Worse, the people who need to be excited about WebM (big corporate media) will actively be repulsed by the FSF's stamp of approval. The way these people think, open = free = piracy, and more open = more free = more piracy. They hear "FSF" and envision a large, bearded hippie with his middle finger raised in their direction.
Even ignoring that problem, unless Google is willing to stand behind it with indemnity from patent suits, those media giants going to see WebM as a giant target painted on their chests. Video compression is a patent minefield, and indemnity is pretty much an absolute requirement these days if you expect to be taken seriously. So now those media giants will see a large bearded hippie flipping them off, with a bomb strapped to his chest.
The FSF putting their stamp on it is just the final nail in WebM's coffin. Stick a fork in it. It's done. Google has really screwed the pooch on this one.
I never said it was my backup. It's just one of the ways I share (loose, release) photos... and not the one I prefer, but the one many of my friends prefer. I would much prefer it if Facebook would let me tag photos on my own web server so that people could view them at full quality (or at least have the photo link to my site). It's fine that FB doesn't want to waste the bandwidth, but I'd like to have the option to substitute my own.:-)
It's more about economics than anything else. Over time, cameras take bigger pictures and flash costs decrease. For an average camera user, even if you're shooting RAW, it will take you at least three or four years to go through a 32GB flash card unless you shoot a *lot* of photos. Most people replace their cameras with a larger camera every four or five years, and that camera will need a bigger flash card to last as long, so you're probably going to buy a new flash card anyway. Also, half the time, they've changed flash card formats and your old one won't work with the new camera anyway. So it's really not a very big leap, assuming that you buy the biggest flash card you can buy.
Put another way, I'm starting to think about swapping to my second 32GB flash card after about three or four years, the last year of which I've been shooting RAW almost exclusively. I'm waiting for Canon DSLRs to offer geotagging before I upgrade, so I figure I'm probably 1-2 years out, but I doubt I'll use up the second 32 GB card by then. Yet most of the folks I know are shocked at how many pictures I take by comparison with what they do, so I'm guessing the average is probably an order of magnitude fewer photos per year than I take. Given that, unless you shoot video with your still camera, an average person (who typically does *not* shoot in RAW) could probably use a 32GB card for the rest of his/her life and never fill it up.
As for the boxes full of flash cards, if you're comparing to hard drives, I think you need to check the numbers. The density of flash cards is actually amazing when you get right down to it. A CF card is about 5.14 mL, and can hold up to 128GB (24.9 GB/mL). A 2.5" hard drive is 87.3125 mL and holds up to 1 TB last I checked (12.3 GB/mL). A desktop hard drive is 389.8 mL and holds up to 3 TB (7.7 GB/mL). Thus, flash cards are substantially more data dense than any hard drive. Man, I never thought I would use GB/mL as a unit of measurement.... Next stop, furlongs per fortnight.
Does unplugging the drive cables from the circuit board count as disassembly? Because you don't really need to have physical access to the inside of the drive, but merely access to the raw analog output of the drive heads on a given track (and, ideally, fractional tracks, which you should be able to fudge by rapidly stepping the drive heads one way and then the other).
True, it can be done easily, but the expert didn't say it couldn't be done. The expert said it wouldn't be done. If you store the key in RAM and there's a power failure, your bug will never work again until someone physically goes in and rebuilds the system from scratch. Determining the difference between an actual detection incident and a harmless condition like a blackout is a nontrivial exercise. And if you have regular access to the area, there's probably no good reason to plant a bug there. You generally would plant a bug in places where you don't regularly go so that you can have access to them when you otherwise wouldn't. Therefore, one can assume that when you amortize the damage over a long enough period of time, a self-destructing bug is a poor value proposition.
In short, the computer expert argued that a self-destructing app was foolish---not because it can't be done, nor even because it is difficult, but because it is self-defeating.
Ack! I've been dealing with product marketing people too much! I'm starting to sound like them! Gaaaaaah!
1- Don't erase any images from the memory card except the useless ones (like those with the lens cap on). Get a new card when full. This is much cheaper than film and developing was just a few years ago.
2- When card is full, or when you get back from a trip like that, copy all the images to an external USB hard disk.
3- Every once in a while (once per year at least), do a system backup to the external USB hard drive, encrypt anything that might be embarrassing, and send the drive to your Mom for off-site storage.
My scheme is similar:
1. Don't erase any images from the flash card. Get a new card when full. Flash is cheap. Photos are priceless.
2. Periodically copy the photos to your laptop HD. You now have one backup.
3. Back your laptop up to a fireproof hard drive.
4. Take the flash cards to work for off-site storage if you're really paranoid.
You're assuming that it's equally easy for all companies to cut down on carbon emission, and that the total carbon credit can't be altered.
Bear in mind that most of the cost is not the equipment, but rather the difference in ongoing cost of operations. The physical equipment costs are a drop in the bucket compared with the long-term cost of using natural gas or electricity instead of coal for pretty much anything. The main driver is the cost of buying the energy from someone, and the cleaner forms of energy are always going to be significantly more expensive by about the same amount, whether you're buying that energy to turn a crank that spits out widgets or buying that energy to heat a vat of metal for smelting. That cost difference is proportional to total energy use and is likely to be fairly comparable across all industries, ignoring small market price differences in the cost of natural gas from region to region.
About the only situation where the cost should be significantly different, ignoring the initial hardware costs, would be in the case where you're talking about smelting, refining, or other operations that require insane amounts of heat. Those are at a disadvantage because producing heat by electricity tends to not be particularly efficient, leaving them pretty much only fossil fuels (including natural gas) as options. In regions where electricity is cheaper than natural gas, they would be at a *slight* cost disadvantage. However, and here's the kicker, they are all at an approximately equal disadvantage, so once the low hanging fruit in other sectors has transitioned to other energy sources, you're in the same boat.
Oh, and the power companies themselves are something of an exception, too, but ignoring early industrial nations, power companies are already working pretty hard to get away from CO2-emitting fossil fuels for cost reasons, making this something of a moot point.
More to the point, it can be solved just as easily if Facebook would:
Require users to accept or reject tagging explicitly before a photo tag becomes visible to anyone other than the tagger and the taggee.
Expire photos after a reasonable period of time unless the user explicitly confirms that it should remain posted (use notifications).
Expire tags in the same fashion.
More importantly, it fails because:
The sorts of people who post pictures of their friends looking like assholes are unlikely to care enough to use a special service that provides expiration.
The sorts of people who post pictures of their friends looking like assholes are unlikely to set a short expiration date.
The person affected by the tagging is not the person deciding on its expiration.
The decision about how long I should be tagged in a photo must be my decision, not the decision of the person who posts the photo. Any scheme that does not achieve this goal is completely missing the point.
5. Due to 4, there are economic incentives to reduce carbon output.
And this is how I know that the folks who came up with this scheme failed high school economics. The mere fact that you can make money selling credits does not inherently provide any incentive to reduce pollution. It only provides incentive to do so if the amount you make by selling those credits exceeds the amount it would cost you to update your equipment to produce less emissions.
Here's the catch: if it would cost you less to update your equipment to produce less emissions, it would also probably cost your buyer less to upgrade their equipment than it would cost to buy the credit, so they will not buy the credit. As soon as you understand that fundamental flaw in this scheme, the whole thing falls apart. The inherent value of the carbon credit must, then, be cheaper than the cost of the upgrades in order for anyone to buy them. So with the exception of a few edge cases (where upgrading is unusually cheap or expensive), these don't provide any incentive at all, and most of those would either already have converted to cleaner energy or would be specifically exempted from pure caps anyway, making those edge cases largely uninteresting.
Thus, the only possible long-term effect of cap and trade is that companies who would have upgraded anyway will do so, but will now be able to sell their credits, allowing other companies to produce more pollution than they would have been able to produce under a strict system of caps.
Worse, this is the case in the short term as well. If initially there are companies that produce substantially less than the cap as you propose (as opposed to each company being capped at its current output), then there will be a glut of these credits, making them much cheaper than the cost of upgrading equipment, thus providing not just incentive not to upgrade, but also incentive to horde credits so that you can continue to pollute more and more at will, in effect eliminating environmental regulation altogether.
In short, cap and trade can only lead to increased pollution in reasonable human timeframes, not decreased. Sure, it does limit the growth to a certain total, which might result in lower total pollution in the really, really long term, but even that requires us to assume that the market will never result in newer, lower-polluting manufacturing technology being cheaper than ancient clunker hardware. Color me unconvinced.
The only way cap and trade is useful is if the current total is substantially less than the total output now, and to the extent that this is the case, the same can be achieved with simple caps, goals, and fines.
Microsoft got tired of people asking when they were going to fully support HTML 4....
Now everyone will be able to say "We support HTML" even though nobody fully supports all aspects of the spec. Just like today, only nobody will be able to point their finger at any sort of milestone that they missed, so companies that drag their heels in standards compliance end up looking better.
How is this a benefit again? It seems to me that we need smaller, more frequent milestones, not elimination of those milestones.
I should add: mathematicians have been saying that NP is much, much harder than P, and it has always seemed logical to say that.
I wouldn't go that far. I would actually have said the opposite---that if a problem can be verified to be correct in polynomial time, it stands to reason that there should be a way to solve it in poly time. After all, there are hundreds of heuristic approaches that solve them for certain subsets of the problem, certain special cases, etc. All that is missing is a generalizable approach. It might be true that no way to generalize them exists, but if it is, it's the sort of right that requires extraordinary proof in my book.
I've been saying for years that I thought it would be downright hilarious if somebody proved P = NP and really messed with the heads of theoretical computer scientists, so I'm kind of hoping this pans out in a schadenfreude sort of way. That said, statistically, given how many attempts people have made to prove this one way or the other, it probably won't.
No, not really. The problem is that there are some areas of commerce that naturally trend towards monopolies, and no amount of deregulation can ever prevent them from degrading into a monopoly. Therefore, the only options are either harsh government regulation to prevent the monopoly from abusing its monopoly status or government ownership of that area of commerce.
I've seen places where the government allowed a second cable company to move in. Invariably, the incumbent cable company, having the advantage of owning all of their infrastructure free and clear, cuts their previously exorbitant rates dramatically to undercut or match the cheaper rates offered by the newcomer. Competition thrives and everyone is happy with their lower rates... until two years later when the new cable company is still operating hopelessly in the red (despite raising rates once or twice) and is forced to cease operation, whereupon it sells its brand new lines, antenna tower, and office space to the incumbent cable company, and exits the market. On the plus side, everyone in those towns now have newer, higher quality coax with newer amplifiers, etc., but there's still no competition in any of those towns, and their cable rates skyrocketed almost immediately after the exit of the competitors.
As for the S&L debacle, I wouldn't say regulation was "uneven", though. That implies that if you just removed a whole lot more regulations, we'd be in better shape. In reality, they merely removed the wrong regulations. Unfortunately, there really aren't a lot of regulations that are safe to remove.
In general, regulations are put in place to prevent abuses that are already happening, so removal of any of those regulations almost invariably leads to bad things except when those regulations truly are no longer relevant due to some significant change in the landscape (e.g. laws about texting while driving will become irrelevant when cars drive themselves; if there were laws requiring cellular carriers to lease access to their towers, they became irrelevant now that we are no longer limited to one A and one B analog carrier in any given area; etc.).
Hah! The Tea Party wants less government interference, not more. They would never agree to a public-owned wire service. Instead, they would eliminate the limits on the number of services with access to public rights of way. Thus, you would have eighteen different companies digging up your streets to run their own lines. Because they would not be required by law to repave the street, the streets would gradually be replaced by a patchwork of heavy steel plates covering the open trenches below. Because the open trenches would fill with water in the rainy season, you would have eighteen companies digging new trenches again the next year.
I would think that the 10-12 msec estimate already took all of that into account. If the bird is right above you, a 1244 mile round trip is a little over 6 msec. at the speed of light. Also, bear in mind that LEO satellites don't use a dish---they use a normal antenna---so no alignment is involved. You can't realistically track a bird whose twenty minute ground path is the size of North America using a directional aerial. The whole point of LEO constellations is that there are always multiple birds overhead, so you talk to the one that provides the strongest signal or whatever.
But you're right about load balancing and cities. The flip side of that coin is that these can cover areas that can't feasibly be covered by cellular coverage due to low population density. You know, like most of Canada, where this company is based. It's a tradeoff.
Won't work. Companies can't really file patents. The inventors technically file the patents and assign them to the company (by prior agreement). Therefore, get a large enough company and you can effectively file a whole lot of patents without the same person ever being the listed inventor.
No, if you want to cut down on the number of junk patents, require that the people listed on the "Inventor" line pay for the patent by personal check and make reimbursement illegal.
Slashdot Mirror
Too late.
Worse, the people who need to be excited about WebM (big corporate media) will actively be repulsed by the FSF's stamp of approval. The way these people think, open = free = piracy, and more open = more free = more piracy. They hear "FSF" and envision a large, bearded hippie with his middle finger raised in their direction.
Even ignoring that problem, unless Google is willing to stand behind it with indemnity from patent suits, those media giants going to see WebM as a giant target painted on their chests. Video compression is a patent minefield, and indemnity is pretty much an absolute requirement these days if you expect to be taken seriously. So now those media giants will see a large bearded hippie flipping them off, with a bomb strapped to his chest.
The FSF putting their stamp on it is just the final nail in WebM's coffin. Stick a fork in it. It's done. Google has really screwed the pooch on this one.
That's why you copy them to your computer and keep the flash card as a backup.
I never said it was my backup. It's just one of the ways I share (loose, release) photos... and not the one I prefer, but the one many of my friends prefer. I would much prefer it if Facebook would let me tag photos on my own web server so that people could view them at full quality (or at least have the photo link to my site). It's fine that FB doesn't want to waste the bandwidth, but I'd like to have the option to substitute my own. :-)
It's more about economics than anything else. Over time, cameras take bigger pictures and flash costs decrease. For an average camera user, even if you're shooting RAW, it will take you at least three or four years to go through a 32GB flash card unless you shoot a *lot* of photos. Most people replace their cameras with a larger camera every four or five years, and that camera will need a bigger flash card to last as long, so you're probably going to buy a new flash card anyway. Also, half the time, they've changed flash card formats and your old one won't work with the new camera anyway. So it's really not a very big leap, assuming that you buy the biggest flash card you can buy.
Put another way, I'm starting to think about swapping to my second 32GB flash card after about three or four years, the last year of which I've been shooting RAW almost exclusively. I'm waiting for Canon DSLRs to offer geotagging before I upgrade, so I figure I'm probably 1-2 years out, but I doubt I'll use up the second 32 GB card by then. Yet most of the folks I know are shocked at how many pictures I take by comparison with what they do, so I'm guessing the average is probably an order of magnitude fewer photos per year than I take. Given that, unless you shoot video with your still camera, an average person (who typically does *not* shoot in RAW) could probably use a 32GB card for the rest of his/her life and never fill it up.
As for the boxes full of flash cards, if you're comparing to hard drives, I think you need to check the numbers. The density of flash cards is actually amazing when you get right down to it. A CF card is about 5.14 mL, and can hold up to 128GB (24.9 GB/mL). A 2.5" hard drive is 87.3125 mL and holds up to 1 TB last I checked (12.3 GB/mL). A desktop hard drive is 389.8 mL and holds up to 3 TB (7.7 GB/mL). Thus, flash cards are substantially more data dense than any hard drive. Man, I never thought I would use GB/mL as a unit of measurement.... Next stop, furlongs per fortnight.
Until you dispose of the panels after they fail.
Does unplugging the drive cables from the circuit board count as disassembly? Because you don't really need to have physical access to the inside of the drive, but merely access to the raw analog output of the drive heads on a given track (and, ideally, fractional tracks, which you should be able to fudge by rapidly stepping the drive heads one way and then the other).
True, it can be done easily, but the expert didn't say it couldn't be done. The expert said it wouldn't be done. If you store the key in RAM and there's a power failure, your bug will never work again until someone physically goes in and rebuilds the system from scratch. Determining the difference between an actual detection incident and a harmless condition like a blackout is a nontrivial exercise. And if you have regular access to the area, there's probably no good reason to plant a bug there. You generally would plant a bug in places where you don't regularly go so that you can have access to them when you otherwise wouldn't. Therefore, one can assume that when you amortize the damage over a long enough period of time, a self-destructing bug is a poor value proposition.
In short, the computer expert argued that a self-destructing app was foolish---not because it can't be done, nor even because it is difficult, but because it is self-defeating.
Ack! I've been dealing with product marketing people too much! I'm starting to sound like them! Gaaaaaah!
You've never been to 4chan, I take it?
x = don't be really evil?
How odd. I loose some of my photos on Facebook every few days. Oh, you meant lose. :-D
Sorry, couldn't resist.
My scheme is similar:
1. Don't erase any images from the flash card. Get a new card when full. Flash is cheap. Photos are priceless.
2. Periodically copy the photos to your laptop HD. You now have one backup.
3. Back your laptop up to a fireproof hard drive.
4. Take the flash cards to work for off-site storage if you're really paranoid.
Bear in mind that most of the cost is not the equipment, but rather the difference in ongoing cost of operations. The physical equipment costs are a drop in the bucket compared with the long-term cost of using natural gas or electricity instead of coal for pretty much anything. The main driver is the cost of buying the energy from someone, and the cleaner forms of energy are always going to be significantly more expensive by about the same amount, whether you're buying that energy to turn a crank that spits out widgets or buying that energy to heat a vat of metal for smelting. That cost difference is proportional to total energy use and is likely to be fairly comparable across all industries, ignoring small market price differences in the cost of natural gas from region to region.
About the only situation where the cost should be significantly different, ignoring the initial hardware costs, would be in the case where you're talking about smelting, refining, or other operations that require insane amounts of heat. Those are at a disadvantage because producing heat by electricity tends to not be particularly efficient, leaving them pretty much only fossil fuels (including natural gas) as options. In regions where electricity is cheaper than natural gas, they would be at a *slight* cost disadvantage. However, and here's the kicker, they are all at an approximately equal disadvantage, so once the low hanging fruit in other sectors has transitioned to other energy sources, you're in the same boat.
Oh, and the power companies themselves are something of an exception, too, but ignoring early industrial nations, power companies are already working pretty hard to get away from CO2-emitting fossil fuels for cost reasons, making this something of a moot point.
More to the point, it can be solved just as easily if Facebook would:
More importantly, it fails because:
The decision about how long I should be tagged in a photo must be my decision, not the decision of the person who posts the photo. Any scheme that does not achieve this goal is completely missing the point.
And this is how I know that the folks who came up with this scheme failed high school economics. The mere fact that you can make money selling credits does not inherently provide any incentive to reduce pollution. It only provides incentive to do so if the amount you make by selling those credits exceeds the amount it would cost you to update your equipment to produce less emissions.
Here's the catch: if it would cost you less to update your equipment to produce less emissions, it would also probably cost your buyer less to upgrade their equipment than it would cost to buy the credit, so they will not buy the credit. As soon as you understand that fundamental flaw in this scheme, the whole thing falls apart. The inherent value of the carbon credit must, then, be cheaper than the cost of the upgrades in order for anyone to buy them. So with the exception of a few edge cases (where upgrading is unusually cheap or expensive), these don't provide any incentive at all, and most of those would either already have converted to cleaner energy or would be specifically exempted from pure caps anyway, making those edge cases largely uninteresting.
Thus, the only possible long-term effect of cap and trade is that companies who would have upgraded anyway will do so, but will now be able to sell their credits, allowing other companies to produce more pollution than they would have been able to produce under a strict system of caps.
Worse, this is the case in the short term as well. If initially there are companies that produce substantially less than the cap as you propose (as opposed to each company being capped at its current output), then there will be a glut of these credits, making them much cheaper than the cost of upgrading equipment, thus providing not just incentive not to upgrade, but also incentive to horde credits so that you can continue to pollute more and more at will, in effect eliminating environmental regulation altogether.
In short, cap and trade can only lead to increased pollution in reasonable human timeframes, not decreased. Sure, it does limit the growth to a certain total, which might result in lower total pollution in the really, really long term, but even that requires us to assume that the market will never result in newer, lower-polluting manufacturing technology being cheaper than ancient clunker hardware. Color me unconvinced.
The only way cap and trade is useful is if the current total is substantially less than the total output now, and to the extent that this is the case, the same can be achieved with simple caps, goals, and fines.
That movie was hilarious. Oh, wait, you weren't talking about "Men in Tights"? Never mind.
Microsoft got tired of people asking when they were going to fully support HTML 4....
Now everyone will be able to say "We support HTML" even though nobody fully supports all aspects of the spec. Just like today, only nobody will be able to point their finger at any sort of milestone that they missed, so companies that drag their heels in standards compliance end up looking better.
How is this a benefit again? It seems to me that we need smaller, more frequent milestones, not elimination of those milestones.
I wouldn't go that far. I would actually have said the opposite---that if a problem can be verified to be correct in polynomial time, it stands to reason that there should be a way to solve it in poly time. After all, there are hundreds of heuristic approaches that solve them for certain subsets of the problem, certain special cases, etc. All that is missing is a generalizable approach. It might be true that no way to generalize them exists, but if it is, it's the sort of right that requires extraordinary proof in my book.
I've been saying for years that I thought it would be downright hilarious if somebody proved P = NP and really messed with the heads of theoretical computer scientists, so I'm kind of hoping this pans out in a schadenfreude sort of way. That said, statistically, given how many attempts people have made to prove this one way or the other, it probably won't.
Besides, P = NP if N = 1. Everyone knows that. :-)
No, not really. The problem is that there are some areas of commerce that naturally trend towards monopolies, and no amount of deregulation can ever prevent them from degrading into a monopoly. Therefore, the only options are either harsh government regulation to prevent the monopoly from abusing its monopoly status or government ownership of that area of commerce.
I've seen places where the government allowed a second cable company to move in. Invariably, the incumbent cable company, having the advantage of owning all of their infrastructure free and clear, cuts their previously exorbitant rates dramatically to undercut or match the cheaper rates offered by the newcomer. Competition thrives and everyone is happy with their lower rates... until two years later when the new cable company is still operating hopelessly in the red (despite raising rates once or twice) and is forced to cease operation, whereupon it sells its brand new lines, antenna tower, and office space to the incumbent cable company, and exits the market. On the plus side, everyone in those towns now have newer, higher quality coax with newer amplifiers, etc., but there's still no competition in any of those towns, and their cable rates skyrocketed almost immediately after the exit of the competitors.
As for the S&L debacle, I wouldn't say regulation was "uneven", though. That implies that if you just removed a whole lot more regulations, we'd be in better shape. In reality, they merely removed the wrong regulations. Unfortunately, there really aren't a lot of regulations that are safe to remove.
In general, regulations are put in place to prevent abuses that are already happening, so removal of any of those regulations almost invariably leads to bad things except when those regulations truly are no longer relevant due to some significant change in the landscape (e.g. laws about texting while driving will become irrelevant when cars drive themselves; if there were laws requiring cellular carriers to lease access to their towers, they became irrelevant now that we are no longer limited to one A and one B analog carrier in any given area; etc.).
Qwest: Ride the shock front?
Hah! The Tea Party wants less government interference, not more. They would never agree to a public-owned wire service. Instead, they would eliminate the limits on the number of services with access to public rights of way. Thus, you would have eighteen different companies digging up your streets to run their own lines. Because they would not be required by law to repave the street, the streets would gradually be replaced by a patchwork of heavy steel plates covering the open trenches below. Because the open trenches would fill with water in the rainy season, you would have eighteen companies digging new trenches again the next year.
Enjoy your train wreck.
I would think that the 10-12 msec estimate already took all of that into account. If the bird is right above you, a 1244 mile round trip is a little over 6 msec. at the speed of light. Also, bear in mind that LEO satellites don't use a dish---they use a normal antenna---so no alignment is involved. You can't realistically track a bird whose twenty minute ground path is the size of North America using a directional aerial. The whole point of LEO constellations is that there are always multiple birds overhead, so you talk to the one that provides the strongest signal or whatever.
But you're right about load balancing and cities. The flip side of that coin is that these can cover areas that can't feasibly be covered by cellular coverage due to low population density. You know, like most of Canada, where this company is based. It's a tradeoff.
Yo, dawg. I herd you liked to program, so I put VB in your Silverlight so you can program while you program.
Wait, what?
Jeez. I just watched that movie last night. What are you, a Peeping Tom?
Won't work. Companies can't really file patents. The inventors technically file the patents and assign them to the company (by prior agreement). Therefore, get a large enough company and you can effectively file a whole lot of patents without the same person ever being the listed inventor.
No, if you want to cut down on the number of junk patents, require that the people listed on the "Inventor" line pay for the patent by personal check and make reimbursement illegal.