taken over by purveyors of propaganda with an agenda to push and an axe to grind, whether it's Rupert or the Kremlin.
I used to do molecular modeling on an SGI machine. What was nice was that you could set up a remote desktop GUI on any Linux computer and work from anywhere.
There was competition at the time between FORTRAN and ALGOL. Physics majors learned ALGOL, which was supposed to be more humane and logical, but the engineers learned FORTRAN, with its brutal efficiency in packing the most computing into the smallest possible space - a big consideration when each line of code was hand-typed on an individual punch card. I was particularly fond of the arithmetic IF: "IF (x-y/z) 10, 15, 20" would take the program to line 10, 15 or 20 depending on whether x-y/z (or any arithmetic expression) was negative, zero or positive.
What's the point of all these "loyalty" programs? Mostly to evade taxes. Highly paid management personnel might go on dozens of business trips a year, accumulating multiple free flights that they then typically use for personal travel. But of course they pay no income tax on the miles earned, even though by any reasonable definition it is compensation. The rest of us ordinary folks are just tagging along for the ride, and the airlines are increasingly finding ways to lock us out and restrict rewards to the high-rollers.
Well, maybe once upon a time, when I was a child, and it really was The Big Screen, 30 or 40 feet wide, and a few hundred people were all there to see the same movie, it really was an event. Now there's 15 screens, each not that much bigger than one you might have at home, and you're just one in a herd of consumers being serviced. Multiplexes destroyed the moviegoing experience decades ago.
1. Allow comments only from subscribers. People love to opine, and will pay for the privilege.
2. At the same time, allow comments on a larger range of articles. Particularly aggravating is that they don't normally allow comments to guest editorials. Why? If they can't take criticism, they shouldn't be writing editorials.
3. Allow subscribers only to turn off animated ads. NYT blocks at least some ad blockers, but without an ad blocker it is all but unreadable due to distraction from animated ads all over the page.
Despite news of falling prices, the cost of the prismatic LiFePO4 cells commonly used in roll-your-own EV conversions has not come down AT ALL. They're still $1.30-1.40/Ah, or about $410-440/kWh, the same as they were in 2010. So a pack for a usable vehicle is still at least $10K. Sad to say, it's now much cheaper to buy a used EV than to build one.
The key to the success and of fake news and the main determinant of its content is not its sources but its consumers. What social media companies have discovered is that giving people whatever news they personally want to hear, regardless of its accuracy, can be a highly lucrative business. Just set up the algorithms, watch the news sources arise like magic, see the subscribers rack up clicks, and let the ad revenue roll in.
So let me get this straight. I can pay $35 and watch, say, 3 hr of streamed wireless video every day, and AT&T will give me the necessary ~100GB of bandwidth free. But if I just buy an extra unrestricted 10GB of wireless data from Verizon, they charge $45. Makes you really wonder what it actually costs the telco's to provide each GB. Seems like either Verizon's data charges must be ~90% profit at least, or else AT&T is so desperate to stay relevant in the content space that they are willing to endure massive losses by providing bandwidth way below their cost. Maybe a little of both?
At least in the sense, Facebook is sucking all the oxygen, i.e. online ad revenue, out of the room, making billions on news without hiring a single reporter or opening a single news bureau. Even the largest legitimate news organizations, NYTimes, Wall Street Journal, are struggling to make a business model in the online world, because people don't need to subscribe to them or look at their ads in order to get news. Instead they get it from Facebook, which selectively feeds readers only the news they want to hear.
What's the beef? With modern compression algorithms, DSL is plenty fast enough for video streaming, who needs more than that? If I wanted, I could potentially stream 250 GB a month, imagine what that would cost on a wireless contract. DSL reliability is incredible; in 10 years it's been out maybe 3 or 4 times for a few hours. When the power has gone out for days at a time, DSL still works, as does my landline phone. There are no rental charges; my modem was free and it still works. Best of all, I don't have to deal with Comcast and their incomprehensible rate structure. Even including all the phoney landline charges, it's less than half what I'd pay otherwise. DSL now, DSL tomorrow, DSL forever.
If they charge $50/month for 10 years, that's $6000. Can it possibly cost anywhere near that to run a wire from one house to the street? Maybe just let the homeowners do it themselves!?
Took a look at their home page https://webpass.net/residentia... Compared to Verizon or Comcast, it's heaven on earth. A flat $550 a year, no asterisks, no teaser rates, no setup charges, no equipment rentals, no bundled content nobody wants, and free installation. I can't even tell what I'd have to pay Verizon to get the same service but I know it's at least twice that.
Surely GOP politicians, of all people, will recognize that unlike IRS, Facebook is a PRIVATE CORPORATION which can rate trending topics by whatever criteria it chooses, including Political Correctness. If wingnuts don’t “like” it, they can use some other network, or better yet start their own.
We've all read the stories of hackers locking up the data of hospitals and demanding ransom payments in Bitcoin, because it's untraceable. If flagrant criminals can clear payments in Bitcoin, why can't US-designated "rogue" regimes use Bitcoin to get around being shut out of international banking? For example, why couldn't Argentina pay their bondholders in Bitcoin, when international bankers refused to process their payments?
Slashdot Mirror
If you only allow publication of effects with p less than 0.005, that means that in order to prevent the publication of one false positive, you are discarding ~190 results that had a true difference in outcome. I'd agree that it is better to publish nothing than to publish a wrong result, but this level of certainty seems to me excessive. Maybe 0.05 is a little too high, but surely at 0.02 or 0.01 (a one-in-a-hundred chance that you are wrong), it is time to move on to the next experiment, not keep doing the same work over and over, trying to reach the magic 0.005. Saying this is "doable and easy" is ludicrous. In biological sciences, a twofold difference could be extremely important, but to get p = 0.005 significance for a twofold difference is highly unusual. You'd have to double or triple the number of replicate experiments, with each replicate often taking weeks and many thousands of dollars to perform. Genome analysis is such a special case, you are just comparing sequences, not measuring quantitative variables in finicky cells in a wet lab. This will slow science to a crawl, probably for the sake of a marginal improvement in reproducibility.
If I go into Font, select Arial, click "Set as Default", select "All documents based on the Normal template", then "OK", the next new document has Arial by default. But if I close the program and restart, it's back to Calibri. Can't you change it permanently?
New York Times invites you, right on their front page, to "Join us on Facebook"
OK, so they save some money by using thinner boreholes, but how? The width of the boreholes is determined by the need to insert a loop of 2 pipes with big enough diameter to handle the coolant. If you decrease the pipe size, the resistance increases dramatically and pretty soon you're so much energy to pump it you're not saving anything. Maybe they are running the refrigerant directly into the loops, instead of water/methanol as is typical, but that's just a guess. In any case, you're not going to save $35,000 on the wells. We put in geothermal 2 years ago, I researched it pretty thoroughly and I've never heard of a system costing $60,000, so that's just a wild exaggeration. The estimates for our house ranged $31,000-36,000 and it's pretty rare for a system to top $45,000. I've never heard of anyone with 1000-ft deep wells, either. We have two 360-ft wells (although the house is small ~1500 sf). In the end, I acted as my own contractor. I paid the driller $14,000 to put in the loops, bought a heat pump on ebay, and paid a plumber to link it to the existing cast iron radiators, so no messing with the ductwork. Total cost was ~$21,000, or ~$15,000 after the (now expired) tax credit. But my point is, the loops aren't the only reason these systems are expensive. The fancy heat pumps they typically use are pretty pricey, especially after a nice markup by the HVAC contractor. If alterations to the ductwork are necessary, that's a lot of expensive labor. If Dandelion can do it cheaper, great, but I remain skeptical of how much they can save just by making the wells thinner.
As free online news sites, and Yahoo in particular, have degenerated to the point of uselessness, I now subscribe to digital New York Times and Washington Post, and don't mind paying for them. Somebody has to pay all those reporters and editors, online advertising clearly isn't going to be sufficient. This, I believe (and hope), is the future of news: a few prominent players, each with tens of millions of subscribers paying a fee small enough that it just isn't worth it to cheat with workarounds. Free sites will be increasingly taken over by purveyors of propaganda with an agenda to push and an axe to grind, whether it's Rupert or the Kremlin.
I used to do molecular modeling on an SGI machine. What was nice was that you could set up a remote desktop GUI on any Linux computer and work from anywhere.
Just to show how serious this is, they should kick him/us out of the G7.
No, but had it been, I would have been prepared, having taken 4 years of Latin at my Catholic high school.
There was competition at the time between FORTRAN and ALGOL. Physics majors learned ALGOL, which was supposed to be more humane and logical, but the engineers learned FORTRAN, with its brutal efficiency in packing the most computing into the smallest possible space - a big consideration when each line of code was hand-typed on an individual punch card. I was particularly fond of the arithmetic IF: "IF (x-y/z) 10, 15, 20" would take the program to line 10, 15 or 20 depending on whether x-y/z (or any arithmetic expression) was negative, zero or positive.
The 1972 Russian version, is the only thing that comes immediately to mind.
What's the point of all these "loyalty" programs? Mostly to evade taxes. Highly paid management personnel might go on dozens of business trips a year, accumulating multiple free flights that they then typically use for personal travel. But of course they pay no income tax on the miles earned, even though by any reasonable definition it is compensation. The rest of us ordinary folks are just tagging along for the ride, and the airlines are increasingly finding ways to lock us out and restrict rewards to the high-rollers.
Well, maybe once upon a time, when I was a child, and it really was The Big Screen, 30 or 40 feet wide, and a few hundred people were all there to see the same movie, it really was an event. Now there's 15 screens, each not that much bigger than one you might have at home, and you're just one in a herd of consumers being serviced. Multiplexes destroyed the moviegoing experience decades ago.
1. Allow comments only from subscribers. People love to opine, and will pay for the privilege. 2. At the same time, allow comments on a larger range of articles. Particularly aggravating is that they don't normally allow comments to guest editorials. Why? If they can't take criticism, they shouldn't be writing editorials. 3. Allow subscribers only to turn off animated ads. NYT blocks at least some ad blockers, but without an ad blocker it is all but unreadable due to distraction from animated ads all over the page.
Despite news of falling prices, the cost of the prismatic LiFePO4 cells commonly used in roll-your-own EV conversions has not come down AT ALL. They're still $1.30-1.40/Ah, or about $410-440/kWh, the same as they were in 2010. So a pack for a usable vehicle is still at least $10K. Sad to say, it's now much cheaper to buy a used EV than to build one.
Instead of a lottery, allocate the limited H-1B slots by auction, then use the money to support tech scholarships for US citizens.
The key to the success and of fake news and the main determinant of its content is not its sources but its consumers. What social media companies have discovered is that giving people whatever news they personally want to hear, regardless of its accuracy, can be a highly lucrative business. Just set up the algorithms, watch the news sources arise like magic, see the subscribers rack up clicks, and let the ad revenue roll in.
So let me get this straight. I can pay $35 and watch, say, 3 hr of streamed wireless video every day, and AT&T will give me the necessary ~100GB of bandwidth free. But if I just buy an extra unrestricted 10GB of wireless data from Verizon, they charge $45. Makes you really wonder what it actually costs the telco's to provide each GB. Seems like either Verizon's data charges must be ~90% profit at least, or else AT&T is so desperate to stay relevant in the content space that they are willing to endure massive losses by providing bandwidth way below their cost. Maybe a little of both?
At least in the sense, Facebook is sucking all the oxygen, i.e. online ad revenue, out of the room, making billions on news without hiring a single reporter or opening a single news bureau. Even the largest legitimate news organizations, NYTimes, Wall Street Journal, are struggling to make a business model in the online world, because people don't need to subscribe to them or look at their ads in order to get news. Instead they get it from Facebook, which selectively feeds readers only the news they want to hear.
Give it the voice of the computer from (the original) Star Trek and I'll buy it in a minute.
What's the beef? With modern compression algorithms, DSL is plenty fast enough for video streaming, who needs more than that? If I wanted, I could potentially stream 250 GB a month, imagine what that would cost on a wireless contract. DSL reliability is incredible; in 10 years it's been out maybe 3 or 4 times for a few hours. When the power has gone out for days at a time, DSL still works, as does my landline phone. There are no rental charges; my modem was free and it still works. Best of all, I don't have to deal with Comcast and their incomprehensible rate structure. Even including all the phoney landline charges, it's less than half what I'd pay otherwise. DSL now, DSL tomorrow, DSL forever.
If they charge $50/month for 10 years, that's $6000. Can it possibly cost anywhere near that to run a wire from one house to the street? Maybe just let the homeowners do it themselves!?
...but can it really be that difficult to have a browser set up so that it pretends to play an ad and simply superimposes white space on top of it??
Took a look at their home page https://webpass.net/residentia... Compared to Verizon or Comcast, it's heaven on earth. A flat $550 a year, no asterisks, no teaser rates, no setup charges, no equipment rentals, no bundled content nobody wants, and free installation. I can't even tell what I'd have to pay Verizon to get the same service but I know it's at least twice that.
Surely GOP politicians, of all people, will recognize that unlike IRS, Facebook is a PRIVATE CORPORATION which can rate trending topics by whatever criteria it chooses, including Political Correctness. If wingnuts don’t “like” it, they can use some other network, or better yet start their own.
We've all read the stories of hackers locking up the data of hospitals and demanding ransom payments in Bitcoin, because it's untraceable. If flagrant criminals can clear payments in Bitcoin, why can't US-designated "rogue" regimes use Bitcoin to get around being shut out of international banking? For example, why couldn't Argentina pay their bondholders in Bitcoin, when international bankers refused to process their payments?