Those situations are pretty rare, usually as a result of databases that store many GB in a single file. And if you're backing up a database it's usually best to use whatever online backup mechanism the database provides anyway, so that you don't have to take it offline every day for backups.
USPS has a government-granted monopoly over first class mail. They have no such monopoly over package delivery, and their whole infrastructure is built around the idea that most of what they deliver is letters. Compare the size of a mail truck to the trucks used by UPS or FedEx. They just aren't set up for that kind of work, and they haven't adapted to it.
It seems clear that IPoAC is a layer 2 protocol. The endpoints shouldn't have to support it at all. If you want it to be used more frequently then what you need is for e.g. Comcast to support it, but given their level of latency and packet loss it is likely that they already do.
The disagreement about the definition of smut is exaggerated: just about everybody "knows it when they see it," to quote Judge Stewart, for the vast majority of cases. To pretend there's vast disagreement about it is just disingenuous.
The problem, as always, is that "knows it when they see it" is not a meaningful legal test. Is a picture of a topless girl smut? What about a video showing two people having sex, but without showing any nudity? What if the same scene is three minutes long out of a 60 minute drama? What if it's three minutes out of 60, but the sex scene does show nudity? What if there is no sex or nudity at all, but a scene shows a woman striking another woman with a whip and presents the implication that she takes pleasure in it?
If you ask different people these questions, you will get different answers. Or just go back to the original example: Does tumblr go on the blacklist or no? Probably well over half the stuff there is not particularly objectionable, but there is no trivial amount of objectionable material.
The question of whether something is objectionable is completely subjective. You might be able to find two people who mostly agree, but you certainly won't be able to come up with a single set of principles to apply when making a decision on behalf of 300 million people. Some people will argue that a lesbian sex scene is not smut because it contains no penis, or that the Michelangelo's David is because it does. There are some people who think that graphic heterosexual sex is completely unobjectionable but even the implication of homosexuality should be prohibited. The idea that everyone agrees is ridiculous.
I think I have what would clear this mess up is fairly clear actually. Go out of business (including chapter 7 and 11) and the patent is public domain. Patents can be assigned to a business from a person within the first 6 months. Then after that they are permanent. Person dies they become public domain. This sort of thing would clip most of this sort of patent trolls. Company buys another company all the patents from the acquired company become public domain.
The trouble with this is that it screws up the economy. Companies that by all rights ought to merge for reasons having nothing to do with patents, but they can't because they would lose patents valuable enough to make it not worth it. Also, it harms start-ups because prospective investors and creditors would know that they couldn't sell the patents to recover part of their costs if the start-up fails.
But the overall idea is in the right direction. I think what would make more sense is to make patents "use it or lose it" -- not in the way most people suggest that you have to practice if you want to assert it, but more like the way trademarks work. If somebody is publicly infringing your patent for e.g. a year without a license and you don't send them a notice detailing which patent you think they're infringing by what activity, and then file a complaint in court if they don't stop or take a license, you permanently lose the right to assert any claims in that patent against that company.
That would put a stop to most of the submarine patents, because you can't wait until the entire industry has been using it for years before trying to assert it since by then the entire industry will be immune for your lack of action. It would also have the added benefit that it would temper the arms race between large companies to accumulate a huge patent arsenal for defensive purposes, because once those patents hadn't been asserted for a year they would no longer function. Which means that having a huge arsenal of low-quality patents is useless, and the only reason to get patents would be if you actually want to exclude competitors from your genuinely valuable invention, as originally intended.
Isnt this a rather good place for TXT records? img.tumbler.com TXT contains:unfiltered-images,usercontent
The trouble is that the tags don't tell you enough. Every website in the world has user content these days. If you block anything that approximately means "unmoderated user content" then you block everything from Slashdot to vendors' tech support forums to Google web search. But if you don't then you let through all the "explicit" content posted by users.
As for first amendment rights, as long as it is not being legislated or pushed into place by the government, what your employer filters is not a violation of your rights at all.
The only way the smut peddlers would actually register their content as smut is if the law required it; they know it means their site will be blocked and they have no interest in that. But if you force them then you have state action -- it's a de facto requirement that their site is blacklisted all over the place. You can look at the list of laws repeatedly passed and struck down by Congress trying to legislate movie ratings before the MPAA member companies got together to do it themselves on their terms so that they could stop having to repeatedly go to court getting the laws struck down. And that solution doesn't work with pornography because there isn't a tight cartel of pornographers that the government can harass into submission by passing a new unconstitutional law each time the old one is struck down -- there is always a new Larry Flint looking to make a point.
Gates is rich enough. But I'm guessing the local sales rep isn't a billionaire yet.
No, but that only makes it even worse: That sales rep sold out an entire country's people in exchange for a few grand worth of commissions.
More to the point, somebody hired a sales rep willing to do that. And that willingness stems from a flaw in Microsoft's business model. The quality and price of their products has never matched their market share; they subsist on inertia and lock-in. The problem with that model is that all it would take to break it is for a single medium-sized country to decide that they would rather spend a billion dollars once to implement all of the APIs, file format converters, migration tools, etc. to make switching from Microsoft to FOSS easy and popular in order to avoid having that same country's government and people continue to pay an even greater amount of money every year to a foreign corporation. The last thing in the world they need is for some oil-rich dictators to conclude that they could implement a feature-complete open-source equivalent of Exchange Server for less than the amount of money their country pays to license it.
So when Tunisia or China or whoever else comes to Microsoft and makes demands, Microsoft bends. Because Microsoft can't afford for those countries to make the path away from Microsoft's ecosystem simple, well-documented and conspicuous. So yes, you can blame the sales rep who did the deed, but that doesn't change the fact that Microsoft has left itself in the position that it has to yield to crackpot dictators who violate human rights in order to maintain its market dominance.
I'd like to see a scheme like the.xxx tld work well- simplifying things for people who don't want to encounter internet smut without error-prone filter setups and without futile attempts to keep that kind of stuff off the web entirely. But it looks like this is being done in the worst way possible.
The trouble is that it can't work that way. You can't exclude all smut to a single set of domains for a large number of reasons. For one thing, nobody really agrees on a definition. For another, any single domain may contain a wide variety of things: You can find a metric ton of non-smut on tumblr, but you can also find plenty of naked women there too. And you basically end up with two choices: Either you banish all of those websites in their entirety to.xxx and then all of their non-smut content ends up behind the filter (and you hit First Amendment problems in the US), or you let websites containing smut use non-.xxx domains, but then the filter doesn't actually block the smut because nobody uses exclusively.xxx when they can reach a larger audience by paying another $8/year to get the equivalent.com domain.
The problem is really with filtering in general, not with domains: You have a trade-off between false negatives and false positives. The only way to have a low number of false negatives is to have a high number of false positives and vice versa. And we decided a long time ago that it's better for government to accept the large number of false negatives and then let people choose for themselves what content they want to consume, than to have a government censorship board that decides what people can see and hear.
I can certainly see why Facebook hates it though: Not only does it deprive them of the tracking information for all the people who don't click the like button, it changes the user's choice in clicking the button from "click this button if you like the story, but you'll be tracked either way" to "click this button to cause Facebook to track you" -- and if it becomes common knowledge that that is how the like button works, fewer people will use it.
"Automatic tracking" can almost entirely be disabled already - and for years now. You just have to DO IT, and most people would rather bitch than spend the 5 minutes it takes.
If I'm just reading the news, I use whatever computer is in front of me. Sometimes that's my PC, or my laptop, or my PC at work, or a school computer, etc. Having to change a setting on every different computer I use is a huge annoyance, to say nothing of the times when I don't have administrative access to make certain changes.
Anything that makes protecting my privacy the default is a win.
the reason why legit ISP's don't get equipment seized is they keep records they give to law enforcement.
No it isn't. It's because if law enforcement seized the ISPs' equipment every time someone committed a crime using an ISP, no ISP would have any equipment. If AT&T decided that they would set up their DHCP servers to require users to change their IP address every two hours and then not keep any records of who had what IP address for any longer than that two hour period, law enforcement would not seize their equipment -- because it has no evidentiary value. The logs don't exist, seizing the equipment doesn't cause them to come into existence.
The thing is, a TOR exit node is the same thing. There is no evidence there to collect. The operator of the exit node is no more likely to be the perpetrator than anyone else on the internet who could have made use of it. Seizing the equipment is nothing more than harassment.
OK, let's see the language in the ISP agreement that has the account holder agreeing to be "legally responsible" for what anyone does with the connection.
If someone does something illegal, that person is the one responsible, not the ISP, the account holder, Microsoft, Obama, the mailman, etc.
The same logic has you impounding the computers of everyone at King's ISP, or any other ISP on the route between King's PC and the destination, since it could just as easily have been any of them.
Do you have any specific cases of a criminal conviction that relied solely on an IP address?
That's not the problem. The problem is that the evidence-gathering process has become so intrusive and expensive that being subjected to an investigation can have higher costs than actually being convicted of the crime. Seizing all of someone's computers can be extraordinarily disruptive to that person's life, so the evidence necessary in order to do so needs to be of a sufficiently reliable degree that we aren't subjecting scores of innocent people to it. An IP address should by no means be considered sufficiently reliable evidence for that.
You are kidding, right? I have had to clean "XP Antivirus 2011" and its ilk off of the computers of almost everyone I know who uses Windows. And the only reason that got removed is because it interfered with the user's use of the computer.
A criminal who wants to hide his tracks will compromise an extremely large number of PCs and route through them using the same sort of onion routing that TOR does so that the criminal can avoid being ensnared by a honeypot or leaving evidence on any given compromised PC sufficient to trace criminal activity back to its origin. The criminal seeking to mask his origin, unlike the scareware author, has no reason to interfere with the normal use of the PC so that the user will never be aware of the malware or think to have it removed. Moreover, the criminal will want to switch which PCs are used as proxies for illegal activity in order to avoid consistently reusing a small set of compromised PCs that would allow law enforcement to over time map the path and trace the traffic back to is origin. The bottom line is that it is extremely likely that a large fraction, perhaps even a significant majority of the Windows PCs on the internet have been or are being used to route illegal traffic for criminals.
On top of that, while it may be the case that for normal internet traffic, the source of the traffic is very likely to be in the household of the internet account holder, for illegal internet traffic the opposite is the case because the criminal knows that what they're doing is illegal and will in the typical case take steps to avoid being caught.
That's a reasonable expectation, because the vast majority of home internet connections are for one household and not shared.
That only applies if people don't do things like run TOR exit nodes. Advising people not to run a TOR exit node because no one else is doing it is self-fulfilling but ultimately destructive. What needs to happen instead is for everyone to run one so that the expectation is changed and no one is harassed for doing it.
(1) Apple products only "just work" if you only want to do what Apple allows you to want to do. Flash doesn't "just work" on iOS. Any given third party app that Apple doesn't approve doesn't "just work" because it requires jailbreaks and all of that. Try to get VoIP over 802.11 working on anything with carrier lockdown.
(2) You're conflating the way to get aircraft safety with what consumers want. Consumers want safe aircraft and working systems. Just because you get "safe aircraft" with closed design doesn't mean the same thing leads to "just works" software. See (1).
You're missing the reliance. It isn't that the computers necessarily create all of the value, although there is a lot of that there. It's that it's so utterly catastrophic if you do it wrong. Computer data is one of the most valuable things a company has that it can't practically insure.
If your software is well-written and your IT staff is competent, when your facility burns to the ground, you call the insurance company, overnight a new server from the OEM to the new office space, restore from the offsite backups and are up and running again inside of 24 hours. If your IT department is some kid who comes by once in a while to reboot the Dell Optiplex GX260 "server" running Windows 2000 Professional, at some point the head beancounter is going to come by and ask why he can't access the Accounts Receivable database. Then you're going to discover that the thing in the conference room that someone accidentally knocked out a fifth story window was a non-redundant storage device with no backups. Then you're going to call a bankruptcy lawyer.
More and more, I think the real challenge is reorganizing society on the basis of a recognition that we ought to be doing a lot less work, that we're producing things needed only in order to support unnecessarily high levels of productivity, and that we'd all be better off if we spent less time and energy on production.
I'm going to be pessimistic here. The problem is this: It's more efficient to have fewer skilled workers who each work for more hours. Given two alternatives, a company will almost always want to pay a single employee $40X for five days / 40 hours of work rather than paying five employees $8X for one day each, because it reduces training costs, benefit costs, etc. And there are plenty of people who are willing to work five days a week for five times the cash. But as automation reduces the number of employees required to produce all that is necessary for the world, it creates a large body of individuals who can't be gainfully employed doing anything productive, because all the production that is necessary is being done by a small fraction of the population.
You end up with a world where there exists a great deal of scarcity, but it isn't a scarcity of labor. It's a scarcity of suitable land, raw materials, energy, etc. You have a mass of unemployed people whose labor is not required to produce anything and so have no way to make a living, but can't acquire anything because products and services still require scarce non-labor resources. And those people aren't exactly willing to stand around while they slowly starve to death, or even go without a second vacation home, so they find ways to acquire resources other than by being productive: Arbitrage, government pork, litigation, war, fraud, nepotism, etc.
And so we "create jobs" for people who are good at stealing the wealth of productive people: High frequency traders, lobbyists, lawyers, defense contractors, criminals, PHBs, etc. The asshats generate their own spawn: If your competitor is making trades faster than you, you need to waste a thousand programmer hours making yours 1ms faster. If your opponent has more lobbyists or lawyers or soldiers than you, you don't have enough. High crime rates lead to purveyors of security theater snake oil. The CEO's nephew can hire all his incompetent drinking buddies to be your bosses.
The trouble is that there is a limit to the amount of labor required to produce everything we actually need, but there is no limit to the amount of human resources we can throw at fighting over the allocation of raw materials that are scarce for reasons entirely unrelated to a shortage of labor. Moreover, the person who retires to the beach to sip drinks out of glasses with little pink umbrellas without wanting to fight anyone for anything is the person who loses all their savings in the stock market, misses weighing in on the vote to deprive them of their social insurance, loses their house to a con man and gets drafted into the oil wars.
I don't want to sell advertising, and I'm not alone.
I suspect the trouble for you is that the intersection of a) Android developers who b) live in a country that Google won't send payments to, c) don't want to sell advertising, and d) won't set up a subsidiary or pay an escrow company to accept and forward payments from the US or Europe, is very small. Small enough that it probably doesn't make any kind of real difference to number of applications available for Android. Which means that if you're arguing that they're losing a lot of developers, you're taking the wrong tack, because they're not.
A tablet by its nature is not a corporate device. They're easy to break, which means that corporate IT would rather provide a stipend for employees to buy one and then let them be responsible for damage. It needs about the same level of IT policy as a television set, and the obvious way to lock one down is at the network level: You put a transparent proxy server on the WiFi, use HTTP-based document storage and retrieval and implement whatever access controls you need there on the server, which is platform agnostic.
It's easier to make the buttons bigger in your application and add some radial menus than to rewrite your application and strip out enough features to get it to run on a honeycomb tablet.
That helps the developer, not the user. Which means you still haven't solved the chicken and egg problem that people make apps for platforms people use and use the platforms that apps already exist for.
The reason people use Windows instead of Linux on the desktop is that they need to run that app that came with their camera which reads the proprietary image format, or they require some ancient cruft constructed using IE6 and Visual Basic, etc. Nobody is going to port one iota of that stuff to any other platform, ever, unless and until that platform has a substantial market share. And maybe not even then. Windows mobile is no exception.
What is far more likely to happen is that users will buy Apple and Google devices, developers will write the large majority mobile applications for those devices, and people will keep around a Windows PC for the legacy software that still requires it, until such time as a new version (or competing product) is released that runs on their Apple or Google device.
If people want integration, Apple is clearly in a far stronger position than Microsoft, because they have both a popular desktop platform and a popular mobile platform. Even Google is better positioned, because new software is actively being written for Google's mobile platform, so if Google starts pushing a desktop platform then it's a lot easier to make a desktop port contemporaneously with a new mobile application than it is to go back and rewrite all the cruft that has been accumulating in the Windows desktop ecosystem after the original developers no longer work for the company, old libraries or frameworks (like IE6) were used that are not supported on the mobile platform, fundamental design decisions were made without regard to mobile requirements, etc.
I seem to remember that the IE 7 fan's answer to that was along the lines of "why should I waste space on my PC's hard drive with two browsers?"
Firefox is around 80MB. I don't think you can still even buy a hard drive smaller than 80GB. Nobody actually cares about 0.1% of their disk space, it sounds like the guy was just looking for a reason to argue.
And besides, without some sort of MSI version of Firefox that can be administered with the Group Policy system, how will larger companies deploy Firefox alongside IE?
You appear not to understand how the United States mobile phone market works. The carriers are completely in control of the whole process because smartphones are priced for subsidized sale through a carrier, not sale directly to end users.
The customers can still choose which make of phone they get. Moreover, the carriers might choose who gets OS updates, but they don't really choose what goes in the app stores. And if people start complaining that iOS Safari doesn't support DNSSEC, Apple could easily silence them by publishing a free version-agnostic DNSSEC plugin in the app store.
I have another question: I want to deploy DNSSEC on a web site once my hosting provider supports it. Is the format of a certificate record standardized yet?
And which plug-in should I recommend to users of Google Chrome?
It appears that they're working on making the Firefox plugin work for Chrome but it isn't ported yet as of April, see here. There is also this which appears to be related to the same.
Slashdot Mirror
Those situations are pretty rare, usually as a result of databases that store many GB in a single file. And if you're backing up a database it's usually best to use whatever online backup mechanism the database provides anyway, so that you don't have to take it offline every day for backups.
Just use rsync with "--link-dest", I've seen multi-TB backups take less than a minute because most of the files are the same.
USPS has a government-granted monopoly over first class mail. They have no such monopoly over package delivery, and their whole infrastructure is built around the idea that most of what they deliver is letters. Compare the size of a mail truck to the trucks used by UPS or FedEx. They just aren't set up for that kind of work, and they haven't adapted to it.
It seems clear that IPoAC is a layer 2 protocol. The endpoints shouldn't have to support it at all. If you want it to be used more frequently then what you need is for e.g. Comcast to support it, but given their level of latency and packet loss it is likely that they already do.
The disagreement about the definition of smut is exaggerated: just about everybody "knows it when they see it," to quote Judge Stewart, for the vast majority of cases. To pretend there's vast disagreement about it is just disingenuous.
The problem, as always, is that "knows it when they see it" is not a meaningful legal test. Is a picture of a topless girl smut? What about a video showing two people having sex, but without showing any nudity? What if the same scene is three minutes long out of a 60 minute drama? What if it's three minutes out of 60, but the sex scene does show nudity? What if there is no sex or nudity at all, but a scene shows a woman striking another woman with a whip and presents the implication that she takes pleasure in it?
If you ask different people these questions, you will get different answers. Or just go back to the original example: Does tumblr go on the blacklist or no? Probably well over half the stuff there is not particularly objectionable, but there is no trivial amount of objectionable material.
The question of whether something is objectionable is completely subjective. You might be able to find two people who mostly agree, but you certainly won't be able to come up with a single set of principles to apply when making a decision on behalf of 300 million people. Some people will argue that a lesbian sex scene is not smut because it contains no penis, or that the Michelangelo's David is because it does. There are some people who think that graphic heterosexual sex is completely unobjectionable but even the implication of homosexuality should be prohibited. The idea that everyone agrees is ridiculous.
I think I have what would clear this mess up is fairly clear actually. Go out of business (including chapter 7 and 11) and the patent is public domain. Patents can be assigned to a business from a person within the first 6 months. Then after that they are permanent. Person dies they become public domain. This sort of thing would clip most of this sort of patent trolls. Company buys another company all the patents from the acquired company become public domain.
The trouble with this is that it screws up the economy. Companies that by all rights ought to merge for reasons having nothing to do with patents, but they can't because they would lose patents valuable enough to make it not worth it. Also, it harms start-ups because prospective investors and creditors would know that they couldn't sell the patents to recover part of their costs if the start-up fails.
But the overall idea is in the right direction. I think what would make more sense is to make patents "use it or lose it" -- not in the way most people suggest that you have to practice if you want to assert it, but more like the way trademarks work. If somebody is publicly infringing your patent for e.g. a year without a license and you don't send them a notice detailing which patent you think they're infringing by what activity, and then file a complaint in court if they don't stop or take a license, you permanently lose the right to assert any claims in that patent against that company.
That would put a stop to most of the submarine patents, because you can't wait until the entire industry has been using it for years before trying to assert it since by then the entire industry will be immune for your lack of action. It would also have the added benefit that it would temper the arms race between large companies to accumulate a huge patent arsenal for defensive purposes, because once those patents hadn't been asserted for a year they would no longer function. Which means that having a huge arsenal of low-quality patents is useless, and the only reason to get patents would be if you actually want to exclude competitors from your genuinely valuable invention, as originally intended.
Isnt this a rather good place for TXT records? img.tumbler.com TXT contains:unfiltered-images,usercontent
The trouble is that the tags don't tell you enough. Every website in the world has user content these days. If you block anything that approximately means "unmoderated user content" then you block everything from Slashdot to vendors' tech support forums to Google web search. But if you don't then you let through all the "explicit" content posted by users.
As for first amendment rights, as long as it is not being legislated or pushed into place by the government, what your employer filters is not a violation of your rights at all.
The only way the smut peddlers would actually register their content as smut is if the law required it; they know it means their site will be blocked and they have no interest in that. But if you force them then you have state action -- it's a de facto requirement that their site is blacklisted all over the place. You can look at the list of laws repeatedly passed and struck down by Congress trying to legislate movie ratings before the MPAA member companies got together to do it themselves on their terms so that they could stop having to repeatedly go to court getting the laws struck down. And that solution doesn't work with pornography because there isn't a tight cartel of pornographers that the government can harass into submission by passing a new unconstitutional law each time the old one is struck down -- there is always a new Larry Flint looking to make a point.
Gates is rich enough. But I'm guessing the local sales rep isn't a billionaire yet.
No, but that only makes it even worse: That sales rep sold out an entire country's people in exchange for a few grand worth of commissions.
More to the point, somebody hired a sales rep willing to do that. And that willingness stems from a flaw in Microsoft's business model. The quality and price of their products has never matched their market share; they subsist on inertia and lock-in. The problem with that model is that all it would take to break it is for a single medium-sized country to decide that they would rather spend a billion dollars once to implement all of the APIs, file format converters, migration tools, etc. to make switching from Microsoft to FOSS easy and popular in order to avoid having that same country's government and people continue to pay an even greater amount of money every year to a foreign corporation. The last thing in the world they need is for some oil-rich dictators to conclude that they could implement a feature-complete open-source equivalent of Exchange Server for less than the amount of money their country pays to license it.
So when Tunisia or China or whoever else comes to Microsoft and makes demands, Microsoft bends. Because Microsoft can't afford for those countries to make the path away from Microsoft's ecosystem simple, well-documented and conspicuous. So yes, you can blame the sales rep who did the deed, but that doesn't change the fact that Microsoft has left itself in the position that it has to yield to crackpot dictators who violate human rights in order to maintain its market dominance.
I'd like to see a scheme like the .xxx tld work well- simplifying things for people who don't want to encounter internet smut without error-prone filter setups and without futile attempts to keep that kind of stuff off the web entirely. But it looks like this is being done in the worst way possible.
The trouble is that it can't work that way. You can't exclude all smut to a single set of domains for a large number of reasons. For one thing, nobody really agrees on a definition. For another, any single domain may contain a wide variety of things: You can find a metric ton of non-smut on tumblr, but you can also find plenty of naked women there too. And you basically end up with two choices: Either you banish all of those websites in their entirety to .xxx and then all of their non-smut content ends up behind the filter (and you hit First Amendment problems in the US), or you let websites containing smut use non-.xxx domains, but then the filter doesn't actually block the smut because nobody uses exclusively .xxx when they can reach a larger audience by paying another $8/year to get the equivalent .com domain.
The problem is really with filtering in general, not with domains: You have a trade-off between false negatives and false positives. The only way to have a low number of false negatives is to have a high number of false positives and vice versa. And we decided a long time ago that it's better for government to accept the large number of false negatives and then let people choose for themselves what content they want to consume, than to have a government censorship board that decides what people can see and hear.
I can certainly see why Facebook hates it though: Not only does it deprive them of the tracking information for all the people who don't click the like button, it changes the user's choice in clicking the button from "click this button if you like the story, but you'll be tracked either way" to "click this button to cause Facebook to track you" -- and if it becomes common knowledge that that is how the like button works, fewer people will use it.
"Automatic tracking" can almost entirely be disabled already - and for years now. You just have to DO IT, and most people would rather bitch than spend the 5 minutes it takes.
If I'm just reading the news, I use whatever computer is in front of me. Sometimes that's my PC, or my laptop, or my PC at work, or a school computer, etc. Having to change a setting on every different computer I use is a huge annoyance, to say nothing of the times when I don't have administrative access to make certain changes.
Anything that makes protecting my privacy the default is a win.
There are public lists of exit nodes. They could trivially check the IP against the list.
the reason why legit ISP's don't get equipment seized is they keep records they give to law enforcement.
No it isn't. It's because if law enforcement seized the ISPs' equipment every time someone committed a crime using an ISP, no ISP would have any equipment. If AT&T decided that they would set up their DHCP servers to require users to change their IP address every two hours and then not keep any records of who had what IP address for any longer than that two hour period, law enforcement would not seize their equipment -- because it has no evidentiary value. The logs don't exist, seizing the equipment doesn't cause them to come into existence.
The thing is, a TOR exit node is the same thing. There is no evidence there to collect. The operator of the exit node is no more likely to be the perpetrator than anyone else on the internet who could have made use of it. Seizing the equipment is nothing more than harassment.
OK, let's see the language in the ISP agreement that has the account holder agreeing to be "legally responsible" for what anyone does with the connection.
If someone does something illegal, that person is the one responsible, not the ISP, the account holder, Microsoft, Obama, the mailman, etc.
The same logic has you impounding the computers of everyone at King's ISP, or any other ISP on the route between King's PC and the destination, since it could just as easily have been any of them.
Do you have any specific cases of a criminal conviction that relied solely on an IP address?
That's not the problem. The problem is that the evidence-gathering process has become so intrusive and expensive that being subjected to an investigation can have higher costs than actually being convicted of the crime. Seizing all of someone's computers can be extraordinarily disruptive to that person's life, so the evidence necessary in order to do so needs to be of a sufficiently reliable degree that we aren't subjecting scores of innocent people to it. An IP address should by no means be considered sufficiently reliable evidence for that.
Malware does happen, but it's also rare.
You are kidding, right? I have had to clean "XP Antivirus 2011" and its ilk off of the computers of almost everyone I know who uses Windows. And the only reason that got removed is because it interfered with the user's use of the computer.
A criminal who wants to hide his tracks will compromise an extremely large number of PCs and route through them using the same sort of onion routing that TOR does so that the criminal can avoid being ensnared by a honeypot or leaving evidence on any given compromised PC sufficient to trace criminal activity back to its origin. The criminal seeking to mask his origin, unlike the scareware author, has no reason to interfere with the normal use of the PC so that the user will never be aware of the malware or think to have it removed. Moreover, the criminal will want to switch which PCs are used as proxies for illegal activity in order to avoid consistently reusing a small set of compromised PCs that would allow law enforcement to over time map the path and trace the traffic back to is origin. The bottom line is that it is extremely likely that a large fraction, perhaps even a significant majority of the Windows PCs on the internet have been or are being used to route illegal traffic for criminals.
On top of that, while it may be the case that for normal internet traffic, the source of the traffic is very likely to be in the household of the internet account holder, for illegal internet traffic the opposite is the case because the criminal knows that what they're doing is illegal and will in the typical case take steps to avoid being caught.
That's a reasonable expectation, because the vast majority of home internet connections are for one household and not shared.
That only applies if people don't do things like run TOR exit nodes. Advising people not to run a TOR exit node because no one else is doing it is self-fulfilling but ultimately destructive. What needs to happen instead is for everyone to run one so that the expectation is changed and no one is harassed for doing it.
(1) Apple products only "just work" if you only want to do what Apple allows you to want to do. Flash doesn't "just work" on iOS. Any given third party app that Apple doesn't approve doesn't "just work" because it requires jailbreaks and all of that. Try to get VoIP over 802.11 working on anything with carrier lockdown.
(2) You're conflating the way to get aircraft safety with what consumers want. Consumers want safe aircraft and working systems. Just because you get "safe aircraft" with closed design doesn't mean the same thing leads to "just works" software. See (1).
You're missing the reliance. It isn't that the computers necessarily create all of the value, although there is a lot of that there. It's that it's so utterly catastrophic if you do it wrong. Computer data is one of the most valuable things a company has that it can't practically insure.
If your software is well-written and your IT staff is competent, when your facility burns to the ground, you call the insurance company, overnight a new server from the OEM to the new office space, restore from the offsite backups and are up and running again inside of 24 hours. If your IT department is some kid who comes by once in a while to reboot the Dell Optiplex GX260 "server" running Windows 2000 Professional, at some point the head beancounter is going to come by and ask why he can't access the Accounts Receivable database. Then you're going to discover that the thing in the conference room that someone accidentally knocked out a fifth story window was a non-redundant storage device with no backups. Then you're going to call a bankruptcy lawyer.
More and more, I think the real challenge is reorganizing society on the basis of a recognition that we ought to be doing a lot less work, that we're producing things needed only in order to support unnecessarily high levels of productivity, and that we'd all be better off if we spent less time and energy on production.
I'm going to be pessimistic here. The problem is this: It's more efficient to have fewer skilled workers who each work for more hours. Given two alternatives, a company will almost always want to pay a single employee $40X for five days / 40 hours of work rather than paying five employees $8X for one day each, because it reduces training costs, benefit costs, etc. And there are plenty of people who are willing to work five days a week for five times the cash. But as automation reduces the number of employees required to produce all that is necessary for the world, it creates a large body of individuals who can't be gainfully employed doing anything productive, because all the production that is necessary is being done by a small fraction of the population.
You end up with a world where there exists a great deal of scarcity, but it isn't a scarcity of labor. It's a scarcity of suitable land, raw materials, energy, etc. You have a mass of unemployed people whose labor is not required to produce anything and so have no way to make a living, but can't acquire anything because products and services still require scarce non-labor resources. And those people aren't exactly willing to stand around while they slowly starve to death, or even go without a second vacation home, so they find ways to acquire resources other than by being productive: Arbitrage, government pork, litigation, war, fraud, nepotism, etc.
And so we "create jobs" for people who are good at stealing the wealth of productive people: High frequency traders, lobbyists, lawyers, defense contractors, criminals, PHBs, etc. The asshats generate their own spawn: If your competitor is making trades faster than you, you need to waste a thousand programmer hours making yours 1ms faster. If your opponent has more lobbyists or lawyers or soldiers than you, you don't have enough. High crime rates lead to purveyors of security theater snake oil. The CEO's nephew can hire all his incompetent drinking buddies to be your bosses.
The trouble is that there is a limit to the amount of labor required to produce everything we actually need, but there is no limit to the amount of human resources we can throw at fighting over the allocation of raw materials that are scarce for reasons entirely unrelated to a shortage of labor. Moreover, the person who retires to the beach to sip drinks out of glasses with little pink umbrellas without wanting to fight anyone for anything is the person who loses all their savings in the stock market, misses weighing in on the vote to deprive them of their social insurance, loses their house to a con man and gets drafted into the oil wars.
I don't want to sell advertising, and I'm not alone.
I suspect the trouble for you is that the intersection of a) Android developers who b) live in a country that Google won't send payments to, c) don't want to sell advertising, and d) won't set up a subsidiary or pay an escrow company to accept and forward payments from the US or Europe, is very small. Small enough that it probably doesn't make any kind of real difference to number of applications available for Android. Which means that if you're arguing that they're losing a lot of developers, you're taking the wrong tack, because they're not.
A tablet by its nature is not a corporate device. They're easy to break, which means that corporate IT would rather provide a stipend for employees to buy one and then let them be responsible for damage. It needs about the same level of IT policy as a television set, and the obvious way to lock one down is at the network level: You put a transparent proxy server on the WiFi, use HTTP-based document storage and retrieval and implement whatever access controls you need there on the server, which is platform agnostic.
It's easier to make the buttons bigger in your application and add some radial menus than to rewrite your application and strip out enough features to get it to run on a honeycomb tablet.
That helps the developer, not the user. Which means you still haven't solved the chicken and egg problem that people make apps for platforms people use and use the platforms that apps already exist for.
The reason people use Windows instead of Linux on the desktop is that they need to run that app that came with their camera which reads the proprietary image format, or they require some ancient cruft constructed using IE6 and Visual Basic, etc. Nobody is going to port one iota of that stuff to any other platform, ever, unless and until that platform has a substantial market share. And maybe not even then. Windows mobile is no exception.
What is far more likely to happen is that users will buy Apple and Google devices, developers will write the large majority mobile applications for those devices, and people will keep around a Windows PC for the legacy software that still requires it, until such time as a new version (or competing product) is released that runs on their Apple or Google device.
If people want integration, Apple is clearly in a far stronger position than Microsoft, because they have both a popular desktop platform and a popular mobile platform. Even Google is better positioned, because new software is actively being written for Google's mobile platform, so if Google starts pushing a desktop platform then it's a lot easier to make a desktop port contemporaneously with a new mobile application than it is to go back and rewrite all the cruft that has been accumulating in the Windows desktop ecosystem after the original developers no longer work for the company, old libraries or frameworks (like IE6) were used that are not supported on the mobile platform, fundamental design decisions were made without regard to mobile requirements, etc.
You're describing how tablets are better than printed catalogs. How are they better than netbooks?
I seem to remember that the IE 7 fan's answer to that was along the lines of "why should I waste space on my PC's hard drive with two browsers?"
Firefox is around 80MB. I don't think you can still even buy a hard drive smaller than 80GB. Nobody actually cares about 0.1% of their disk space, it sounds like the guy was just looking for a reason to argue.
And besides, without some sort of MSI version of Firefox that can be administered with the Group Policy system, how will larger companies deploy Firefox alongside IE?
Such things exist.
You appear not to understand how the United States mobile phone market works. The carriers are completely in control of the whole process because smartphones are priced for subsidized sale through a carrier, not sale directly to end users.
The customers can still choose which make of phone they get. Moreover, the carriers might choose who gets OS updates, but they don't really choose what goes in the app stores. And if people start complaining that iOS Safari doesn't support DNSSEC, Apple could easily silence them by publishing a free version-agnostic DNSSEC plugin in the app store.
I have another question: I want to deploy DNSSEC on a web site once my hosting provider supports it. Is the format of a certificate record standardized yet?
There is an RFC.
And which plug-in should I recommend to users of Google Chrome?
It appears that they're working on making the Firefox plugin work for Chrome but it isn't ported yet as of April, see here. There is also this which appears to be related to the same.