My course in VLSI design was many, many years in the past, but what I do remember is that early integrated circuits used metal gates in the fabrication process. That process was later abandoned in favor of polysilicon because poly was much easier to work with at smaller feature sizes (I'm a bit foggy on this one). Gee, so now we're going back to metal gate processes, and we'll have real metal-oxide-semiconductor field effect transistors again?
If this is becoming easier to do at deep submicron level, I suppose processes for making deep submicron feature-sized Gallium-Arsenide MESFET's also got easier? Now wouldn't we just love to have such GaAs chips on our desktops... (I do know I'm forgetting another difficulty in working with GaAs, anyone care to remind me why GaAs is not as common as silicon today?)
Right, but none of the tactics they try using against Bayesian filtering can be provably shown to work. Concealing "high dollar words" with HTML entities or white on off white text are things that your corpus of normal ham mail doesn't have, but spam certainly does. These tricks have certainly not increased my false positive or false negative rates one jot, although I see them all the time in my corpus of spam, correctly classified by the Bayesian filter as having a 99-100% probability of being spam. These tricks only work against less-sophisticated fixed ruleset-based filters.
Blacklists are no easier and no harder to evade than Bayesian content based-filters in my experience, but they do suffer from the annoying characteristic of having a lot of false positives, especially if the blacklist you're using is not as responsible as it could be, or perhaps has an overzealous tendency to "vigilantism". They're totally worthless if your IP block is from a region or ISP that is (rightly or wrongly) considered to be spam-friendly.
But as for SMTP with decent authentication, that raises the bar sufficiently that most spammers will be forced to engage in totally illegal activity just to do their business. As of now, sending unsolicited commercial email is perfectly legal (or at least semi-legal) in most jurisdictions, but cracking peoples' computers and hijacking them for your own nefarious purposes is not.
How does heavy blocking of "spam friendly countries and ISP's" serve to deter more spam? I imagine that can only happen if such blocking becomes ubiquitous, and in the same way, if content-based Bayesian filters that fight back become equally ubiquitous, that would serve as an even stronger deterrent, without the same kind of collateral damage that accompanies blacklisting.
Frankly, the only serious long-term solution I can see for the problem of spam is to totally redesign SMTP to provide at the very least strong authentication of mail servers. Until then, IMHO content-based filtering is still a far better interim solution.
I've been using SpamAssassin's Bayesian filtering features to get rid of my spam for good. I've turned off SpamAssassin's use of any of the antispam sites like spamhaus, spews, and spamcop, mainly because some of them have been foolish enough to sweep such a wide net that turning on use of these sites causes SpamAssassin to filter legitimate mail that comes from my own domain! (that's what I get for living in a country whose ccTLD is run by a brain-damaged registrar...) I've been running almost totally on Bayesian filters after having trained them carefully for a month, and have thus far had zero false positives and false negatives. I mainly keep the spam around to further strengthen the training of my filters and for occasional entertainment value. Those Nigerian scams can be really funny sometimes, you know.:)
These blacklists could go away tomorrow and my Bayesian filters will only keep getting better and better at weeding out the spam. In my experience, these antispam sites are actually more part of the problem than the solution, because they filter more mail than they should.
I get your point, but what I don't get is why the frequency should be so low. Wouldn't it have to be unbelievably high? At the time of the big bang, the universe was a small fraction of the size of a proton, so any vibrations in the primordial soup at the time would have to have a wavelength even smaller than that, and hence a frequency with a value so great as to beggar the mind to even think about. If there was a sound wave whose wavelength was greater than the size of the universe at the time, wouldn't it eventually destructively interfere with itself, no matter what the topology of the universe as a whole was at the time?
But the early universe at the moments they're talking about was crammed into a space less than a quarter the size of a proton. Any vibrations in the primordial soup would have to have a wavelength even smaller than this, and hence a frequency whose value in Hertz would boggle the mind. If it had a wavelength bigger than the size of the universe at the time, then the "sound wave" would destructively interfere with itself.
By definition sound is pressure variations in air. Obviously when the big bang happened 15-20 billion years ago there was no air, there was nothing at all besides the expanding universe, which certainly was not made of air. The ultradense, unbelievably hot primordial soup could not have made any sound as it expanded.
As you've explained it, the public domain grants you every possible freedom, including the freedom to restrict the freedom of others from using your derivative works, as your example of TWITW eloquently illustrates. The GPL on the other hand, grants almost all these possible freedoms, with the sole exception of that ability to restrict others from use of derivative works. Saying that the public domain is more free than the GPL is like saying that a society that permits slavery is more free than one that does not because the former gives some members of its society the freedom to restrict the freedom of others, which is of course absolutely ridiculous.
The "framers" of the GPL, Richard M. Stallman and the Free Software Foundation, believe that the freedom to share and change software is an inalienable right, as they like to say, and the GPL is designed to guarantee that right. They believe that proprietary software is a form of slavery, and in one sense they are absolutely correct, as anyone forced to use Microsoft software knows all too well.
Athlon 64 was never speced for multiprocessor, in fact it will probably come out cheaper to build a dual processor Opteron board than a dual processor Athlon 64 board (which doesn't have a chance in hell of being in any other way better). AMD created the Opteron with three HyperTransport buses just for this purpose, and the Athlon 64 is handicapped with only one for the same reason. There are lots of Opteron multiprocessor boards out by now.
Gee, I wonder how many frequent flier miles Steve Ballmer's been getting lately? Earlier this month we've been hearing South Korea shifting from MS-based products to Free Software, and now the UK Government is considering expanding the use of Free Software. Heh.
A few years ago (in the wake of the Iridium fiasco) they already spun off part of their Semiconductor Products Sector (specifically the division that made discrete components, SSI glue logic, power electronics, and similar stuff) into the company that eventually became ON Semiconductor. Now the rest of SPS is following! At this rate, what will be left of their company?
In short, it doesn't. That has to be done by the filesystem layer or application layer itself. A Fibre Channel-based SAN setup such as is common in enterprise deployments doesn't normally provide for this either (Fibre Channel zoning is an exception, but that's a feature not normally required by most setups). If you really want to do concurrent filesystem access by multiple machines, you need a distributed lock manager of some kind, similar to what you have in Oracle RAC's cluster filesystem, Sistina's Global Filesystem, or OpenGFS.
Works that come out of a lathe, or microprocessor, or some universal Turing machine, like music, drawings, and works of text, can be no more patentable than the configuration of the lathe or Turing machine tape. They might be copyrightable, but patentable? They're expressions of ideas, not inventions, so they don't qualify, not even under the twisted US patent regime.
Too bad Limbo is too deeply tied to Inferno. It's unfortunate. From looking at how it's been designed it seems that Dennis Ritchie still hasn't lost his touch as an exceptional language designer. The Dis virtual machine that goes with it is supposedly designed to make a JIT simpler and more efficient, and well, according to Vita Nuova it gets something like 50% to 75% of the performance of compiled C/C++ code, which, if the article and Vita Nuova's benchmarks are accurate, totally blows C# and Java out of the water performance-wise. Now if only Vita Nuova would care to make it half as platform-neutral as Java... But hey, who cares, I'm already trying to do that.:)
In my experience the hardest part about setting Oracle 9i RAC (or even Oracle 8 parallel server) on Linux is the Fibre Channel arbitrated loop system that generally accompanies it. I've performed such a setup both on IBM FastT-type and Compaq/HP StorageWorks hardware, with both QLogic and Emulex HBA cards, and that's where all the dancing around happens. Driver support on Linux (especially for HP) is spotty, and not exactly straightforward to come by. However, once the Fibre setup works perfectly, everything else follows quickly enough.
And these things are godawful expensive as well. The fabric switch alone usually costs more than the entire cluster of computers which are connected to it (and these are by no means cheap machines... they're IBM xSeries or Compaq/HP DL-type enterprise servers), as does the actual disk array, even if it's as small as 200 GB. The entire shared storage setup can come up as the largest single cost in an Oracle RAC or other proprietary enterprise database installation that makes use of a similar system.
However, I imagine that none of these open source database clustering solutions can hold a candle to RAC in terms of performance. They all seem to make use of replication, which means that they will be far, far cheaper than a shared storage setup, but that is where a fibre channel disk array is difficult to beat. Fibre Channel has data transfer rates measured in GB/sec... RAC has a dedicated cluster filesystem or makes use of raw devices, with its own distributed lock manager that prevents the machines in the cluster from simultaneously writing to the storage and corrupting it. As I imagine, such a system is difficult to write and test in an open source context, as it again requires any of the horrendously expensive enterprise shared storage setups on the market (even multi-tailed SCSI disk arrays, while cheaper than fibre, are still quite pricey nevertheless).
GSM crypto was always suspected to be weak
on
Cracking GSM
·
· Score: 2, Interesting
From what I remember, the design of the GSM A5 cipher was always suspected to be weak. From Applied Cryptography:
A lot of strange politics surrounds [A5]. Originally it was thought that GSM's cryptography would prohibit export of the phones to some countries. Now some officials are discussing whether A5 might harm export sales, implying that it is so weak as to be an embarrasment. Rumor has it that the various NATO intelligence agencies had a catfight in the mid-1980's over whether GSM encryption should be strong or weak. The Germans wanted strong cryptography, as they were sitting near the Soviet Union. The other countries overruled them, and A5 is a French design. [emphasis mine]
Bruce Schneier then goes on to say that "There is a trivial attack requiring 240 encryptions." 240 is only some 1 trillion, definitely in reach using today's computers.
Yeah, the NSA has already been doing it, you can be sure of that, and further rumors about GSM crypto that I've been hearing say that the NSA applied pressure on the French as well to insert deliberate weaknesses. Maybe Biham & Co. just managed to find out some of the NSA's "easter eggs".
Well, I suppose you know about the red shift due to the general expansion of the universe? The most distant objects in the universe are now receeding away from us at such a massive rate that the visible light they emitted has been so far red shifted as to wind up in the infrared region. There's a Doppler effect for light that causes light from an object moving very quickly away from an observer to reach the observer at a lower frequency than what was transmitted (the red shift), just like a car moving away from you makes sounds at a lower pitch than were it standing still or moving towards you. Because of Hubble's law, the farther away an object is, the faster it's moving away from us, and consequently, the greater the Doppler effect. This infrared probe is designed to view objects that have been so far "red shifted" as to apparently be emitting infrared radiation.
typical question for an american!
on
Flaming Cellphones
·
· Score: 2, Interesting
...So, if she bought her phone from a carrier store, and they gave her the battery, then would that carrier assume the liability for this happening, since it was not a Nokia battery involved?
First thing that enters people's minds out there... Liability! As they say, America is a nation of lawyers and order. Fortunately, the woman's Dutch, and the fact that the burns were superficial should ensure that nothing of that sort is going to happen. She'll probably just buy a new phone and get on with her life. Only buying original accessories for her new phone, I trust.
Word's equation editor can't hold a candle to the way (La)TeX can format complicated equations. The equations typeset by Word using that system are not only plagued by a clumsy interface (one of the few instances where I find that WYSIWYG is a loser), they also look plain amateurish when compared to that created by LaTeX.
Certainly, there's a learning curve, but after a few days of practicing it becomes as natural as reading a formula. Most of the authors of articles for peer-reviewed journals use it to typeset them. Many of the newer technical books I own also mention that they were typeset using TeX or one of its many macro packages. TeX and LaTeX are the way to go for professional mathematical and scientific typesetting. I haven't seen anything else that comes even remotely close.
The trouble is, the distribution points for these phones are just like the distribution points for new PC's with Windows. The smartphones aren't generally distributed the way most normal handsets like those manufactured by Nokia, Ericsson, and the other big-name brands are. From what I know, these phones are generally sold by a telco as part of their subscriber plans. From what I've heard, this is what Orange is doing, and out here in my country Smart Communications is also distributing these phones along with subscriber plans, even low-level subscriber plans. From what I remember, Smart gives away such an "Amazing Phone" (as they call it) for a plan that amounts to the equivalent of US$16 per month.
If more telcos adopt this strategy, I imagine that Microsoft could actually manage to drive all of the other cellular phone manufacturers out of business in the same way they are able to kill off competing operating systems on the desktop by preinstalling Windows on every new PC. That's the scary part.
I don't see it as hypocrisy on their part. The FSF, contrary to many peoples' perceptions of them here on/. are more pragmatic than most people here give them credit for. In the case of Ogg Vorbis, the problem was the MP3 format is as deeply entrenched as the GIF format, and just as encumbered by patents. By using a more permissive license like The modified BSD-style license rather than the GPL, they would be encouraging proprietary software to use the Ogg Vorbis format, so that the patent-unencumbered Ogg Vorbis format could more quickly gain an edge over MP3 than were it GPLed.
The FSF and RMS have always understood that software patents are a graver threat to Free Software than proprietary software. If some proprietary software company like Microsoft decides not to support use of Ogg Vorbis because the code is GPL and thus cannot be taken proprietary, that would serve to further entrench the MP3 format, which is even more counterproductive to the cause of Free Software, as it is a format that cannot be used by Free Software. RMS justifies this strategy in another post.
It's not hypocrisy, but pragmatism that drives these decisions.
I have to correct myself a bit here. The Ogg Vorbis toolkit was originally licensed under the GPL, from what I remember, and they later shifted to a BSD-style license, which move was not begrudgingly accepted by RMS and the rest of the Free Software Foundation. They actively encouraged the move, IIRC, as Ogg Vorbis is a technologically superior format unencumbered by patents, unlike the dominant MP3 format, for which a legal codec would be impossible for Free Software (LAME and Bladeenc are legally a gray area, and that isn't a good thing). Think GIF vs. PNG. RMS and the FSF have always understood that software patents pose an even greater threat to the cause of Free Software than proprietary software does. The GPL is designed to protect against software from going proprietary, but is of no help at all when dealing with patents (for which there can be no effective legal defense, short of having your own cross-licensable patent pool or having software patents abolished totally, which the FSF and the LPF are actively working to do).
Care to give a link that shows that RMS and the Free Software Foundation did not fully endorse Xiph's decision to move the licensing from GPL to BSD-style? Another link I've found, again RMS's own words, shows more pragmatism than anything. For reference, here's the original link from which i got the first RMS quote.
You are right of course that yes, rare are the cases where another license would serve the cause of Free Software better than the GPL would, but these cases are not unknown. For another example, someone else points out that the FSF actually discourages people from GPLing components at the core of the X Window System. The FSF as a whole and even Stallman in particular are not as inflexible and unpragmatic as many here seem to think.
Slashdot Mirror
My course in VLSI design was many, many years in the past, but what I do remember is that early integrated circuits used metal gates in the fabrication process. That process was later abandoned in favor of polysilicon because poly was much easier to work with at smaller feature sizes (I'm a bit foggy on this one). Gee, so now we're going back to metal gate processes, and we'll have real metal-oxide-semiconductor field effect transistors again?
If this is becoming easier to do at deep submicron level, I suppose processes for making deep submicron feature-sized Gallium-Arsenide MESFET's also got easier? Now wouldn't we just love to have such GaAs chips on our desktops... (I do know I'm forgetting another difficulty in working with GaAs, anyone care to remind me why GaAs is not as common as silicon today?)
Right, but none of the tactics they try using against Bayesian filtering can be provably shown to work. Concealing "high dollar words" with HTML entities or white on off white text are things that your corpus of normal ham mail doesn't have, but spam certainly does. These tricks have certainly not increased my false positive or false negative rates one jot, although I see them all the time in my corpus of spam, correctly classified by the Bayesian filter as having a 99-100% probability of being spam. These tricks only work against less-sophisticated fixed ruleset-based filters.
Blacklists are no easier and no harder to evade than Bayesian content based-filters in my experience, but they do suffer from the annoying characteristic of having a lot of false positives, especially if the blacklist you're using is not as responsible as it could be, or perhaps has an overzealous tendency to "vigilantism". They're totally worthless if your IP block is from a region or ISP that is (rightly or wrongly) considered to be spam-friendly.
But as for SMTP with decent authentication, that raises the bar sufficiently that most spammers will be forced to engage in totally illegal activity just to do their business. As of now, sending unsolicited commercial email is perfectly legal (or at least semi-legal) in most jurisdictions, but cracking peoples' computers and hijacking them for your own nefarious purposes is not.
How does heavy blocking of "spam friendly countries and ISP's" serve to deter more spam? I imagine that can only happen if such blocking becomes ubiquitous, and in the same way, if content-based Bayesian filters that fight back become equally ubiquitous, that would serve as an even stronger deterrent, without the same kind of collateral damage that accompanies blacklisting.
Frankly, the only serious long-term solution I can see for the problem of spam is to totally redesign SMTP to provide at the very least strong authentication of mail servers. Until then, IMHO content-based filtering is still a far better interim solution.
I've been using SpamAssassin's Bayesian filtering features to get rid of my spam for good. I've turned off SpamAssassin's use of any of the antispam sites like spamhaus, spews, and spamcop, mainly because some of them have been foolish enough to sweep such a wide net that turning on use of these sites causes SpamAssassin to filter legitimate mail that comes from my own domain! (that's what I get for living in a country whose ccTLD is run by a brain-damaged registrar...) I've been running almost totally on Bayesian filters after having trained them carefully for a month, and have thus far had zero false positives and false negatives. I mainly keep the spam around to further strengthen the training of my filters and for occasional entertainment value. Those Nigerian scams can be really funny sometimes, you know. :)
These blacklists could go away tomorrow and my Bayesian filters will only keep getting better and better at weeding out the spam. In my experience, these antispam sites are actually more part of the problem than the solution, because they filter more mail than they should.
I get your point, but what I don't get is why the frequency should be so low. Wouldn't it have to be unbelievably high? At the time of the big bang, the universe was a small fraction of the size of a proton, so any vibrations in the primordial soup at the time would have to have a wavelength even smaller than that, and hence a frequency with a value so great as to beggar the mind to even think about. If there was a sound wave whose wavelength was greater than the size of the universe at the time, wouldn't it eventually destructively interfere with itself, no matter what the topology of the universe as a whole was at the time?
But the early universe at the moments they're talking about was crammed into a space less than a quarter the size of a proton. Any vibrations in the primordial soup would have to have a wavelength even smaller than this, and hence a frequency whose value in Hertz would boggle the mind. If it had a wavelength bigger than the size of the universe at the time, then the "sound wave" would destructively interfere with itself.
By definition sound is pressure variations in air. Obviously when the big bang happened 15-20 billion years ago there was no air, there was nothing at all besides the expanding universe, which certainly was not made of air. The ultradense, unbelievably hot primordial soup could not have made any sound as it expanded.
Our AC is for once absolutely correct. As that line from "Once Upon a Time in Mexico" goes: "The best way to win is to rig the game."
As you've explained it, the public domain grants you every possible freedom, including the freedom to restrict the freedom of others from using your derivative works, as your example of TWITW eloquently illustrates. The GPL on the other hand, grants almost all these possible freedoms, with the sole exception of that ability to restrict others from use of derivative works. Saying that the public domain is more free than the GPL is like saying that a society that permits slavery is more free than one that does not because the former gives some members of its society the freedom to restrict the freedom of others, which is of course absolutely ridiculous.
The "framers" of the GPL, Richard M. Stallman and the Free Software Foundation, believe that the freedom to share and change software is an inalienable right, as they like to say, and the GPL is designed to guarantee that right. They believe that proprietary software is a form of slavery, and in one sense they are absolutely correct, as anyone forced to use Microsoft software knows all too well.
Athlon 64 was never speced for multiprocessor, in fact it will probably come out cheaper to build a dual processor Opteron board than a dual processor Athlon 64 board (which doesn't have a chance in hell of being in any other way better). AMD created the Opteron with three HyperTransport buses just for this purpose, and the Athlon 64 is handicapped with only one for the same reason. There are lots of Opteron multiprocessor boards out by now.
If someone doesn't want to use GPLed software, that's their loss.
Gee, I wonder how many frequent flier miles Steve Ballmer's been getting lately? Earlier this month we've been hearing South Korea shifting from MS-based products to Free Software, and now the UK Government is considering expanding the use of Free Software. Heh.
A few years ago (in the wake of the Iridium fiasco) they already spun off part of their Semiconductor Products Sector (specifically the division that made discrete components, SSI glue logic, power electronics, and similar stuff) into the company that eventually became ON Semiconductor. Now the rest of SPS is following! At this rate, what will be left of their company?
In short, it doesn't. That has to be done by the filesystem layer or application layer itself. A Fibre Channel-based SAN setup such as is common in enterprise deployments doesn't normally provide for this either (Fibre Channel zoning is an exception, but that's a feature not normally required by most setups). If you really want to do concurrent filesystem access by multiple machines, you need a distributed lock manager of some kind, similar to what you have in Oracle RAC's cluster filesystem, Sistina's Global Filesystem, or OpenGFS.
Works that come out of a lathe, or microprocessor, or some universal Turing machine, like music, drawings, and works of text, can be no more patentable than the configuration of the lathe or Turing machine tape. They might be copyrightable, but patentable? They're expressions of ideas, not inventions, so they don't qualify, not even under the twisted US patent regime.
Too bad Limbo is too deeply tied to Inferno. It's unfortunate. From looking at how it's been designed it seems that Dennis Ritchie still hasn't lost his touch as an exceptional language designer. The Dis virtual machine that goes with it is supposedly designed to make a JIT simpler and more efficient, and well, according to Vita Nuova it gets something like 50% to 75% of the performance of compiled C/C++ code, which, if the article and Vita Nuova's benchmarks are accurate, totally blows C# and Java out of the water performance-wise. Now if only Vita Nuova would care to make it half as platform-neutral as Java... But hey, who cares, I'm already trying to do that. :)
In my experience the hardest part about setting Oracle 9i RAC (or even Oracle 8 parallel server) on Linux is the Fibre Channel arbitrated loop system that generally accompanies it. I've performed such a setup both on IBM FastT-type and Compaq/HP StorageWorks hardware, with both QLogic and Emulex HBA cards, and that's where all the dancing around happens. Driver support on Linux (especially for HP) is spotty, and not exactly straightforward to come by. However, once the Fibre setup works perfectly, everything else follows quickly enough.
And these things are godawful expensive as well. The fabric switch alone usually costs more than the entire cluster of computers which are connected to it (and these are by no means cheap machines... they're IBM xSeries or Compaq/HP DL-type enterprise servers), as does the actual disk array, even if it's as small as 200 GB. The entire shared storage setup can come up as the largest single cost in an Oracle RAC or other proprietary enterprise database installation that makes use of a similar system.
However, I imagine that none of these open source database clustering solutions can hold a candle to RAC in terms of performance. They all seem to make use of replication, which means that they will be far, far cheaper than a shared storage setup, but that is where a fibre channel disk array is difficult to beat. Fibre Channel has data transfer rates measured in GB/sec... RAC has a dedicated cluster filesystem or makes use of raw devices, with its own distributed lock manager that prevents the machines in the cluster from simultaneously writing to the storage and corrupting it. As I imagine, such a system is difficult to write and test in an open source context, as it again requires any of the horrendously expensive enterprise shared storage setups on the market (even multi-tailed SCSI disk arrays, while cheaper than fibre, are still quite pricey nevertheless).
From what I remember, the design of the GSM A5 cipher was always suspected to be weak. From Applied Cryptography:
Bruce Schneier then goes on to say that "There is a trivial attack requiring 240 encryptions." 240 is only some 1 trillion, definitely in reach using today's computers.
Yeah, the NSA has already been doing it, you can be sure of that, and further rumors about GSM crypto that I've been hearing say that the NSA applied pressure on the French as well to insert deliberate weaknesses. Maybe Biham & Co. just managed to find out some of the NSA's "easter eggs".
Gee, Tufts University? Now THAT is a truly appropriate place for them to have made such a breakthrough discovery in the science of spider silk. ;)
Well, I suppose you know about the red shift due to the general expansion of the universe? The most distant objects in the universe are now receeding away from us at such a massive rate that the visible light they emitted has been so far red shifted as to wind up in the infrared region. There's a Doppler effect for light that causes light from an object moving very quickly away from an observer to reach the observer at a lower frequency than what was transmitted (the red shift), just like a car moving away from you makes sounds at a lower pitch than were it standing still or moving towards you. Because of Hubble's law, the farther away an object is, the faster it's moving away from us, and consequently, the greater the Doppler effect. This infrared probe is designed to view objects that have been so far "red shifted" as to apparently be emitting infrared radiation.
First thing that enters people's minds out there... Liability! As they say, America is a nation of lawyers and order. Fortunately, the woman's Dutch, and the fact that the burns were superficial should ensure that nothing of that sort is going to happen. She'll probably just buy a new phone and get on with her life. Only buying original accessories for her new phone, I trust.
Word's equation editor can't hold a candle to the way (La)TeX can format complicated equations. The equations typeset by Word using that system are not only plagued by a clumsy interface (one of the few instances where I find that WYSIWYG is a loser), they also look plain amateurish when compared to that created by LaTeX.
Certainly, there's a learning curve, but after a few days of practicing it becomes as natural as reading a formula. Most of the authors of articles for peer-reviewed journals use it to typeset them. Many of the newer technical books I own also mention that they were typeset using TeX or one of its many macro packages. TeX and LaTeX are the way to go for professional mathematical and scientific typesetting. I haven't seen anything else that comes even remotely close.
The trouble is, the distribution points for these phones are just like the distribution points for new PC's with Windows. The smartphones aren't generally distributed the way most normal handsets like those manufactured by Nokia, Ericsson, and the other big-name brands are. From what I know, these phones are generally sold by a telco as part of their subscriber plans. From what I've heard, this is what Orange is doing, and out here in my country Smart Communications is also distributing these phones along with subscriber plans, even low-level subscriber plans. From what I remember, Smart gives away such an "Amazing Phone" (as they call it) for a plan that amounts to the equivalent of US$16 per month.
If more telcos adopt this strategy, I imagine that Microsoft could actually manage to drive all of the other cellular phone manufacturers out of business in the same way they are able to kill off competing operating systems on the desktop by preinstalling Windows on every new PC. That's the scary part.
I don't see it as hypocrisy on their part. The FSF, contrary to many peoples' perceptions of them here on /. are more pragmatic than most people here give them credit for. In the case of Ogg Vorbis, the problem was the MP3 format is as deeply entrenched as the GIF format, and just as encumbered by patents. By using a more permissive license like The modified BSD-style license rather than the GPL, they would be encouraging proprietary software to use the Ogg Vorbis format, so that the patent-unencumbered Ogg Vorbis format could more quickly gain an edge over MP3 than were it GPLed.
The FSF and RMS have always understood that software patents are a graver threat to Free Software than proprietary software. If some proprietary software company like Microsoft decides not to support use of Ogg Vorbis because the code is GPL and thus cannot be taken proprietary, that would serve to further entrench the MP3 format, which is even more counterproductive to the cause of Free Software, as it is a format that cannot be used by Free Software. RMS justifies this strategy in another post.
It's not hypocrisy, but pragmatism that drives these decisions.
I have to correct myself a bit here. The Ogg Vorbis toolkit was originally licensed under the GPL, from what I remember, and they later shifted to a BSD-style license, which move was not begrudgingly accepted by RMS and the rest of the Free Software Foundation. They actively encouraged the move, IIRC, as Ogg Vorbis is a technologically superior format unencumbered by patents, unlike the dominant MP3 format, for which a legal codec would be impossible for Free Software (LAME and Bladeenc are legally a gray area, and that isn't a good thing). Think GIF vs. PNG. RMS and the FSF have always understood that software patents pose an even greater threat to the cause of Free Software than proprietary software does. The GPL is designed to protect against software from going proprietary, but is of no help at all when dealing with patents (for which there can be no effective legal defense, short of having your own cross-licensable patent pool or having software patents abolished totally, which the FSF and the LPF are actively working to do).
Care to give a link that shows that RMS and the Free Software Foundation did not fully endorse Xiph's decision to move the licensing from GPL to BSD-style? Another link I've found, again RMS's own words, shows more pragmatism than anything. For reference, here's the original link from which i got the first RMS quote.
You are right of course that yes, rare are the cases where another license would serve the cause of Free Software better than the GPL would, but these cases are not unknown. For another example, someone else points out that the FSF actually discourages people from GPLing components at the core of the X Window System. The FSF as a whole and even Stallman in particular are not as inflexible and unpragmatic as many here seem to think.