Take an economics course. Government mandates HURT ECONOMIES. There is no exception to this rule.
I have taken an economics course. There are a number of very important exceptions to this rule. If you really have taken an economics course, and learned something from it, you would remember these exceptions.
One important exception is monopoly power. Another one is externalities. A third exception is asymmetric information. A fourth exception is public goods. All of these exceptions are relevant in one way or another to the auto market. I will leave it as an exercise to figure out why each of these classifications applies. If you really took an economics course, it should be very easy.
I do agree that very many develop Linux for "themselves", but I don't think geeks are more interested in struggling with network or sound or wifi or video tearing or printer drivers or webcams than other people. Nor are they particularly tolerant of buggy or inflexible software, even though they may have their own ideas of UI design.
The difference is subtle but it definitely exists. Here's one example that comes to mind. In GNOME or KDE, it's really easy to remove the main menu from the panel - just right click the thing and select "Remove". In Windows or Mac OS, however, it is nigh-on impossible to remove the Start/Apple menu from the taskbar/menubar. You can temporarily hide it, but it can't be removed entirely unless you start writing code. Geeks much prefer the flexibility of Linux, whereas most regular users are better off not having this flexibility, since a missing main menu is not a situation they can easily recover from.
Obviously there'll always be distributions for geeks by geeks, but I have no problem imagining distributions for joe average by companies that want to sell services to the residential or business market. The gap, the changes they'd have to make aren't that big compared to everything they could reuse.
There is a market here, and the required changes are indeed individually very small. (For example, the behavior I mentioned above would be easy to correct.) However, the changes are also numerous, and there are all sorts of rare cases where one can easily fail to notice a needed change. In order to do a good job, you need a large team -- not as large as the Windows or Mac OS teams, but still a good sized team. And, since few geeks are interested in volunteering for such a task, you're back to the situation where you need a revenue source in order to support that team.
It is instructive to look at the netbook market, which of course is the kind of Linux deployment that we're talking about -- selling Linux to Joe average. Netbooks may not be as profitable as laptops, but they bring in more than enough revenue to support a dedicated development team. The EeePC, for example, has a very inflexible desktop by any geek's standards: you can't even move or alter the icons, and forget about configuring the main menu, since there isn't any. In other words, it's just what grandma needs. But even this approach has its limits. One big problem is that it's hard to keep a modified platform in sync with the main trunk. Every time the upstream developers publish a security update, you have to integrate that update with the changes that you have made. This process requires continuous engineering effort far beyond the short product life of a particular piece of hardware. For a major distribution like Debian, there are ample volunteers to handle the task of integrating security patches, even for older versions of Debian. In the case of the EeePC, especially the older models, more often than not this work is not done.
To be fair, things are not entirely bleak. Ubuntu Netbook Remix is an example of a non-geek desktop interface which is integrated into a major distribution. (I'm not sure how much of its development was funded by Canonical vs. contributed by volunteers.) I can see such projects offering hope for Linux on the desktop, but it will never be a natural fit. You'll always need some effort in order to cajole Linux developers into contributing to mass market products. Even worse, such efforts have to be continuous and ongoing in order to best take advantage of traditional Linux strengths such as security.
I'm sorry, but this is just really, really stupid. The crux of your post is that Linux is doing better than it is, but nobody knows because Linux companies are so goshdarn humble. Do you actually believe that? And you wrote it on a site that proclaims every year to be the year of Linux on the desktop?
The crux of my post is that "doing better" does not mean the same thing for Linux as it does for Windows or Apple.
You speak of the mythical Linux on the desktop. I honestly don't think Linux will ever achieve success in the desktop market. I'm not saying this out of elitism -- as far as I can tell, it is cold objective fact. Most Linux developers are not traditional desktop users. By this I mean that even though Linux developers do use the desktop, they are not representative of the average desktop user. The Linux desktop will always be optimized for the average Linux developer rather than the average desktop user. Obviously, the ideal desktop for an average Linux developer is very different from the ideal desktop for an average user. That's why Linux will have a hard time in the desktop market -- because it's written by, and targeted to, developers, not traditional desktop users.
At the same time, Linux does not need to succeed on the desktop in order to be relevant. The vast majority of Linux developers will continue to contribute to Linux even if it fails in the wider desktop market. (Note that this is where Linux differs significantly from Windows -- if Windows fails in the consumer desktop market, it's game over.) The fact that you continue to use desktop penetration as your measure of success shows that you are still trying to judge Linux using Windows's standards. Such reasoning is faulty, and even more so now that large segments of the computing market (i.e. netbooks) are breaking away from the traditional desktop paradigm.
I didn't think Linux would survive. I thought that Linux and its users would be slaughtered. It has, and it has grown and improved. I am honestly astounded at the level of advancement that Linux has made
Any proprietary platform with the kind of market share that Linux has would be dead by now. It happened to OS/2, it happened to the various proprietary Unices, and many others (BeOS etc.). Hell, even Apple would have died in the 90s had it not been saved by, ironically, Microsoft.
The difference is that a free platform can advance even with a very small userbase, as long as enough skilled volunteers are willing to contribute. Linux probably represents more total man-years of work than Windows, even though its market share is much smaller. That's because Linux can easily recruit volunteer developers for free, whereas Microsoft has to pay huge salaries to attract top talent.
It's a mistake to judge Linux by the same standards as other commercial vendors. Free software really changes the rules of the game, and that's why Microsoft is on its heels.
Perhaps after the success of the switch adds Apple inched ahead of Linux on the desk top. But if you look over the last 15 years, I believe that there has been more Linux on the desktop than Apple OS's.
It hasn't been in anyone's interest to say that. I think that is even true of the Linux companies. For a long time they wanted to be under the radar under dogs. Perhaps because they didn't want a fight to the death with Microsoft.
This is an excellent observation and it lies at the heart of Linux's success. Anyone who fails to grasp this point is not even in the right ballgame.
In terms of economics, the most important difference between free and proprietary platforms is that free platforms do not require a large userbase in order to thrive. The number of skilled developers willing to work on Linux for free would remain very high even if Linux's userbase were to drastically shrink. (This is incidentally the main reason why Microsoft cannot win against Linux, at least not by any means available in the marketplace.)
By contrast, a proprietary platform requires a large userbase in order to even survive, since the only way for a proprietary platform to get developers is to pay them a salary, and salaries require money, which requires users. There are no volunteer developers who are even able, let alone willing, to contribute code to Microsoft Windows for free, because of the locked-in nature of the platform.
That's why proprietary vendors routinely inflate their usage numbers. Larger numbers are necessary in order to convince new users that the platform is worthy of adoption. If a proprietary platform does not have a lot of users, then it has no future, and rational users would not risk selecting that platform. That's why OS/2 died, and that's why Solaris (despite being made free recently) is about to die.
We thus have a situation where every vendor, other than the Linux vendors, has a huge economic incentive to inflate the reported size of their userbase. I don't necessarily mean illegal activity here; there are well known legal methods by which usage numbers can be inflated. For example, Microsoft counts every Vista OEM license as a Vista sale even if the user exercises downgrade rights to XP, and so on. In any case, it doesn't surprise me at all that the actual market share of Linux is far higher than what is being reported.
More like several thousand if you burn up all the uranium and the thorium we can reasonably extract and don't waste any of it.
It is actually billions of years, not thousands or even millions. The reason is that, unlike oil, the raw material for nuclear power (uranium or thorium) is so energy-dense that we can spend extraordinary efforts on extraction and still come out ahead, both financially and energetically. Thorium, in particular, is about as common as lead: there is a LOT of it in the ground. See Cohen for the calculation that produces the billions of years figure.
How many journals have you read? This seems to be par for the course, as far as my experience goes.
I am a research mathematician. Part of my job is to read journals. Admittedly, these are mostly math and computer science journals, which you might not count as science. If you count only journals which are strictly (non-computer) science, then I read Nature and Science, but very rarely any others, and this level of bias is definitely not typical among what I read.
So basically, to make Nuclear just fall off his chart, he assumes that building more powerplants will lead to nuclear war, and calculates how much stuff that will burn. Is that not completely absurd?
Yes, it is completely absurd.
Nuclear weapons have been used exactly two times in all of human history. Both times occurred in 1945, before the invention of nuclear power plants. There is absolutely no scientific or empirical evidence linking nuclear power plants to nuclear war.
Nuclear power production is responsible for exactly two civilian deaths worldwide in the last 22 years. It is by far the safest energy source available under current technology.
The authors rate nuclear energy negatively because it takes 10-19 years to build a nuclear power plant. The vast majority of those 10-19 years is consumed in regulatory oversight, not technical construction. I wouldn't be surprised if these same authors were responsible for lobbying in favor of said regulations.
The authors cite waste disposal as a disadvantage of nuclear energy. They fail to mention that uranium reprocessing dramatically reduces the quantity of nuclear waste. They also fail to mention that, with reprocessing, there is enough nuclear fuel to last for billions of years, i.e. longer than the sun will shine. They do link reprocessing to nuclear war, a notion which I thoroughly debunked in two sentences above. This treatment strikes me as incredibly biased and ignorant of scientific reality. While all of their points about regulatory and legal obstacles are factually true, these points belong in a policy paper or a legal draft. A scientific paper ought to focus on technical and scientific matters rather than legal issues. Again, I wouldn't be surprised if these same authors were responsible for lobbying in favor of the existing laws which ban reprocessing.
All in all, this paper does not belong in any sort of science journal. It is that bad.
Don't you think that a woman who loves and respects her husband would WANT to at least find out how he votes and why?
That's the same mentality as "If I'm not breaking the law, then what have I got to hide?" However reasonable the loss of secrecy is in this case, it is the first step in a slippery slope.
There is a very big difference between permitting spouses to reveal their votes to each other, vs. requiring it. As much as you might like to see the world through rose-tinted glasses, the reality is that a large number of households are not as harmonious as yours.
So how is such a vote, cast at home and sent in by mail not respect the voter's secrecy?
This is a very common question, and the very fact that you asked it is an excellent illustration of my point -- people in general do not realize that absentee ballots threaten the secret ballot system. The whole point of a step-by-step erosion is that each step by itself seems unthreatening.
The answer is: it is not possible to keep an absentee vote secret from other members in your household, especially if there is a disparity in power and voting tendencies within the household.
The usual example cited is that of Mormon women, which comprise a huge fraction of the electorate in Utah. If you mandate absentee ballots, it amounts to requiring that ballots be filled out in the home, under the eyes of a watchful husband. (In theory, ballots can be filled out in places other than your home, but in practice the only common case is that absentee ballots are filled out at home.) Under such a scenario, women are much more likely to vote in accordance with their husband's wishes than they otherwise would under a true secret ballot system.
Of course, coercion is also possible in a voting booth, but it is harder (e.g. voters can vote one way and say they voted the other way). Regardless of whatever flaws a voting booth may have, it is certainly at least as secret as voting at home, and very possibly more secret.
in the united states, indeed in every western democracy, ballots are secret. no one questions this anonymity -- indeed, it's mandated by law.
I totally agree with the need for secret ballots in general elections (even though I'm not posting anonymously).
I would like to point out in addition that absentee voting is becoming increasingly popular in the US. An absentee ballot is fundamentally incompatible with the mandate that ballots be secret, because there is no way (no constitutional way, at least) to enforce secrecy of ballots within the privacy of an individual's own home.
If we ever as a society lose our right to secret ballots, I suspect it will happen gradually, one step at a time, with mandatory absentee voting (already implemented in Oregon, and due by 2010 in Washington) being the first step.
I don't quite follow your reasoning. In order to convince you that some proof languages are readable I would need to be able to write formal proofs efficiently?... As I mentioned, languages that are designed to make writing the proofs easier tend to be less readable.
Readability and writeability are separate and diametrically opposed issues. Writeability is necessary in order to speed up the development of new mathematics. Readability is necessary in order to improve the general public's understanding of and accessibility to the project. We apparently both agree that readability and writeability are conflicting goals. I am of the opinion that machine readable proofs are currently inadequate on both counts, i.e. machine readable proofs are not useful for developing new math, and they are also not useful for teaching mathematics to the general public. You on the other hand seem to be convinced that machine readable proofs are adequate on both counts, despite the conflicting nature of these goals.
I hope that the above accurately summarizes our positions. If not, feel free to correct any mistakes that I have made.
How many formal proof languages have you seen? Most of them are designed for software verification, not formalized mathematics. They are optimized to make it easier for the author, not for the reader.
Indeed, write-only code is a key aspect of perl which is shared by every formal proof language.
Vdash will (probably) use Isar which is as easy to read for a mathematician as a standard proof.
In order for this claim to be true to the extent that it would affect my assertion about advancing (or not advancing) the development of new mathematics, one would at a minimum need to be able to efficiently write existing proofs in the formal language (surely one cannot hope to accelerate the discovery of new proofs if even the existing literature cannot be efficiently handled). That means formal proofs for things like Fermat's Last Theorem, Poincare Conjecture, and so on, which I honestly don't ever expect to see in my lifetime.
If they'd just make it a bit more accessible for the general public, and provide more examples of what they are doing, instead of abstracting everything and assuming everyone knew as much as them (or rather what symbols they are using to represent what they know), then maybe mathematics would be further advanced.
Disclosure: I am a mathematician. I also graduated from Harvard, and I know Cam Freer personally.
You make several interesting but separate points in your post. Let me respond to them one at a time. Regarding pay-to-view websites, I can assure you that many mathematicians are just as unhappy as you are with the lack of public access to publications. This issue is larger than just mathematics (see for example musicians vs. RIAA, which is essentially same issue), and to pin the blame on mathematicians is unfair. This is a problem that society as a whole needs to solve in all its incarnations.
About making mathematics more accessible to the public, I agree that this step is necessary for ethical and practical reasons, and it's one of the reasons why I left a pure research position (with no teaching responsibilities) in exchange for a research position having some teaching responsibilities. However, I really doubt that public education would result in mathematics being further advanced. The cutting edge of new mathematics today is so far beyond the pale of the public that even a herculean effort at education won't be enough to bring very many people into the fold. There is a saying that the more you know, the more you know you don't know, and having been through a Ph.D program I can assure you that what you perceive as elitism is nothing of the sort -- the gap between current research and the general public is simply too wide for even the best educational efforts to bridge.
In fact, it's not even clear to me that increasing the rate of advancement of mathematics is the right goal. Perhaps we should even be holding back the development of new mathematics to give the public a chance to catch up with what we've done.
The project being discussed won't do any of these things. It won't make mathematics more accessible; anyone who's ever seen a machine-verifiable proof knows that they're even more obfuscated than perl code. It won't affect the overall rate of advancement in mathematics, because writing machine-verifiable proofs is much slower than writing standard mathematical proofs. It will of course increase our knowledge in the area of machine-verified proofs, but it's hard to argue that this area of mathematics is any more important than any other area like number theory. If you want to target the general public, the best way is still via sites like Wikipedia or Planetmath.org where the focus is on human-readable presentation.
The dumb part here is that they send the whole movie to your computer even if you're just watching the free two-minute preview. The two-minute restriction is only enforced in the flash applet. Now, no amount of DRM can stop a paying customer from copying the movie, but a smartly designed system could certainly make the customer pay for the movie before giving the whole movie to them.
it needs to provide the user feedback that the claim implicit in the URL is not supported.
Blocking the site completely (and yes, the firefox behavior amounts to a total block) is a far cry from simply providing user feedback. It's throwing out the baby with the bathwater.
Only a relatively sophisticated user will recognize the so-called implicit claims that you describe. They are not the ones harmed by this behavior, because they know how to add security exceptions. The ones who suffer are the totally ignorant users, who are not savvy enough to understand implied claims, and are not knowledgeable enough to add an exception.
I wonder if you have ever used ssh. Anyone can generate an ssh public key without contacting a CA. In fact, an ssh public key comes with no certificate whatsoever, not even a self-signed one. The only thing ssh checks is whether the key is unchanged from what it was before. Are you therefore suggesting that ssh should be blocked, and unencrypted telnet should be allowed? For extra credit, compare the market penetration of ssh vs. https and tell me which one has done more in practice to improve user security.
Unlike sites with self-signed certs, sites with vanilla HTTP make no claim about their security.
Even if a self-signed cert involves a claim about security, why can't firefox just ignore that claim?
In other words, why doesn't firefox just treat self-signed certs as equivalent to vanilla HTTP?
Right now, firefox treats self-signed certs much more harshly than sites with vanilla HTTP. This is counterintuitive, counterproductive, and makes absolutely zero sense.
someone could hijack your bank website, use a self-signed certificate and Firefox would just ignore the authentication error.
What's to stop somebody from hijacking the bank website, redirecting to a website that uses no SSL at all, and waiting for the passwords to roll in?
Firefox and IE will, by default, warn you about sending unencrypted passwords. Once. And no more than once.
Of course, many or perhaps even most people will notice that the site is unencrypted, but the attacker doesn't need to fool everybody. Even a 20% success rate is plenty good enough.
I can't believe that out of five hundred comments no one has mentioned the more cynical viewpoint yet. Obama has a very good chance of being our next president. The Democrats have a very good chance of putting a Democrat in the White House in six months. When that happens, it's the Democrats who will get to use these new spy powers.
Why wouldn't Obama support this? Why wouldn't the Democrats support this?
By law, as of January 1, 2001, all firearms in Canada must legally be registered with the Canadian gun registry.
To purchase a handgun or other restricted firearms, a person must have a restricted licence and be a member of a certified range. To use restricted firearms a person must also obtain long term authorization to transport (LTATT) from their provincial Chief Firearms Officer (CFO) to move the firearm to and from the range.
... section 17 of Firearms Act makes it an offence for any person, including a security guard, to possess prohibited or restricted firearms (i.e. handguns) anywhere outside of his or her home.
Compare this with the US, where guns do not need to be registered, most people can legally purchase handguns without licensing, and (in some states at least) most people can easily get concealed carry licenses.
Yeah, it seems to me that it was an oversight that networking wasn't encrypted in the first place. When lots of these protocols were being developed, security didn't seem to be much of a consideration.
You may be too young to remember this, but until 1997, it was for all practical purposes illegal to transmit cryptography software over the internet because of ITAR regulations.
As a result, during the formative years of the internet when networking protocols were being designed, there was no practical way to include security as a requirement. A cynic would interpret this state of affairs as being exactly the goal that the US government had in mind when they made cryptography illegal.
There is a very limited supply of easily accessable fissable material on earth.
Actually, there is billions of years worth of economically accessible uranium in the earth. That's not even including thorium, which is (literally) as common as lead.
The myth of limited supply of fission fuel has to die. Whether or not you personally support nuclear power, I would expect (and hope) that you are in favor of giving the public accurate facts so that they can intelligently decide. The facts are that nuclear fuel supplies are enormous.
set up two printers in CUPS, one which takes postscript input and sends it to the printer shared by the VM and another which takes the resulting file and sends it as raw data to the printer.
This method doesn't always work, because some printers are so spectacularly bad that not even the raw data is enough to make it print -- you also have to send the raw data down the wire according to strict timing requirements.
Having said that, the Epson R1800 should mostly work in Linux; if all you can print is text, then you're doing something wrong.
I've tried the "Ubuntu live usb with persistence" approach, and it still runs into problems. For starters, persistence is implemented as an overlay on top of a (compressed) base OS, which means that any security updates that you install are not compressed. As a consequence, the amount of space available for system files decreases over time as you overwrite more and more system files with updated versions. On small or even moderate size (4GB) drives, you can easily run out of space in this way. Configuring disk compression in the overlay is surely possible, but it is not easy.
The other problem with Ubuntu on a pendrive is memory pressure. If you're going to run your main OS off a removable pendrive (as the ggp post suggests), that drive needs to work with a wide variety of machines. Some of those machines will inevitably have lower amounts of RAM. Running a regular-sized linux distro like the most recent Ubuntu on a low RAM machine is quite limiting, for two reasons. One reason is that running off a live USB usually involves using a RAM drive (for performance, and to save wear on your device). The other reason is that you cannot put swap space on a removable pendrive (it will wear out the flash memory too fast). Memory optimization, in the form of smaller binaries compared to Ubuntu, is one of the key features that makes FaunOS and similar projects usable on diverse machines.
So, the problem in a nutshell is that small footprint linux distributions like FaunOS inevitably lack security infrastructure (by which I mean not only "aptitude upgrade" but also community and institutional commitment towards timely and long-term security updates). Large footprint linux distributions like Ubuntu lack the various optimizations that are really needed for using thumbdrive-based OSes as a replacement for traditional OSes. There is no solution available that offers both long term credible security support and small footprint, at least as far as I know. (I would love to be informed otherwise, if you have something else to suggest.)
Again, "yum update" works wonderfully on Redhat and similar systems, but yum is not included in FaunOS. The entire problem that I'm trying to point out is that none of the USB-key based linux distributions include an easy mechanism for security updates, and certainly none of them make the issue of security updates very much of an organizational priority to the degree of mailing lists, bug reporting, and so on.
Slashdot Mirror
Uranium is a finite, non-renewable resource
Uranium is finite in the same sense that solar power is finite. With breeder reactors, the uranium on earth will be sufficient for billions of years.
Yes, I said billions.
Take an economics course. Government mandates HURT ECONOMIES. There is no exception to this rule.
I have taken an economics course. There are a number of very important exceptions to this rule. If you really have taken an economics course, and learned something from it, you would remember these exceptions.
One important exception is monopoly power. Another one is externalities. A third exception is asymmetric information. A fourth exception is public goods. All of these exceptions are relevant in one way or another to the auto market. I will leave it as an exercise to figure out why each of these classifications applies. If you really took an economics course, it should be very easy.
I do agree that very many develop Linux for "themselves", but I don't think geeks are more interested in struggling with network or sound or wifi or video tearing or printer drivers or webcams than other people. Nor are they particularly tolerant of buggy or inflexible software, even though they may have their own ideas of UI design.
The difference is subtle but it definitely exists. Here's one example that comes to mind. In GNOME or KDE, it's really easy to remove the main menu from the panel - just right click the thing and select "Remove". In Windows or Mac OS, however, it is nigh-on impossible to remove the Start/Apple menu from the taskbar/menubar. You can temporarily hide it, but it can't be removed entirely unless you start writing code. Geeks much prefer the flexibility of Linux, whereas most regular users are better off not having this flexibility, since a missing main menu is not a situation they can easily recover from.
Obviously there'll always be distributions for geeks by geeks, but I have no problem imagining distributions for joe average by companies that want to sell services to the residential or business market. The gap, the changes they'd have to make aren't that big compared to everything they could reuse.
There is a market here, and the required changes are indeed individually very small. (For example, the behavior I mentioned above would be easy to correct.) However, the changes are also numerous, and there are all sorts of rare cases where one can easily fail to notice a needed change. In order to do a good job, you need a large team -- not as large as the Windows or Mac OS teams, but still a good sized team. And, since few geeks are interested in volunteering for such a task, you're back to the situation where you need a revenue source in order to support that team.
It is instructive to look at the netbook market, which of course is the kind of Linux deployment that we're talking about -- selling Linux to Joe average. Netbooks may not be as profitable as laptops, but they bring in more than enough revenue to support a dedicated development team. The EeePC, for example, has a very inflexible desktop by any geek's standards: you can't even move or alter the icons, and forget about configuring the main menu, since there isn't any. In other words, it's just what grandma needs. But even this approach has its limits. One big problem is that it's hard to keep a modified platform in sync with the main trunk. Every time the upstream developers publish a security update, you have to integrate that update with the changes that you have made. This process requires continuous engineering effort far beyond the short product life of a particular piece of hardware. For a major distribution like Debian, there are ample volunteers to handle the task of integrating security patches, even for older versions of Debian. In the case of the EeePC, especially the older models, more often than not this work is not done.
To be fair, things are not entirely bleak. Ubuntu Netbook Remix is an example of a non-geek desktop interface which is integrated into a major distribution. (I'm not sure how much of its development was funded by Canonical vs. contributed by volunteers.) I can see such projects offering hope for Linux on the desktop, but it will never be a natural fit. You'll always need some effort in order to cajole Linux developers into contributing to mass market products. Even worse, such efforts have to be continuous and ongoing in order to best take advantage of traditional Linux strengths such as security.
I'm sorry, but this is just really, really stupid. The crux of your post is that Linux is doing better than it is, but nobody knows because Linux companies are so goshdarn humble. Do you actually believe that? And you wrote it on a site that proclaims every year to be the year of Linux on the desktop?
The crux of my post is that "doing better" does not mean the same thing for Linux as it does for Windows or Apple.
You speak of the mythical Linux on the desktop. I honestly don't think Linux will ever achieve success in the desktop market. I'm not saying this out of elitism -- as far as I can tell, it is cold objective fact. Most Linux developers are not traditional desktop users. By this I mean that even though Linux developers do use the desktop, they are not representative of the average desktop user. The Linux desktop will always be optimized for the average Linux developer rather than the average desktop user. Obviously, the ideal desktop for an average Linux developer is very different from the ideal desktop for an average user. That's why Linux will have a hard time in the desktop market -- because it's written by, and targeted to, developers, not traditional desktop users.
At the same time, Linux does not need to succeed on the desktop in order to be relevant. The vast majority of Linux developers will continue to contribute to Linux even if it fails in the wider desktop market. (Note that this is where Linux differs significantly from Windows -- if Windows fails in the consumer desktop market, it's game over.) The fact that you continue to use desktop penetration as your measure of success shows that you are still trying to judge Linux using Windows's standards. Such reasoning is faulty, and even more so now that large segments of the computing market (i.e. netbooks) are breaking away from the traditional desktop paradigm.
I didn't think Linux would survive. I thought that Linux and its users would be slaughtered. It has, and it has grown and improved. I am honestly astounded at the level of advancement that Linux has made
Any proprietary platform with the kind of market share that Linux has would be dead by now. It happened to OS/2, it happened to the various proprietary Unices, and many others (BeOS etc.). Hell, even Apple would have died in the 90s had it not been saved by, ironically, Microsoft.
The difference is that a free platform can advance even with a very small userbase, as long as enough skilled volunteers are willing to contribute. Linux probably represents more total man-years of work than Windows, even though its market share is much smaller. That's because Linux can easily recruit volunteer developers for free, whereas Microsoft has to pay huge salaries to attract top talent.
It's a mistake to judge Linux by the same standards as other commercial vendors. Free software really changes the rules of the game, and that's why Microsoft is on its heels.
Perhaps after the success of the switch adds Apple inched ahead of Linux on the desk top. But if you look over the last 15 years, I believe that there has been more Linux on the desktop than Apple OS's.
It hasn't been in anyone's interest to say that. I think that is even true of the Linux companies. For a long time they wanted to be under the radar under dogs. Perhaps because they didn't want a fight to the death with Microsoft.
This is an excellent observation and it lies at the heart of Linux's success. Anyone who fails to grasp this point is not even in the right ballgame.
In terms of economics, the most important difference between free and proprietary platforms is that free platforms do not require a large userbase in order to thrive. The number of skilled developers willing to work on Linux for free would remain very high even if Linux's userbase were to drastically shrink. (This is incidentally the main reason why Microsoft cannot win against Linux, at least not by any means available in the marketplace.)
By contrast, a proprietary platform requires a large userbase in order to even survive, since the only way for a proprietary platform to get developers is to pay them a salary, and salaries require money, which requires users. There are no volunteer developers who are even able, let alone willing, to contribute code to Microsoft Windows for free, because of the locked-in nature of the platform.
That's why proprietary vendors routinely inflate their usage numbers. Larger numbers are necessary in order to convince new users that the platform is worthy of adoption. If a proprietary platform does not have a lot of users, then it has no future, and rational users would not risk selecting that platform. That's why OS/2 died, and that's why Solaris (despite being made free recently) is about to die.
We thus have a situation where every vendor, other than the Linux vendors, has a huge economic incentive to inflate the reported size of their userbase. I don't necessarily mean illegal activity here; there are well known legal methods by which usage numbers can be inflated. For example, Microsoft counts every Vista OEM license as a Vista sale even if the user exercises downgrade rights to XP, and so on. In any case, it doesn't surprise me at all that the actual market share of Linux is far higher than what is being reported.
More like several thousand if you burn up all the uranium and the thorium we can reasonably extract and don't waste any of it.
It is actually billions of years, not thousands or even millions. The reason is that, unlike oil, the raw material for nuclear power (uranium or thorium) is so energy-dense that we can spend extraordinary efforts on extraction and still come out ahead, both financially and energetically. Thorium, in particular, is about as common as lead: there is a LOT of it in the ground. See Cohen for the calculation that produces the billions of years figure.
How many journals have you read? This seems to be par for the course, as far as my experience goes.
I am a research mathematician. Part of my job is to read journals. Admittedly, these are mostly math and computer science journals, which you might not count as science. If you count only journals which are strictly (non-computer) science, then I read Nature and Science, but very rarely any others, and this level of bias is definitely not typical among what I read.
So basically, to make Nuclear just fall off his chart, he assumes that building more powerplants will lead to nuclear war, and calculates how much stuff that will burn. Is that not completely absurd?
Yes, it is completely absurd.
Nuclear weapons have been used exactly two times in all of human history. Both times occurred in 1945, before the invention of nuclear power plants. There is absolutely no scientific or empirical evidence linking nuclear power plants to nuclear war.
Nuclear power production is responsible for exactly two civilian deaths worldwide in the last 22 years. It is by far the safest energy source available under current technology.
The authors rate nuclear energy negatively because it takes 10-19 years to build a nuclear power plant. The vast majority of those 10-19 years is consumed in regulatory oversight, not technical construction. I wouldn't be surprised if these same authors were responsible for lobbying in favor of said regulations.
The authors cite waste disposal as a disadvantage of nuclear energy. They fail to mention that uranium reprocessing dramatically reduces the quantity of nuclear waste. They also fail to mention that, with reprocessing, there is enough nuclear fuel to last for billions of years, i.e. longer than the sun will shine. They do link reprocessing to nuclear war, a notion which I thoroughly debunked in two sentences above. This treatment strikes me as incredibly biased and ignorant of scientific reality. While all of their points about regulatory and legal obstacles are factually true, these points belong in a policy paper or a legal draft. A scientific paper ought to focus on technical and scientific matters rather than legal issues. Again, I wouldn't be surprised if these same authors were responsible for lobbying in favor of the existing laws which ban reprocessing.
All in all, this paper does not belong in any sort of science journal. It is that bad.
Don't you think that a woman who loves and respects her husband would WANT to at least find out how he votes and why?
That's the same mentality as "If I'm not breaking the law, then what have I got to hide?" However reasonable the loss of secrecy is in this case, it is the first step in a slippery slope.
There is a very big difference between permitting spouses to reveal their votes to each other, vs. requiring it. As much as you might like to see the world through rose-tinted glasses, the reality is that a large number of households are not as harmonious as yours.
So how is such a vote, cast at home and sent in by mail not respect the voter's secrecy?
This is a very common question, and the very fact that you asked it is an excellent illustration of my point -- people in general do not realize that absentee ballots threaten the secret ballot system. The whole point of a step-by-step erosion is that each step by itself seems unthreatening.
The answer is: it is not possible to keep an absentee vote secret from other members in your household, especially if there is a disparity in power and voting tendencies within the household.
The usual example cited is that of Mormon women, which comprise a huge fraction of the electorate in Utah. If you mandate absentee ballots, it amounts to requiring that ballots be filled out in the home, under the eyes of a watchful husband. (In theory, ballots can be filled out in places other than your home, but in practice the only common case is that absentee ballots are filled out at home.) Under such a scenario, women are much more likely to vote in accordance with their husband's wishes than they otherwise would under a true secret ballot system.
Of course, coercion is also possible in a voting booth, but it is harder (e.g. voters can vote one way and say they voted the other way). Regardless of whatever flaws a voting booth may have, it is certainly at least as secret as voting at home, and very possibly more secret.
in the united states, indeed in every western democracy, ballots are secret. no one questions this anonymity -- indeed, it's mandated by law.
I totally agree with the need for secret ballots in general elections (even though I'm not posting anonymously).
I would like to point out in addition that absentee voting is becoming increasingly popular in the US. An absentee ballot is fundamentally incompatible with the mandate that ballots be secret, because there is no way (no constitutional way, at least) to enforce secrecy of ballots within the privacy of an individual's own home.
If we ever as a society lose our right to secret ballots, I suspect it will happen gradually, one step at a time, with mandatory absentee voting (already implemented in Oregon, and due by 2010 in Washington) being the first step.
I don't quite follow your reasoning. In order to convince you that some proof languages are readable I would need to be able to write formal proofs efficiently? ... As I mentioned, languages that are designed to make writing the proofs easier tend to be less readable.
Readability and writeability are separate and diametrically opposed issues. Writeability is necessary in order to speed up the development of new mathematics. Readability is necessary in order to improve the general public's understanding of and accessibility to the project. We apparently both agree that readability and writeability are conflicting goals. I am of the opinion that machine readable proofs are currently inadequate on both counts, i.e. machine readable proofs are not useful for developing new math, and they are also not useful for teaching mathematics to the general public. You on the other hand seem to be convinced that machine readable proofs are adequate on both counts, despite the conflicting nature of these goals.
I hope that the above accurately summarizes our positions. If not, feel free to correct any mistakes that I have made.
How many formal proof languages have you seen? Most of them are designed for software verification, not formalized mathematics. They are optimized to make it easier for the author, not for the reader.
Indeed, write-only code is a key aspect of perl which is shared by every formal proof language.
Vdash will (probably) use Isar which is as easy to read for a mathematician as a standard proof.
In order for this claim to be true to the extent that it would affect my assertion about advancing (or not advancing) the development of new mathematics, one would at a minimum need to be able to efficiently write existing proofs in the formal language (surely one cannot hope to accelerate the discovery of new proofs if even the existing literature cannot be efficiently handled). That means formal proofs for things like Fermat's Last Theorem, Poincare Conjecture, and so on, which I honestly don't ever expect to see in my lifetime.
Disclosure: I am a mathematician. I also graduated from Harvard, and I know Cam Freer personally.
You make several interesting but separate points in your post. Let me respond to them one at a time. Regarding pay-to-view websites, I can assure you that many mathematicians are just as unhappy as you are with the lack of public access to publications. This issue is larger than just mathematics (see for example musicians vs. RIAA, which is essentially same issue), and to pin the blame on mathematicians is unfair. This is a problem that society as a whole needs to solve in all its incarnations.
About making mathematics more accessible to the public, I agree that this step is necessary for ethical and practical reasons, and it's one of the reasons why I left a pure research position (with no teaching responsibilities) in exchange for a research position having some teaching responsibilities. However, I really doubt that public education would result in mathematics being further advanced. The cutting edge of new mathematics today is so far beyond the pale of the public that even a herculean effort at education won't be enough to bring very many people into the fold. There is a saying that the more you know, the more you know you don't know, and having been through a Ph.D program I can assure you that what you perceive as elitism is nothing of the sort -- the gap between current research and the general public is simply too wide for even the best educational efforts to bridge.
In fact, it's not even clear to me that increasing the rate of advancement of mathematics is the right goal. Perhaps we should even be holding back the development of new mathematics to give the public a chance to catch up with what we've done.
The project being discussed won't do any of these things. It won't make mathematics more accessible; anyone who's ever seen a machine-verifiable proof knows that they're even more obfuscated than perl code. It won't affect the overall rate of advancement in mathematics, because writing machine-verifiable proofs is much slower than writing standard mathematical proofs. It will of course increase our knowledge in the area of machine-verified proofs, but it's hard to argue that this area of mathematics is any more important than any other area like number theory. If you want to target the general public, the best way is still via sites like Wikipedia or Planetmath.org where the focus is on human-readable presentation.
The dumb part here is that they send the whole movie to your computer even if you're just watching the free two-minute preview. The two-minute restriction is only enforced in the flash applet. Now, no amount of DRM can stop a paying customer from copying the movie, but a smartly designed system could certainly make the customer pay for the movie before giving the whole movie to them.
it needs to provide the user feedback that the claim implicit in the URL is not supported.
Blocking the site completely (and yes, the firefox behavior amounts to a total block) is a far cry from simply providing user feedback. It's throwing out the baby with the bathwater.
Only a relatively sophisticated user will recognize the so-called implicit claims that you describe. They are not the ones harmed by this behavior, because they know how to add security exceptions. The ones who suffer are the totally ignorant users, who are not savvy enough to understand implied claims, and are not knowledgeable enough to add an exception.
I wonder if you have ever used ssh. Anyone can generate an ssh public key without contacting a CA. In fact, an ssh public key comes with no certificate whatsoever, not even a self-signed one. The only thing ssh checks is whether the key is unchanged from what it was before. Are you therefore suggesting that ssh should be blocked, and unencrypted telnet should be allowed? For extra credit, compare the market penetration of ssh vs. https and tell me which one has done more in practice to improve user security.
Unlike sites with self-signed certs, sites with vanilla HTTP make no claim about their security.
Even if a self-signed cert involves a claim about security, why can't firefox just ignore that claim?
In other words, why doesn't firefox just treat self-signed certs as equivalent to vanilla HTTP?
Right now, firefox treats self-signed certs much more harshly than sites with vanilla HTTP. This is counterintuitive, counterproductive, and makes absolutely zero sense.
someone could hijack your bank website, use a self-signed certificate and Firefox would just ignore the authentication error.
What's to stop somebody from hijacking the bank website, redirecting to a website that uses no SSL at all, and waiting for the passwords to roll in?
Firefox and IE will, by default, warn you about sending unencrypted passwords. Once. And no more than once.
Of course, many or perhaps even most people will notice that the site is unencrypted, but the attacker doesn't need to fool everybody. Even a 20% success rate is plenty good enough.
Why wouldn't Obama support this? Why wouldn't the Democrats support this?
Excuse me? I can't speak for Switzerland, but Canada's gun control laws are stricter than the US.
To quote from the Wikipedia page:
Compare this with the US, where guns do not need to be registered, most people can legally purchase handguns without licensing, and (in some states at least) most people can easily get concealed carry licenses.You may be too young to remember this, but until 1997, it was for all practical purposes illegal to transmit cryptography software over the internet because of ITAR regulations.
As a result, during the formative years of the internet when networking protocols were being designed, there was no practical way to include security as a requirement. A cynic would interpret this state of affairs as being exactly the goal that the US government had in mind when they made cryptography illegal.
Actually, there is billions of years worth of economically accessible uranium in the earth. That's not even including thorium, which is (literally) as common as lead.
The myth of limited supply of fission fuel has to die. Whether or not you personally support nuclear power, I would expect (and hope) that you are in favor of giving the public accurate facts so that they can intelligently decide. The facts are that nuclear fuel supplies are enormous.
This method doesn't always work, because some printers are so spectacularly bad that not even the raw data is enough to make it print -- you also have to send the raw data down the wire according to strict timing requirements.
Having said that, the Epson R1800 should mostly work in Linux; if all you can print is text, then you're doing something wrong.
The other problem with Ubuntu on a pendrive is memory pressure. If you're going to run your main OS off a removable pendrive (as the ggp post suggests), that drive needs to work with a wide variety of machines. Some of those machines will inevitably have lower amounts of RAM. Running a regular-sized linux distro like the most recent Ubuntu on a low RAM machine is quite limiting, for two reasons. One reason is that running off a live USB usually involves using a RAM drive (for performance, and to save wear on your device). The other reason is that you cannot put swap space on a removable pendrive (it will wear out the flash memory too fast). Memory optimization, in the form of smaller binaries compared to Ubuntu, is one of the key features that makes FaunOS and similar projects usable on diverse machines.
So, the problem in a nutshell is that small footprint linux distributions like FaunOS inevitably lack security infrastructure (by which I mean not only "aptitude upgrade" but also community and institutional commitment towards timely and long-term security updates). Large footprint linux distributions like Ubuntu lack the various optimizations that are really needed for using thumbdrive-based OSes as a replacement for traditional OSes. There is no solution available that offers both long term credible security support and small footprint, at least as far as I know. (I would love to be informed otherwise, if you have something else to suggest.)
Again, "yum update" works wonderfully on Redhat and similar systems, but yum is not included in FaunOS. The entire problem that I'm trying to point out is that none of the USB-key based linux distributions include an easy mechanism for security updates, and certainly none of them make the issue of security updates very much of an organizational priority to the degree of mailing lists, bug reporting, and so on.