The point being missed is that the cost of a CD has to cover much more than just the manufacturing and distribution of that disk. It has to cover the expenses in finding and developing talent, recording and touring, marketing and advertising. These expenses far exceed the costs of pressing a CD.
Of course those costs are not fixed in stone. There's no fundamental reason that the record companies have to spend the way they do. They could instead try to sign lots and lots of bands, use the cheapest promotions they can come up with, and see which ones happen to work out well. They could, for instance, offer selected tracks by the bands for free download over existing P2P systems instead of advertizing them through pay per ad media.
I don't see anyone, not indy labels, not bands selling their own music, who sells CDs for a few bucks each.
Then you're looking in the wrong genres. In classical music, for instance, there are plenty of dirt cheap disks produced by relatively unknown orchestras. They spend very little on advertizing, have very simple labels, and sell because some people want to get copies of the classics without having to pay a top dollar price. It's a working model. If a major label adopted a similar model for their unknown bands- don't pay for hype, sell the disks for $4-5, and wait for some of the bands to get good word of mouth- I'm sure they could make it work. They could advertize their unknowns label- "Get music by new, upcoming groups for dirt cheap!"- to some extent, but mostly they'd try to keep the budget as low as possible so that the label would pay for itself before the profits from the groups that do make it. It could definitely work.
Come to think of it, why have we never seen an aerial photo of the Apollo 11 landing site taken from Earth or Earth orbit? It can't be too difficult, can it?
Yes, it can be. As an approximation you can figure the resolution of a telescope at a given distance as:
size of mirror/wavelength of light = distance to object/size of object
Given that the distance to the moon is about 500 million meters, the lunar lander is about 10 meters across, and visible light has a wavelength of about 5e-7 meters, that means that you need a mirror about 25 meters across to see it. That's about 10 times the size of the Hubble mirror, and 2.5 times the size of the Keck mirror. Of course you can't see it with a ground based telescope anyway because they'll have problems with atmospheric distortion. And that's just the resolution you'd need to be able to spot it as a speck. You'd need to multiply the size of the mirror by the number of pixels you want in your picture, so a 10x10 pixel picture (still not exactly detailed) would need an optical telescope about the size of the Arecibo radio telescope.
Sadly that's not the case. You just can't convince somebody who truly wants to believe that it's all a conspiracy. They'll point out that this supposedly independent private company had to get government approval to do so, and that's proof that NASA got to themm and forced them to take part in the deception. It took me all of about 2 seconds to come up with that explanation. These are people who wouldn't believe that it was possible to go to the moon if you blasted them into space and landed them there. They'd still come up with some elaborate explanation about how it was all faked.
There are none so blind as those who will not see. Sadly this applies as much to physical proof as anything.
Well, there are potentially three points. One is that hopefully after a while the filter will work well enough that you can develop some real confidence in it and you won't have to check every time to see that it's working right. I'm pretty close to that point with bogofilter; I so rarely see any false positives that I can almost afford to flush the messages without checking. Actually, I assume that what I'll really do is to change the rules a bit so that alleged spam is sent to a waiting folder and doesn't even show up in my main inbox.
That gets to point two: now I'll be able to check for spam in batch mode. Instead of going through my inbox every time I look for messages, marking some as spam and reading others, I'll be able to read just about everything in my inbox without worrying about spam. Then once a week I can check my spam box and see if there's actually anything legitimate there. This is going to be faster than doing it every time a new message shows up in my inbox.
I'm not a compulsive mail reader, but for some people this would also be really useful because it would protect them from distractions. They are working on something and then their mailbox beeps them to let them know that a message has arrived. Unfortunately, when they check it out it turns out that their train of thought has been needlessly disrupted by another spam. If they can filter out the spam before the notification while still being alerted promptly when a real message shows up, that's a big win.
With some cleverness, you can use any outside filter with the most recent version (i.e. the develpment fork) of Evolution. They've added the ability to pipe incoming messages to an outside program and read back the exit code. So if the program is written using standard Unixisms- i.e. it reads on standard input and returns a different value depending on whether the incoming message is spam or not- it can be used with Evolution. I know that bogofilter can do this because I'm using it with Evolution and it works pretty well.
Well, I guess it just depends on what you consider to be a usefully low false negative rate. I'd guess that mine flags 90+% of incomming spam correctly, and that's with just a few weeks worth of spam to learn from. I think that reducing spam by 90% without a serious problem with false positives (I've had one; a html message from my new ISP. After telling the filter that it was wanted, future messages made it through fine.) is useful. I have every reason to think that it will continue to get better, and that's good enough for me. I also suspect that if my spam load gets higher, the filter will improve quickly enough to catch up.
Free hint: perl runs under Windows, too. True, it's not pre-installed on every system, and it's a bit bigger than you'd probably want as part of a simple install, but there's no real barrier to installing perl on a Windows box. With a bit of cleverness, I'd bet that you could make an installer figure out if the system has perl and install it if it's not there already.
But perl scripts are just as easy to run as.exe files, so long as you have the perl interpreter installed. So now it's just a two step process:
Install perl.
Install the perl script.
This is not exactly brain surgery. Perl can be installed on essentially any system you choose to name, with no more trouble than installing any other executable. For those people running Windows, there's an excellent port available from Activestate. As somebody else pointed out, this means that a perl script is actually available to more people than a.exe would be, because it's truly cross-platform.
You're wrong, though. The whole point of this kind of filter is that it develops its rules based on the information that you give it, not what somebody else thinks. If you tell it that mails from your legitimate business partners aren't spam, it learns to tell them apart. I use a Bayesian filter on my mail, and it has no trouble telling my legitimate business mail, like messages from Amazon about books I've been waiting for, from illegitimate ones. Some of that is that the legitimate mail is written with a very different style from the illegitimate stuff, but I assume that the filter has also learned that mail with amazon.com as the sender is OK. In any case, I find that it just plain works.
Bogofilter comes close to this. It has an operating mode where each file that it filters is automatically added to the appropriate corpus, either of spam or non-spam. Since it's correct the vast majority of the time, that means that there's very little for the user to do. When it is wrong, you just take the messages that it miscategorized and feed them back into the system with the notation that they were originally marked incorrectly, and it backs out the changes to the wrong category and adds them to the correct category.
I'm using bogofilter with Evolution, and it works very well. I just have two extra folders, one for false negatives and one for false positives. When I notice mail that's been flagged incorrectly, I drag it into the appropriate folder and run a script that tells bogofilter to correct its mistake. Then I either flush the mail (if it was spam marked as non-spam) or process it normally (if it was non-spam marked as spam). I've only been using it for about two weeks and it already has a nearly zero false positive rate (i.e. incorrectly flagged as spam) and a usefully low false negative rate (i.e. incorrectly flagged as legitimate).
Another important point is that there are some things that they can't hide, at least not in their current working model. If they're trying to sell you something, they have to describe what that thing is and where you can get it, and those descriptions are unlikely to be in any legitimate email. If they want to advertize a web site, they have to include its URL in the message, and the filter can catch that. If they advertize a physical address or phone number, the system can catch those, too. If they don't repeat the message, it means that there's inherently less spam, because I'm only seeing each add once.
It's also not possible to disguise everything in their headers, so things like their posting host (either the one they pay for legitimately or any open relay they're taking advantage of) will wind up being a pointer to who they are. They certainly can't change anything about the headers that's added downstream of their posting host, so as long as they keep using the same one it's likely that there will be characteristic stamps there that the spammers absolutely can't change. I know that analysis of the headers is part of bogofilter, another Bayesian filter that I've been using to good effect.
I haven't ever worked in a place where the need for computing power has varied wildly over time (which is the only scenario for which this model seems to make any sense) so I don't know how common this market is or how valuable the service will be.
It might be more common than you think. In my work, for instance, we have occasional spurts where we generate a large amount of scientific data that needs to be processed followed by long periods where it doesn't. We're limited to running on the fastest box we can reasonably afford, but it might be cheaper and faster to buy just the clock cycles we need. One thing that's unclear is whether our processed data would be secure on our own machine or would be on IBM's farm, though. We'd definitely need to keep it on our own machine.
The other thing to consider is that it's possible that there are lots of applications where computer use might vary wildly from time to time, but nobody is thinking about them because they're uneconomical. Most places can't afford to have a supercomputer sitting around idle 95+% of the time, so instead they buy a machine that can process all of their data without the long idle times. The result is that there's a long lag between when the processing starts and when it finishes. If a system were available where they could buy the power to process that data rapidly when needed, though, it might make more sense to do it that way. They might very well wind up with about the same total cost but much faster results.
Because it's not their beach. Beaches in California (up to the mean high tide mark) are public property. They don't have the right to dump those boulders on public property, or in a way that would damage public property (at least without a permit). In California, they aren't legally allowed to deny beach access, either, again because the beach is public property.
Wrong. Because the cell phone owner pays for incoming calls, it's illegal to make an unsolicited call to a cell phone. The telemarketers know this and don't call cell phones. Of course as cell phone rates go down I expect for there to be heavy lobying against this rule, but as it stands right now there are legal obstacles to telemarketing to cell phones.
How much you save probably depends a lot on how you go about stripping out the code. I assume that there are also drivers for hardware that isn't available on x86 machines, so if you stripped out those you'd save some more space. There have to be some drivers specific to IBM mainframes, for instance, that are irrelevant to people using commodity hardware.
What would be really nice would be a system that would let you download a minimal configuration system, run you through the configuration, and then only download the files you need to compile the things you specified. If you run SCSI only, it wouldn't download the IDE drivers. If you run an x86, it wouldn't download the files specific to other processors. If you don't want to talk to Macs, it wouldn't download the HFS+ filesystem or appletalk drivers. I realize that adding a build system like that wouldn't be practical for the kernel developers, but it's nice to dream.
No. It's not bloated. Bloat is unnecessary size. The kernel isn't bloated because that size is necessary. The kernel has a huge number of features: it runs on a zillion different architectures, it can handle almost any hardware you throw at it, and it can handle almost any protocol and format that you might want your kernel to do. Those aren't things that are in there because some marketroid decided to put them in for buzzword compliance. They're things that are in there because actual users wanted them enough to write the code. You're not going to fit all of those available features into a small package, no matter how hard you try.
At the same time, the end product of a compiled kernel can be very compact because it doesn't have to include any features that the user doesn't want. The existence of code for dozens of architectures doesn't result in a bigger compiled kernel because those architectures aren't compiled in. Neither are drivers for hardware not present on your system, or protocols that you're not going to use. The net result is a kernel that's capable of doing whatever you want it to do without carrying the weight of the things that you don't want to do.
But the 32MB is for the compressed source, not for the binary. By the time you've compiled and compressed the binary image, it's typically down to about 5-700 KB, and I'd assume that if you have an embedded system with clearly defined minimal hardware you can probably get it a bit smaller than that. It's certainly not tiny, but the size of the end product (which is what really matters for the vast majority of people) is very reasonable.
Good enough that you have a quote mentioning it in your sig. That said, I'd easily add recommendations for Evangelion and Escaflowne. I'd almost recommend the original Macross, but even with AnimEigo's excellent restoration it hasn't aged that well- certainly don't watch the cheesy Robotech version.
Gundam is probably as close to legendary as any anime series can be. It radically altered the whole climate for SF anime. Perhaps it would be better to call it seminal or groundbreaking, but it was certainly very important in the development of the genre.
#2. How many "publically elected" seats are left? The story just says they eliminated 5 without elaborating.
None. ICANN seems to have decided that having directors who were elected by the net population at large was interfering with their nice, cozy, corrupt way of doing things. Auerbach was only the most obvious example of this.
I'm no expert on law by any means, but doesn't a website have to follow the laws in the country where its server is physically located?
Legally, the question seems to be rather murky. There are some cases where it's clear that you have to obey more than one set of laws. If you're selling products to a country other than where you're located you still have to obey their laws, the same way that you'd have to if you sold their by mail or telephone order. If you're just providing information gratis, the answer seems to be more questionable. Certainly there are a lot of countries that claim that you have to obey their laws if you want to serve web pages to people there.
Practially, you may have to obey the laws of countries other than the one where your servers are located because they can make things very unpleasant for you if you don't. If you want to do business in country X, you have to listen to what they tell you, or they're likely to do nasty things like seize any assets you have in there or arrest your employees when they visit. That may not be a big problem for some countries- Google probably doesn't do much business with North Korea, and their employees probably don't visit there very often- but with places like the EU this is likely to be a problem.
Unfortunately, pressure from the outside is a consequence of Google's influence, and would likely wind up applying to any search engine that reaches Google's caliber. They're internationally significant, so any entity that wants to shut down opposing points of view is likely to target them. (Google may have some additional problems here because they cache the sites and make them available, making them a source of the information directly as well as indirectly.) Unfortunately, being globally significant also means that Google has to obey a lot of different laws in various jurisdictions. That means that they can be legally forced to de-list some sites- Holocaust denial sites in Germany, places selling Nazi memorabilia in France, sites that are alleged to violate copyright in the U.S., or even anything the government disapproves of in some more repressive countries. If you have a suggestion for how they can continue listing everything in spite of the legal restrictions on doing so, I'm sure that Google would be very interested in knowing about it.
It is a vaguely interesting thing. A coworker once mentioned a word that she typed entirely with one hand and speculated about what the longest such word was. Being a geek, I thought that the obvious solution was to write a perl script to find out. I found that the longest words that could be typed entirely with one hand (the left) were 12 letters long; the only one that might actually be used in ordinary usage was stewardesses. ISTR that the longest word typed entirely with the right hand was only 9 letters, but I don't remember any examples.
Slashdot Mirror
Of course those costs are not fixed in stone. There's no fundamental reason that the record companies have to spend the way they do. They could instead try to sign lots and lots of bands, use the cheapest promotions they can come up with, and see which ones happen to work out well. They could, for instance, offer selected tracks by the bands for free download over existing P2P systems instead of advertizing them through pay per ad media.
Then you're looking in the wrong genres. In classical music, for instance, there are plenty of dirt cheap disks produced by relatively unknown orchestras. They spend very little on advertizing, have very simple labels, and sell because some people want to get copies of the classics without having to pay a top dollar price. It's a working model. If a major label adopted a similar model for their unknown bands- don't pay for hype, sell the disks for $4-5, and wait for some of the bands to get good word of mouth- I'm sure they could make it work. They could advertize their unknowns label- "Get music by new, upcoming groups for dirt cheap!"- to some extent, but mostly they'd try to keep the budget as low as possible so that the label would pay for itself before the profits from the groups that do make it. It could definitely work.
Yes, it can be. As an approximation you can figure the resolution of a telescope at a given distance as:
size of mirror/wavelength of light = distance to object/size of object
Given that the distance to the moon is about 500 million meters, the lunar lander is about 10 meters across, and visible light has a wavelength of about 5e-7 meters, that means that you need a mirror about 25 meters across to see it. That's about 10 times the size of the Hubble mirror, and 2.5 times the size of the Keck mirror. Of course you can't see it with a ground based telescope anyway because they'll have problems with atmospheric distortion. And that's just the resolution you'd need to be able to spot it as a speck. You'd need to multiply the size of the mirror by the number of pixels you want in your picture, so a 10x10 pixel picture (still not exactly detailed) would need an optical telescope about the size of the Arecibo radio telescope.
Sadly that's not the case. You just can't convince somebody who truly wants to believe that it's all a conspiracy. They'll point out that this supposedly independent private company had to get government approval to do so, and that's proof that NASA got to themm and forced them to take part in the deception. It took me all of about 2 seconds to come up with that explanation. These are people who wouldn't believe that it was possible to go to the moon if you blasted them into space and landed them there. They'd still come up with some elaborate explanation about how it was all faked.
There are none so blind as those who will not see. Sadly this applies as much to physical proof as anything.
Well, there are potentially three points. One is that hopefully after a while the filter will work well enough that you can develop some real confidence in it and you won't have to check every time to see that it's working right. I'm pretty close to that point with bogofilter; I so rarely see any false positives that I can almost afford to flush the messages without checking. Actually, I assume that what I'll really do is to change the rules a bit so that alleged spam is sent to a waiting folder and doesn't even show up in my main inbox.
That gets to point two: now I'll be able to check for spam in batch mode. Instead of going through my inbox every time I look for messages, marking some as spam and reading others, I'll be able to read just about everything in my inbox without worrying about spam. Then once a week I can check my spam box and see if there's actually anything legitimate there. This is going to be faster than doing it every time a new message shows up in my inbox.
I'm not a compulsive mail reader, but for some people this would also be really useful because it would protect them from distractions. They are working on something and then their mailbox beeps them to let them know that a message has arrived. Unfortunately, when they check it out it turns out that their train of thought has been needlessly disrupted by another spam. If they can filter out the spam before the notification while still being alerted promptly when a real message shows up, that's a big win.
With some cleverness, you can use any outside filter with the most recent version (i.e. the develpment fork) of Evolution. They've added the ability to pipe incoming messages to an outside program and read back the exit code. So if the program is written using standard Unixisms- i.e. it reads on standard input and returns a different value depending on whether the incoming message is spam or not- it can be used with Evolution. I know that bogofilter can do this because I'm using it with Evolution and it works pretty well.
Well, I guess it just depends on what you consider to be a usefully low false negative rate. I'd guess that mine flags 90+% of incomming spam correctly, and that's with just a few weeks worth of spam to learn from. I think that reducing spam by 90% without a serious problem with false positives (I've had one; a html message from my new ISP. After telling the filter that it was wanted, future messages made it through fine.) is useful. I have every reason to think that it will continue to get better, and that's good enough for me. I also suspect that if my spam load gets higher, the filter will improve quickly enough to catch up.
Free hint: perl runs under Windows, too. True, it's not pre-installed on every system, and it's a bit bigger than you'd probably want as part of a simple install, but there's no real barrier to installing perl on a Windows box. With a bit of cleverness, I'd bet that you could make an installer figure out if the system has perl and install it if it's not there already.
But perl scripts are just as easy to run as .exe files, so long as you have the perl interpreter installed. So now it's just a two step process:
This is not exactly brain surgery. Perl can be installed on essentially any system you choose to name, with no more trouble than installing any other executable. For those people running Windows, there's an excellent port available from Activestate. As somebody else pointed out, this means that a perl script is actually available to more people than a .exe would be, because it's truly cross-platform.
You're wrong, though. The whole point of this kind of filter is that it develops its rules based on the information that you give it, not what somebody else thinks. If you tell it that mails from your legitimate business partners aren't spam, it learns to tell them apart. I use a Bayesian filter on my mail, and it has no trouble telling my legitimate business mail, like messages from Amazon about books I've been waiting for, from illegitimate ones. Some of that is that the legitimate mail is written with a very different style from the illegitimate stuff, but I assume that the filter has also learned that mail with amazon.com as the sender is OK. In any case, I find that it just plain works.
Bogofilter comes close to this. It has an operating mode where each file that it filters is automatically added to the appropriate corpus, either of spam or non-spam. Since it's correct the vast majority of the time, that means that there's very little for the user to do. When it is wrong, you just take the messages that it miscategorized and feed them back into the system with the notation that they were originally marked incorrectly, and it backs out the changes to the wrong category and adds them to the correct category.
I'm using bogofilter with Evolution, and it works very well. I just have two extra folders, one for false negatives and one for false positives. When I notice mail that's been flagged incorrectly, I drag it into the appropriate folder and run a script that tells bogofilter to correct its mistake. Then I either flush the mail (if it was spam marked as non-spam) or process it normally (if it was non-spam marked as spam). I've only been using it for about two weeks and it already has a nearly zero false positive rate (i.e. incorrectly flagged as spam) and a usefully low false negative rate (i.e. incorrectly flagged as legitimate).
Another important point is that there are some things that they can't hide, at least not in their current working model. If they're trying to sell you something, they have to describe what that thing is and where you can get it, and those descriptions are unlikely to be in any legitimate email. If they want to advertize a web site, they have to include its URL in the message, and the filter can catch that. If they advertize a physical address or phone number, the system can catch those, too. If they don't repeat the message, it means that there's inherently less spam, because I'm only seeing each add once.
It's also not possible to disguise everything in their headers, so things like their posting host (either the one they pay for legitimately or any open relay they're taking advantage of) will wind up being a pointer to who they are. They certainly can't change anything about the headers that's added downstream of their posting host, so as long as they keep using the same one it's likely that there will be characteristic stamps there that the spammers absolutely can't change. I know that analysis of the headers is part of bogofilter, another Bayesian filter that I've been using to good effect.
It might be more common than you think. In my work, for instance, we have occasional spurts where we generate a large amount of scientific data that needs to be processed followed by long periods where it doesn't. We're limited to running on the fastest box we can reasonably afford, but it might be cheaper and faster to buy just the clock cycles we need. One thing that's unclear is whether our processed data would be secure on our own machine or would be on IBM's farm, though. We'd definitely need to keep it on our own machine.
The other thing to consider is that it's possible that there are lots of applications where computer use might vary wildly from time to time, but nobody is thinking about them because they're uneconomical. Most places can't afford to have a supercomputer sitting around idle 95+% of the time, so instead they buy a machine that can process all of their data without the long idle times. The result is that there's a long lag between when the processing starts and when it finishes. If a system were available where they could buy the power to process that data rapidly when needed, though, it might make more sense to do it that way. They might very well wind up with about the same total cost but much faster results.
Because it's not their beach. Beaches in California (up to the mean high tide mark) are public property. They don't have the right to dump those boulders on public property, or in a way that would damage public property (at least without a permit). In California, they aren't legally allowed to deny beach access, either, again because the beach is public property.
I'm more worried by how badly out of date that website is. Anything that's trying to convince me to run Mandrake 6.1 needs a bit of updating.
Wrong. Because the cell phone owner pays for incoming calls, it's illegal to make an unsolicited call to a cell phone. The telemarketers know this and don't call cell phones. Of course as cell phone rates go down I expect for there to be heavy lobying against this rule, but as it stands right now there are legal obstacles to telemarketing to cell phones.
How much you save probably depends a lot on how you go about stripping out the code. I assume that there are also drivers for hardware that isn't available on x86 machines, so if you stripped out those you'd save some more space. There have to be some drivers specific to IBM mainframes, for instance, that are irrelevant to people using commodity hardware.
What would be really nice would be a system that would let you download a minimal configuration system, run you through the configuration, and then only download the files you need to compile the things you specified. If you run SCSI only, it wouldn't download the IDE drivers. If you run an x86, it wouldn't download the files specific to other processors. If you don't want to talk to Macs, it wouldn't download the HFS+ filesystem or appletalk drivers. I realize that adding a build system like that wouldn't be practical for the kernel developers, but it's nice to dream.
No. It's not bloated. Bloat is unnecessary size. The kernel isn't bloated because that size is necessary. The kernel has a huge number of features: it runs on a zillion different architectures, it can handle almost any hardware you throw at it, and it can handle almost any protocol and format that you might want your kernel to do. Those aren't things that are in there because some marketroid decided to put them in for buzzword compliance. They're things that are in there because actual users wanted them enough to write the code. You're not going to fit all of those available features into a small package, no matter how hard you try.
At the same time, the end product of a compiled kernel can be very compact because it doesn't have to include any features that the user doesn't want. The existence of code for dozens of architectures doesn't result in a bigger compiled kernel because those architectures aren't compiled in. Neither are drivers for hardware not present on your system, or protocols that you're not going to use. The net result is a kernel that's capable of doing whatever you want it to do without carrying the weight of the things that you don't want to do.
But the 32MB is for the compressed source, not for the binary. By the time you've compiled and compressed the binary image, it's typically down to about 5-700 KB, and I'd assume that if you have an embedded system with clearly defined minimal hardware you can probably get it a bit smaller than that. It's certainly not tiny, but the size of the end product (which is what really matters for the vast majority of people) is very reasonable.
Good enough that you have a quote mentioning it in your sig. That said, I'd easily add recommendations for Evangelion and Escaflowne. I'd almost recommend the original Macross, but even with AnimEigo's excellent restoration it hasn't aged that well- certainly don't watch the cheesy Robotech version.
Gundam is probably as close to legendary as any anime series can be. It radically altered the whole climate for SF anime. Perhaps it would be better to call it seminal or groundbreaking, but it was certainly very important in the development of the genre.
None. ICANN seems to have decided that having directors who were elected by the net population at large was interfering with their nice, cozy, corrupt way of doing things. Auerbach was only the most obvious example of this.
Legally, the question seems to be rather murky. There are some cases where it's clear that you have to obey more than one set of laws. If you're selling products to a country other than where you're located you still have to obey their laws, the same way that you'd have to if you sold their by mail or telephone order. If you're just providing information gratis, the answer seems to be more questionable. Certainly there are a lot of countries that claim that you have to obey their laws if you want to serve web pages to people there.
Practially, you may have to obey the laws of countries other than the one where your servers are located because they can make things very unpleasant for you if you don't. If you want to do business in country X, you have to listen to what they tell you, or they're likely to do nasty things like seize any assets you have in there or arrest your employees when they visit. That may not be a big problem for some countries- Google probably doesn't do much business with North Korea, and their employees probably don't visit there very often- but with places like the EU this is likely to be a problem.
Unfortunately, pressure from the outside is a consequence of Google's influence, and would likely wind up applying to any search engine that reaches Google's caliber. They're internationally significant, so any entity that wants to shut down opposing points of view is likely to target them. (Google may have some additional problems here because they cache the sites and make them available, making them a source of the information directly as well as indirectly.) Unfortunately, being globally significant also means that Google has to obey a lot of different laws in various jurisdictions. That means that they can be legally forced to de-list some sites- Holocaust denial sites in Germany, places selling Nazi memorabilia in France, sites that are alleged to violate copyright in the U.S., or even anything the government disapproves of in some more repressive countries. If you have a suggestion for how they can continue listing everything in spite of the legal restrictions on doing so, I'm sure that Google would be very interested in knowing about it.
Not quite. Polyphony is longer.
It is a vaguely interesting thing. A coworker once mentioned a word that she typed entirely with one hand and speculated about what the longest such word was. Being a geek, I thought that the obvious solution was to write a perl script to find out. I found that the longest words that could be typed entirely with one hand (the left) were 12 letters long; the only one that might actually be used in ordinary usage was stewardesses. ISTR that the longest word typed entirely with the right hand was only 9 letters, but I don't remember any examples.