Club cards are only useful if you can't chose to shop elsewhere.
Which often is the case.
Which is why I said it...
Actually, I don't think it's OFTEN the case. Depending on where you are and what stores are around, driving 50 miles out of your way might actually save you a lot of money, and allow you to maintain some privacy as well. I'm not in a big city, but it's still only a 15 mile drive to get to a much better supermarket than the 2 around here.
Why do people say that so often? It's obvious that nobody understands what it means.
It doesn't mean the customer can do no wrong, or that they should get whatever they want. It means you don't make an issue out of something that really doesn't matter. When they say they want a better pair of speakers for their Panasony stereo surround-sound system, you don't tell them the speakers that came with it are as good at it gets, you don't tell them that they've confused the brand name, and you don't tell them that stereo isn't surround-sound. You just let them go right along being right, and sell them the fancy looking expensive speakers. THAT is what "the customer is always right" really means.
They buy products, apply for rebates, return the purchases, then buy them back at returned-merchandise discounts. They load up on "loss leaders," severely discounted merchandise designed to boost store traffic, then flip the goods at a profit on eBay. They slap down rock-bottom price quotes from Web sites and demand that Best Buy make good on its lowest-price pledge.
It really pisses me off that he groups those together, like people who bought a discounted computer are just as bad as those who are getting rebate checks from items they've already returned to the store. It also pisses me off that they are angry that some customers are forcing them to really honor the terms THEY MADE UP about price matching... if you don't like it, change your policy. Either way, accept the consequences of your own terms and shut up. Hell, the customers are forced to accept the terms even if they don't want to, why shouldn't Best Buy?
My mother regularly saves 30%~50% on groceries because she clips coupons and uses her frequent shopper card.
All stores lie to their customers to PRETEND they're getting a discount.
The other day I was in a store and saw a sign for Levi's 501s for $38, MSRP $42. Then I went to the store next door and saw a similar sign: Levi's 501s for $34, MSRP $38. There isn't any way to reconcile this, other than realizing that stores are lying through their teeth.
Back to the point, I also saw a documentary a few months ago about shopping clubs. They went to a city with two supermarkets with shopping clubs and bought a good list of various items with their club cards. Then they went to a neighboring city, bought the same items from a similar supermarket with no shopping club, and the non-discounted items without a card were a LOT cheaper.
The moral of this story... Stores are lying to you. Shopping clubs only pretend to save you money. You have to compare not to the non-discounted price at the same store, but to the normal price at a non-shopping club store. Club cards are only useful if you can't chose to shop elsewhere.
Besides, the customer tracking is completely unacceptable. It only serves to find out how to squeeze more money out of you. Just think about it, they can find out what products are selling in aggregate without tracking each buyer. They only need per-buyer info to figure out how to make their store less attractive to customers who aren't making them much money. Club cards don't magically make it cheaper for the supermarket to buy that head of lettuce, or can of soup.
First, let me say that white in general is not cheap. I would be interested in what technology you would want to use to turn hundreds of millions of acres of rooftops and roads white.
If you don't like white, any reflective surface would be fine. How about chrome, nickel, etc.? Hell, any light color would be tremendously better than the current color: dark black tar.
As for rooftops, within a few decades they will mostly be solar.
Yes, well, in the intervening century, we can make better use of them, and quite cheaply.
Probably within ten years, we will have reasonably efficienct, cheap plastic solar cells.
I hope so as much as anybody, but I'm certainly not optimistic about that. Developments in solar are going quite slow, and I expect it to take several decades, not just one.
where each process runs on its own processor, and the bandwidth through the pipes is very large.
That's nice and all, but as I said, those use such a trivial percentage of the CPU power, that it wouldn't be worth $10 for a 100% speed-up on non-compression tasks. Most CPU cycles in video playback are wasted on displaying the content, not on the actual decompression. Scaling takes very little. Croping takes almost no CPU time. etc.
Video processing often *is* highly amenable to parallel processing: the coordinating process ships off a discrete chunk of video -- a GOP perhaps, or a chapter -- and gets back a compressed version at some point in the future.
If you were to do it in huge chunks (chapters) it could work, but not easily, and not without losses.
First of all, you'd have to do a first-pass on a single processor with all the motion estimation, scene-change detection, b-frame placement, etc. Then, you can't just cut it up at the I-frames because then the frames at the end of one group can't insert some b-frames which would depend on the I-frame of the next group, which would hurt quality a bit.
The same is true for I-frame placement. Most codecs dynamically decide where to place them, depending on how close to the specified bitrate it currently is, as well as scene-changes. You'd also risk exceeding your specified I-frame interval, or rather, placing extra I-frames where they aren't really needed.
You'd also be completely throwing-off ratecontrol, since one chip really won't know what average bitrate has been used by (eg.) the 7 segements before it. Requiring ratecontrol to make each segment of the video come out to exactly the ABR would maintain the appropriate bitrate, but would lead to even worse quality.
So, basically, you've moving most of the processing burden from the (currently slow) 2nd pass, into the (currently fast) first pass. Since, the first pass can be made very fast by disabling or lowering the strength of some of these computations (the the "turbo" option in Xvid, lavc, etc), I can't see the net speed being much of a gain, and again, you have still lowered quality slightly, this would only work well on very long videos, and you'd be paying a lot for the privlidge.
It occurs to me that, once you've written software that could manage all of this, you could just send each segment of video to a different machine over the network (and back) instead of to a cell, as delay in this design wouldn't be so important.
Finally there is a per-frame parallelization in a streaming approach that can *help* with quality, notably exhaustively searching fullpixel motion estimation in the compression stage is likely to be made quicker by doing different algorithms on different SPUs in parallel, and discarding the lower-quality results returned from them.
Well, that much may work, if the uncompressed frame along with motion vectors and other info can be transfered quickly enough from one chip to several others, run on those chips, and returned all in less than, say 1/100th of a second. However, perhaps a single (second) CPU could already do all that, as could a single ASIC. So, perhaps an advantage, but likely a very small one.
A set of SPUs can do the cropping, scaling, static denoise, temporal denoise, gamma adjustment and compression while another set of SPUs can adjust audio volume, dynamic range and compression. This approach would blow the socks off a fast standard core + single SIMD engine.
The compression is the CPU-intensive process. All the rest in your list uses practically no CPU time on a standard processor. Denoising can waste a lot of cycles (a small fraction compared to compression), but it doesn't have-to, since you can combine it with compression and avoid doing the same number-crunching twice in a row.
Additionally, video compression doesn't work well in parallel. Threading hurts quality a few percent at a given bitrate, and I can't imagine any way that could be eliminated, to make this very useful.
Well, it is pretty hard to wrestle useful information out of the spec.org website... From what I found, it looks like the SpecFP of the fastest Opteron is 75% of the fastest Itanium 2, with the fastest POWER being a bit better still.
Right now the Pentium 4s and Athlons invest a huge number of transistors to coming up with useful microcode
"huge" doesn't tell me much. Would you like to attach a number to that? I don't follow CPU development closely, but it is commonly said that the instruction translator is now really quite tiny and insignificant in comparison to the rest of the chip. Sources would be welcomed.
1) decreased die size 2) increased cache
See above. If it's only a difference of 0.5%, there's no reason to worry about it.
3) increased functional clock speed
Reality hasn't borne this out. RISC processors have been clocked much slower for many years now. The final Alphas were clocked at about 1.2GHz when AMD/Intel was up to 2GHz. Look at Power or SPARC processors, and you don't see high clock speeds.
We aren't talking about general-purpose vs custom chips. We're talking about x86 vs. (eg.) Alpha. Multipurpose chip vs multipurpose chip.
But, if you want to produce realtimee graphiccs for video games, Cell is the way to go...
That's nice and all, but I don't want my x86 processor to render videogames... I want my videocard's ASIC to do that. Videocards are getting very impressive, particularly in comparison to consoles.
Who knows, maybe in a few years, we'll have some variation on cell processors built right in to videocards.
It's always wonderful to see that some of the slashdot-readers are so much smarter than people that actually do research in the area.
To tell you the truth, that wouldn't entirely surprise me. Large-scale group-think is more pervasive in science than practically any other field.
Watch the National Geographic channel on any given day to hear about numerous cases in the not-so-distant past, where all the scientists in the world were convinced their theory was correct, and they would completely write-off all data that was not in agreement, and shun anyone who proposed alternatives to well-accepted theories.
For the most recent example, just ask yourself, what changed about eggs, that suddenly they aren't bad for you anymore? Remember, until just a few years ago, cholesterol was bad for you. Now it has been divided up into "good" and "bad" cholestorol, and some of the things that were terrible for you, is now health-food.
Scientific concensus, at least in the short-term, is not a convincing argument. In the case of global warming, as much as a 30% increase in solar output, combined with the new findings that plants and soil actually RELEASE greehouse gases (like nitrogen) when they experience higher tempuratures, certainly casts reasonable doubt on the theory. I think it wasn't until just this past year that scientists were publicly admitting that global warming isn't ENTIRELY caused by human activities. Maybe next year we'll find out that global warming is good for polar bears... (okay, not likely)
It's only the vastly ignorant that will ignore alternative theories, because it doesn't happen to agree with the theory you currently like.
Simply put, the economics of global warming solutions are just terrible. You really have to stretch to come up with a cost-benefit that justifies actually doing much about global warming.
Really? Because I can think of quite a few.
How about some regulations saying roofs must be painted white? The dark (commonly black tar) roofs in cities have been shown to actually cause a local rise in tempuratures of several degrees.
How about changing road materials to cement, or at least putting an additive into asphalt that will make it white, or at least much lighter?
How about expanding energy production by windmills, which not only produce rather economical electricity without polluting, but actually do it by taking energy out of the atmosphere? If that's too much, just STOP ADDINIG REGULATIONS which make it more difficult to put up windmills.. A nearby city ordinance, just passed, now requires $600 in fees to cover the paperwork and inspection before you can install a windmill. This was done only because neighbors had complaints about the small ammount of noise they produce. And we're talking about a rural area out in the desert, where neighbors have several acres between them...
How about subsudies for home-owners to replace their oil-based heating systems with a ground-source heatpump? Not only would that save home owners lots of money, it would also reduce demand for oil (reducing prices slightly) and vastly reduce the overall ammount of energy needed for home heating/cooling.
How about requiring everyone in the world to have a shaved head? Thereby reflecting much of the sunlight back into space, as well as eliminating billions of cans of aerosol hairspray each year...
These are all very good and rather inexpensive ideas, which would have a very clear benefit, even if global warming isn't really happening (okay, okay, except for that last one, but I just had to throw it in there). It always drives me up a wall when I hear greenpeace or the sierra club, or any other eco-crazies telling us that must all start riding our bicycles to work, or else the earth will be turned into a charred hell-scale in 10 years. There are oh-so-many, many things we can do, which are quite conventional, and will have much more impact... all without freezing your balls off.
Last Gasp for Big Iron? [...] Today's super computers are parallel computing down with 64bit Gen x86 processors, like the AMD Opteron.
Clusters are only really good for embarassingly parallel problems. The interconnects just can't be as fast as a local bus.
How's this for an alternative: "Commodity Big Iron"?
Why can't a supercomputer be based largely on off-the-shelf CPUs, RAM, Drives, etc.? I know important pieces are missing, but it should be possible to open this stuff up, and have it become a commodity.
Too bad HP won't spend $$$ to bring back the Alpha.
I miss architecture diversity....
It seems to me, just about all the huge advantages that alternative architectures (like the Alpha) held over x86, have been washed away in the past few years.
64-bit memory space. Insanely large cache. Very low-latency access to RAM. Incredible memory throughput. PCI-X/PCI Express slots on cheap motherboards. Seriously high-end graphics. DMA. SMP. Built-in 1000Mbps NICs. RAID. etc.
What advantages could something like Alpha have over x86 now? A few years ago, I was anxious to jump ship to another platform, but with the introduction of the Opteron and kin, I'd say I'm quite happy with x86 now.
The only feature I really want now is a new way to handle interrupts... Then simple things like copying CDs, or a little network traffic won't bring PCs to a crawl. Perhaps add a socket for an FPGA or other simple processor to specifically handle those tasks, like the math coprocessors of the old days.
"Alternatives to novocaine" is an ever-evolving field of study. Every month, new information is published in places like The New England Journal of Medicine with updated test results and data about the safety and success rates of such drugs.
No, no, no, no. I never asked about the viability of the alternatives, I just asked WHAT THEY ARE. A list of them, not a medial paper on their pros and cons.
It's not really something that is wise of the layman to "dabble" in.
What are you talking about? How can a little bit of knowledge on the subject be dangerous. We aren't talking about experimental treatments, or urban myths about safety equipment, we are talking about anestesia, which has to be administed by a doctor, anyhow. Unless you are self-medicating (which makes no sense in the context of novocaine) you can't possibly know "just enough to be dangerous."
For example, the post which kicked this off (somebody who didn't know what deus ex machina meant [...]) is exactly the sort of thing one turns to an encyclopedia for.
No, as a matter of fact, definitions of terms are exactly the sort of thing ones turns to a dictionary for. Information on drugs and their derivitives and alternatives is exactly the kind of thing one turns to an encyclopedia for.
Simply author a file that you are the copy holder and then sue them for restraint of trade or insterstate commerce problems.
No. They aren't outlawing filesharing, as the good-ole/. editors would like you to think.
The RIAA are trying to make it illegal to make copyrighted files available for sharing, even if NOBODY has downloaded them from you yet. Obviously, right now, it's not copyright infringement until you can prove someone has actually COPIED them.
Because we aren't talking about a typo here, we're talking about entire subjects being screwed-up beyond belief.
I've contributed to wikipedia plenty, as a matter of fact. But I'm certainly not an expert on everything I look up, and unlike many others, I'm not going to directly plagarize another website. Hell, even if I was an expert on everything, I wouldn't even think of spending hours every day looking through Wikipedia, and fixing every mistake, only to have it re-fucked-up by some idiot.
Reverts are rare on all but the most popular articles, and it's not because nobody screws-up the less popular ones... quite the opposite in-fact.
For two, Google, after all, is a business. They are not a NGO, charity, or some other organization that's in existance to make this planet a better World (TM). They are here to make their shareholders (and themselves) a return on their investment.
Thankfully, the two overlap. If people trust google less, the stock price will fall.
The real question is whether they're going to go for the short-term profit or the long-term profit. In the short-term, selling user information will make money, but you will quickly lose customers, and in the long-term, money.
Google, being a homogeneous information service, depends far more on customer trust than (nearly) any other type of company (except perhaps banks). Losing face with their customers will cause them to lose a lot of money for their shareholders as well.
I think it says a lot that Wikipedia is now the default first choice for looking up damn near any non-controversial subect.
Wikipedia is a good source for things that/.ers are likely to know a lot about... Video games, radio communications, computers, etc. However, it really falls flat on non-technology subjects.
One I came across most recently: Using only wikipedia, try to come up with a list of all the alternatives to novocaine... You will be graded on accuracy.
I've seen a lot of articles that were innaccurate just because they didn't plagarize them well enough, changing a few words here and there, which the writer obviously didn't understand.
Then there are plenty of subjects that are largely decent, but somebody threw in a couple sentences of complete nonsense in the middle of them, and it just stays there for years.
Wikipedia has a lot of shortcommings, which will become more obvious as it's popularity among the general public increases. Something akin to USENET or E-MAIL in the beginning.
Okay, I agree that this may not be a great approach, but I don't think that a procedure that requires skilled technicians and a cleanroom to do reliably qualifies as "trivial".
Cleanroom? What cleanroom?
The circuit boards for hard drives are just held on by a few screws, and a simple plug attaches it to the voice coil, heads, and motor. All it requires is a star-shaped screwdriver (usually T-10) and a wrist-strap.
I'm tired of being told it is illegal to play DVDs on my linux-based laptop even though I own the DVD and have no DVD ripping libraries on my computer.
Circumventing CSS for interoperability is explicitly legal... The original DeCSS/CSS-auth code, however, was declared illegal due to the reverse-engineering of commercial software to get the key, not due to the circumvention clause of the DMCA.
Nobody uses DeCSS/css-auth anymore. It was replaced by libdvdcss, which was a clean-room implimentation. There shouldn't be any reason that using libdvdcss should be inherently illegal.
Ironically, the law most Linux users are running afoul of is not the DMCA, but patent law. The guys that wrote those MPEG-2 and AC3 decoder routines did not pay the necessary license fees to make it legal for you to use.
So, did you send-in your $2.50 to the MPEG-LA and $1.98 to Dolby to license MPEG-2 and AC3? If you encode to MPEG-4 with Xvid or lavc, and re-encode the audio to MP3, you've got even more patent-license fees to pay. You HAVE sent-in your fees, right???
I only point that out to illustrate that selling out your rights to further rapacious corporate profits is not, and never has been, exclusively a Republican trait.
Living in CA, with two democratic senators, both under the thumb of Hollywood, I know that's certainly true. However, you have to admit that Republicans are overwhelmingly ahead of Democrats in this regard.
Slashdot Mirror
Which is why I said it...
Actually, I don't think it's OFTEN the case. Depending on where you are and what stores are around, driving 50 miles out of your way might actually save you a lot of money, and allow you to maintain some privacy as well. I'm not in a big city, but it's still only a 15 mile drive to get to a much better supermarket than the 2 around here.
Why do people say that so often? It's obvious that nobody understands what it means.
It doesn't mean the customer can do no wrong, or that they should get whatever they want. It means you don't make an issue out of something that really doesn't matter. When they say they want a better pair of speakers for their Panasony stereo surround-sound system, you don't tell them the speakers that came with it are as good at it gets, you don't tell them that they've confused the brand name, and you don't tell them that stereo isn't surround-sound. You just let them go right along being right, and sell them the fancy looking expensive speakers. THAT is what "the customer is always right" really means.
It really pisses me off that he groups those together, like people who bought a discounted computer are just as bad as those who are getting rebate checks from items they've already returned to the store. It also pisses me off that they are angry that some customers are forcing them to really honor the terms THEY MADE UP about price matching... if you don't like it, change your policy. Either way, accept the consequences of your own terms and shut up. Hell, the customers are forced to accept the terms even if they don't want to, why shouldn't Best Buy?
All stores lie to their customers to PRETEND they're getting a discount.
The other day I was in a store and saw a sign for Levi's 501s for $38, MSRP $42. Then I went to the store next door and saw a similar sign: Levi's 501s for $34, MSRP $38. There isn't any way to reconcile this, other than realizing that stores are lying through their teeth.
Back to the point, I also saw a documentary a few months ago about shopping clubs. They went to a city with two supermarkets with shopping clubs and bought a good list of various items with their club cards. Then they went to a neighboring city, bought the same items from a similar supermarket with no shopping club, and the non-discounted items without a card were a LOT cheaper.
The moral of this story... Stores are lying to you. Shopping clubs only pretend to save you money. You have to compare not to the non-discounted price at the same store, but to the normal price at a non-shopping club store. Club cards are only useful if you can't chose to shop elsewhere.
Besides, the customer tracking is completely unacceptable. It only serves to find out how to squeeze more money out of you. Just think about it, they can find out what products are selling in aggregate without tracking each buyer. They only need per-buyer info to figure out how to make their store less attractive to customers who aren't making them much money. Club cards don't magically make it cheaper for the supermarket to buy that head of lettuce, or can of soup.
If you don't like white, any reflective surface would be fine. How about chrome, nickel, etc.? Hell, any light color would be tremendously better than the current color: dark black tar.
Yes, well, in the intervening century, we can make better use of them, and quite cheaply.
I hope so as much as anybody, but I'm certainly not optimistic about that. Developments in solar are going quite slow, and I expect it to take several decades, not just one.
That's nice and all, but as I said, those use such a trivial percentage of the CPU power, that it wouldn't be worth $10 for a 100% speed-up on non-compression tasks. Most CPU cycles in video playback are wasted on displaying the content, not on the actual decompression. Scaling takes very little. Croping takes almost no CPU time. etc.
If you were to do it in huge chunks (chapters) it could work, but not easily, and not without losses.
First of all, you'd have to do a first-pass on a single processor with all the motion estimation, scene-change detection, b-frame placement, etc. Then, you can't just cut it up at the I-frames because then the frames at the end of one group can't insert some b-frames which would depend on the I-frame of the next group, which would hurt quality a bit.
The same is true for I-frame placement. Most codecs dynamically decide where to place them, depending on how close to the specified bitrate it currently is, as well as scene-changes. You'd also risk exceeding your specified I-frame interval, or rather, placing extra I-frames where they aren't really needed.
You'd also be completely throwing-off ratecontrol, since one chip really won't know what average bitrate has been used by (eg.) the 7 segements before it. Requiring ratecontrol to make each segment of the video come out to exactly the ABR would maintain the appropriate bitrate, but would lead to even worse quality.
So, basically, you've moving most of the processing burden from the (currently slow) 2nd pass, into the (currently fast) first pass. Since, the first pass can be made very fast by disabling or lowering the strength of some of these computations (the the "turbo" option in Xvid, lavc, etc), I can't see the net speed being much of a gain, and again, you have still lowered quality slightly, this would only work well on very long videos, and you'd be paying a lot for the privlidge.
It occurs to me that, once you've written software that could manage all of this, you could just send each segment of video to a different machine over the network (and back) instead of to a cell, as delay in this design wouldn't be so important.
Well, that much may work, if the uncompressed frame along with motion vectors and other info can be transfered quickly enough from one chip to several others, run on those chips, and returned all in less than, say 1/100th of a second. However, perhaps a single (second) CPU could already do all that, as could a single ASIC. So, perhaps an advantage, but likely a very small one.
Okay, we'll count that as #1
The compression is the CPU-intensive process. All the rest in your list uses practically no CPU time on a standard processor. Denoising can waste a lot of cycles (a small fraction compared to compression), but it doesn't have-to, since you can combine it with compression and avoid doing the same number-crunching twice in a row.
Additionally, video compression doesn't work well in parallel. Threading hurts quality a few percent at a given bitrate, and I can't imagine any way that could be eliminated, to make this very useful.
Well, it is pretty hard to wrestle useful information out of the spec.org website... From what I found, it looks like the SpecFP of the fastest Opteron is 75% of the fastest Itanium 2, with the fastest POWER being a bit better still.
"huge" doesn't tell me much. Would you like to attach a number to that? I don't follow CPU development closely, but it is commonly said that the instruction translator is now really quite tiny and insignificant in comparison to the rest of the chip. Sources would be welcomed.
See above. If it's only a difference of 0.5%, there's no reason to worry about it.
Reality hasn't borne this out. RISC processors have been clocked much slower for many years now. The final Alphas were clocked at about 1.2GHz when AMD/Intel was up to 2GHz. Look at Power or SPARC processors, and you don't see high clock speeds.
That's nice and all, but I don't want my x86 processor to render videogames... I want my videocard's ASIC to do that. Videocards are getting very impressive, particularly in comparison to consoles.
Who knows, maybe in a few years, we'll have some variation on cell processors built right in to videocards.
To tell you the truth, that wouldn't entirely surprise me. Large-scale group-think is more pervasive in science than practically any other field.
Watch the National Geographic channel on any given day to hear about numerous cases in the not-so-distant past, where all the scientists in the world were convinced their theory was correct, and they would completely write-off all data that was not in agreement, and shun anyone who proposed alternatives to well-accepted theories.
For the most recent example, just ask yourself, what changed about eggs, that suddenly they aren't bad for you anymore? Remember, until just a few years ago, cholesterol was bad for you. Now it has been divided up into "good" and "bad" cholestorol, and some of the things that were terrible for you, is now health-food.
Scientific concensus, at least in the short-term, is not a convincing argument. In the case of global warming, as much as a 30% increase in solar output, combined with the new findings that plants and soil actually RELEASE greehouse gases (like nitrogen) when they experience higher tempuratures, certainly casts reasonable doubt on the theory. I think it wasn't until just this past year that scientists were publicly admitting that global warming isn't ENTIRELY caused by human activities. Maybe next year we'll find out that global warming is good for polar bears... (okay, not likely)
It's only the vastly ignorant that will ignore alternative theories, because it doesn't happen to agree with the theory you currently like.
Really? Because I can think of quite a few.
How about some regulations saying roofs must be painted white? The dark (commonly black tar) roofs in cities have been shown to actually cause a local rise in tempuratures of several degrees.
How about changing road materials to cement, or at least putting an additive into asphalt that will make it white, or at least much lighter?
How about expanding energy production by windmills, which not only produce rather economical electricity without polluting, but actually do it by taking energy out of the atmosphere? If that's too much, just STOP ADDINIG REGULATIONS which make it more difficult to put up windmills.. A nearby city ordinance, just passed, now requires $600 in fees to cover the paperwork and inspection before you can install a windmill. This was done only because neighbors had complaints about the small ammount of noise they produce. And we're talking about a rural area out in the desert, where neighbors have several acres between them...
How about subsudies for home-owners to replace their oil-based heating systems with a ground-source heatpump? Not only would that save home owners lots of money, it would also reduce demand for oil (reducing prices slightly) and vastly reduce the overall ammount of energy needed for home heating/cooling.
How about requiring everyone in the world to have a shaved head? Thereby reflecting much of the sunlight back into space, as well as eliminating billions of cans of aerosol hairspray each year...
These are all very good and rather inexpensive ideas, which would have a very clear benefit, even if global warming isn't really happening (okay, okay, except for that last one, but I just had to throw it in there). It always drives me up a wall when I hear greenpeace or the sierra club, or any other eco-crazies telling us that must all start riding our bicycles to work, or else the earth will be turned into a charred hell-scale in 10 years. There are oh-so-many, many things we can do, which are quite conventional, and will have much more impact... all without freezing your balls off.
Clusters are only really good for embarassingly parallel problems. The interconnects just can't be as fast as a local bus.
How's this for an alternative: "Commodity Big Iron"?
Why can't a supercomputer be based largely on off-the-shelf CPUs, RAM, Drives, etc.? I know important pieces are missing, but it should be possible to open this stuff up, and have it become a commodity.
It seems to me, just about all the huge advantages that alternative architectures (like the Alpha) held over x86, have been washed away in the past few years.
64-bit memory space. Insanely large cache. Very low-latency access to RAM. Incredible memory throughput. PCI-X/PCI Express slots on cheap motherboards. Seriously high-end graphics. DMA. SMP. Built-in 1000Mbps NICs. RAID. etc.
What advantages could something like Alpha have over x86 now? A few years ago, I was anxious to jump ship to another platform, but with the introduction of the Opteron and kin, I'd say I'm quite happy with x86 now.
The only feature I really want now is a new way to handle interrupts... Then simple things like copying CDs, or a little network traffic won't bring PCs to a crawl. Perhaps add a socket for an FPGA or other simple processor to specifically handle those tasks, like the math coprocessors of the old days.
No, no, no, no. I never asked about the viability of the alternatives, I just asked WHAT THEY ARE. A list of them, not a medial paper on their pros and cons.
What are you talking about? How can a little bit of knowledge on the subject be dangerous. We aren't talking about experimental treatments, or urban myths about safety equipment, we are talking about anestesia, which has to be administed by a doctor, anyhow. Unless you are self-medicating (which makes no sense in the context of novocaine) you can't possibly know "just enough to be dangerous."
No, as a matter of fact, definitions of terms are exactly the sort of thing ones turns to a dictionary for. Information on drugs and their derivitives and alternatives is exactly the kind of thing one turns to an encyclopedia for.
No. They aren't outlawing filesharing, as the good-ole
The RIAA are trying to make it illegal to make copyrighted files available for sharing, even if NOBODY has downloaded them from you yet. Obviously, right now, it's not copyright infringement until you can prove someone has actually COPIED them.
Because we aren't talking about a typo here, we're talking about entire subjects being screwed-up beyond belief.
I've contributed to wikipedia plenty, as a matter of fact. But I'm certainly not an expert on everything I look up, and unlike many others, I'm not going to directly plagarize another website. Hell, even if I was an expert on everything, I wouldn't even think of spending hours every day looking through Wikipedia, and fixing every mistake, only to have it re-fucked-up by some idiot.
Reverts are rare on all but the most popular articles, and it's not because nobody screws-up the less popular ones... quite the opposite in-fact.
Thankfully, the two overlap. If people trust google less, the stock price will fall.
The real question is whether they're going to go for the short-term profit or the long-term profit. In the short-term, selling user information will make money, but you will quickly lose customers, and in the long-term, money.
Google, being a homogeneous information service, depends far more on customer trust than (nearly) any other type of company (except perhaps banks). Losing face with their customers will cause them to lose a lot of money for their shareholders as well.
No, he used-to be that. Now he's just a loud-mouthed child with limitless unfunny gags about repressed homosexuality.
He hasn't been interesting in a long time.
Wikipedia is a good source for things that
One I came across most recently: Using only wikipedia, try to come up with a list of all the alternatives to novocaine... You will be graded on accuracy.
I've seen a lot of articles that were innaccurate just because they didn't plagarize them well enough, changing a few words here and there, which the writer obviously didn't understand.
Then there are plenty of subjects that are largely decent, but somebody threw in a couple sentences of complete nonsense in the middle of them, and it just stays there for years.
Wikipedia has a lot of shortcommings, which will become more obvious as it's popularity among the general public increases. Something akin to USENET or E-MAIL in the beginning.
Cleanroom? What cleanroom?
The circuit boards for hard drives are just held on by a few screws, and a simple plug attaches it to the voice coil, heads, and motor. All it requires is a star-shaped screwdriver (usually T-10) and a wrist-strap.
Would you believe I did that on purpose?
Good luck. OS X isn't even X11-based, so you're going to have a hell of a porting job to do, or else endless layers of compatibility software.
Circumventing CSS for interoperability is explicitly legal... The original DeCSS/CSS-auth code, however, was declared illegal due to the reverse-engineering of commercial software to get the key, not due to the circumvention clause of the DMCA.
Nobody uses DeCSS/css-auth anymore. It was replaced by libdvdcss, which was a clean-room implimentation. There shouldn't be any reason that using libdvdcss should be inherently illegal.
Ironically, the law most Linux users are running afoul of is not the DMCA, but patent law. The guys that wrote those MPEG-2 and AC3 decoder routines did not pay the necessary license fees to make it legal for you to use.
So, did you send-in your $2.50 to the MPEG-LA and $1.98 to Dolby to license MPEG-2 and AC3? If you encode to MPEG-4 with Xvid or lavc, and re-encode the audio to MP3, you've got even more patent-license fees to pay. You HAVE sent-in your fees, right???
Living in CA, with two democratic senators, both under the thumb of Hollywood, I know that's certainly true. However, you have to admit that Republicans are overwhelmingly ahead of Democrats in this regard.