Except that multiple webpages often share content and rely on caching to prevent repeated requests, HTTP 1.1 has allowed multiple requests to go over the same connection for ages, clients use multiple connections mainly because they specifically DON'T want users to have to wait until the whole page is loaded before starting to see other content appear, a significant portion of pages today are dynamically generated - in which case a packaging system like you suggest would actually add load to the server -, and it already IS possible to compress pages if the client sents the appropriate Accept header.
And what about clients that don't want or don't need the flash, javascript, css files, images or other things you think your page "depends on"?
What you are suggesting is a poor solution to something that isn't a real problem that would likely do more harm than good.
Profiling is just a form of benchmarking specifically intended to identify parts of code. When the code path taken is well understood and various parts can be exercised by changing parameters passed in, normal benchmarking techniques can easily be used to gather the same data as "proper" profiling.
Uhuh. And how many people do you think will be comfortable with calling up their ISP and tell them to access porn? People are worried enough about stuff like that showing up on their card statements, or buying soft mens magazines over the counter - that's one of the reasons online porn has grown so quickly in the first place.
This is and attempt at mandating filters that any parent can put in place by themselves. And nobody would be complaining if an ISP offered this as an opt-in service. Or for that matter if they were required to offer it as an opt-in service.
That raises the question of the motive. Why not just mandate it as an opt-in service? Or require ISP's to inform about the availability of filtering software? People can already filter - people who would like a filtered internet connection but don't have one is in that situation either because of cost, ignorance, or not being able to figure out how to install the software. Mandating ISP's to inform and to offer a server side opt-in alternative would solve those concerns.
So why opt-out?
The only reasonable explanation is that this guy doesn't trust parents to do what he thinks is right, so he wants to do it for them, but since he know he'd never get through a mandatory service, he's going for the next best thing: An opt-out solution that requires people to actively contact their ISP and ask for something that most will find embarrassing.
And as for your "protecting your kids against harm", even putting aside the discussion about whether or not porn is harmful: Kids who wants porn will get hold of it, it's that easy. Ever since my C64 days I remember how kids then would exchange floppy disks with badly pixelated, badly dithered, badly colored porn, a few pics per disk. The only thing filtering will do if your kids wants porn is drive them to either exchange CD's or USB keys, or into chatrooms or web forums that have avoided filtering, but where people will happily exchange files in private.
The only thing filtering will do is give you a false sense of security and teach your kids that they certainly can't come to you to talk about stuff they see that worries them, seeing as what they'd come across would be something they'd gotten hold off behind your back.
No, I'm not advocating shoving it in their faces or recommending it to them - I'm advocating sitting down with them and telling them about how there's stuff online they might find disturbing and how you'd prefer if they'd close it down and tell you about it if they see something that troubles them. Kids don't usually act like idiots unless you treat them as idiots.
You must be new here... It's tradition for someone on Slashdot to bash the French any opportunity they get, particularly insinuating they'll surrender whenever they get a chance.
Do you seriously think a politician is stupid enough to think a corporation like IBM is doing things just because it's better for most people?
They're being honest and telling him why they as a company have a stake in ODF, and why that should matter to Gov. Romney, and that's good. It may look like a sales pitch, but it's a sales pitch that targets the right "customer" very well - along with all the other reasons corporate spending in his state is a compelling addition.
Personally, I'd have to drop OpenOffice if I couldn't read and write MS Office files. Why? Because that's what I receive and that's what people expect 90% of the time or more. That is THE feature I need in a word processor - anything else I can compromise on. Sad, I know, but that is what we have to deal with until Open Office has a far greater market share than it currently has.
Messing around with an external converter would piss me off, possibly enough that I'd write a patch to make open office automatically use it to get back to the current usability.
In other words, the functionality is there because it is what users want. And this is open source - you'd be able to take it out of the "official" version, but my bet is that practically any distro that includes Open Office would apply patches to include automatic MS Office filters.
Fighting your users is only an acceptable strategy if you're a monopolist and your users have nowhere they can reasonably go. It doesn't work if your users can just go elsewhere, because they will.
You completely miss the point. I explicitly made the point that I was not suggesting moderation for this exact reason. I don't want anyone to be prevented from editing Wikipedia. I've edited the Wikipedia anonymously several times - not because I felt a need to be anonymous, but because I couldn't be bothered registering at the time.
My suggestion was a way of prioritising review based on peoples opinions of the person making the edit. And as someone not intent on destruction I'd welcome having my contribution looked over by someone else so I wouldn't be worried about my anonymous edits got flagged as having been done by someone with low reputation.
If that someone else want to revert my change or edit it further, then so be it. If I don't agree with it I can always come back and make further changes.
This is nothing that doesn't already happen - people already review articles and either revert changes or make new changes all the time, after all that's what Wikipedia is all about -, however today close scrutiny is mainly reserved for high profile articles that people care about and specific users that other contributors have strong feelings about.
A reputation system would go a long way to ensure "neglected" pages gets attention when someone with a low reputation have edited them - whether that low reputation is because they actively ruin pages or just because they are new -, in addition to the frequent attention that highly controversial and/or popular pages get.
One additional suggestion, though, based on your message: It would be easy enough to keep a list of the voting patterns for registered users (if they want to), and match them against each other, so that in addition to the list of edits by low reputation users, you could also get a list of edits by users that have been given a low rating by you and other users with substantially matching voting patterns.
It's a bit of number crunching, but it's simple to implement and it would make the tool more useful for someone with non-mainstream viewpoints.
It's all about creating mechanisms for automating what a lot of Wikipedia contributors already do manually: Scrutinise pages more heavily if they've been edited by someone they don't trust or know to do good work. I know if I see a lot of anyomous edits I am going to be a lot more sceptical than if I see a lot of edits by a person I know and/or respect the work of based on past experience.
At the same time I know anonymous contributions can have just as much value, so I don't want to lose them as long as they receive the scrutiny needed to weed out the junk.
It it's not "nothing" wrong and paying back money. It's "you claim I've done something wrong, and I claim I haven't, but I think it's worth it (for whatever reasons) to accept X as punishment right now if it will make the case go away without wasting either of our times with a protracted court case"
Of course you'd stand a better chance getting a settlement accepted if you admit wrongdoing, but often what the other party is after is mainly the punishment, and they couldn't care less if you admit doing anything wrong if you're willing to pay.
One reason for being prepared to take the punishment without accepting wrongdoing may be if you worry that being convicted may leave you open to lawsuits from other parties related to what you'd admit to.
I trust Wikipedia to give me accurate enough information for when I am looking up something that isn't important to me. Such as when I just happen to wonder how something works. Most of the time this is fine because the alternatives are either worse (looking up some random webpage that never gets any scrutiny, or where corrections aren't accepted) or because it's too tedious (looking up primary sources or spending time hunting down authorative sources and verifying their quality). Combined with the talk page and checking the edits for potentially controversial subjects it works well enough.
And yes, I do believe most papers have more lies and inaccuracies than Wikipedia. For starters, just from looking and the regular contradictions between the 4-5 major papers I flick through on a daily basis, I can say for certain that it's bloody hard to find a reliable newspaper. THAT is why I still read them - I've yet to find one I can trust enough to stick to as my only source of news. Instead I read multiple sources with different viewpoints that I have a reasonable idea of the bias and strengths and weaknesses of, and get an idea of where there are controversies or inaccurate reporting.
In Wikipedia the same thing usually plays out in the form of repeated edits or discussions on the talk pages quite quickly, which will also tell me if the page has received enough attention for me to be able to reasonably rely on grave errors or intentional misleading stuff getting caught, as well as allow me to easily see a summary of what controversies there are around a subject.
Really? Why do you think such a reputation system would be hard to implement? I think it would be tremendously simple (I'll explain why below), but apparently another (very similar) mechanism is already in beta testing. They're testing a mechanism to allow people to rate specific pages on several different criteria, and a system for looking up pages based on how they've been rated.
Implementing a reputation system for Wikipedia would simply mean:
Add a set of ratings for people to choose between on the history page next to each edit.
Each time someone click on a rating, add that rating to the user id (if logged in) AND an entry for the AS number (easily looked up from the IP) AND an entry for the C-net AND an entry for the IP address. Give higher weight to a user id than the other factors once it has received a certain number of ratings.
Assign the current rating for the user/AS number/C net/IP that did the most recent edit to the page, and if it is below a certain threshold display a warning.
Add a page that will list in ascending order from the lowest reputation of the most recent edit.
Add a button for a user to "bless" an edit - "lending it" their reputation (in which case future negative ratings for people to that edit will reflect on the user who blessed it as well, not just the author)
That's all. It's simple. It doesn't involve moderation. The worst case abuse of it is if someone manage to lower the reputation of an innocent party, and the worst consequence of that is more frequent reviews by other users. It makes it hard to avoid reviews by IP hopping, unless you happen to have access to lots of different ISP's (thanks to penalising the whole C-net and AS for bad behaviour), while protecting genuine users by weighting a registered user which builds up a history higher.
You need to keep in mind that you are dealing with people that HAVE the interest needed to put in the "effort" of registering new addressed - they are already willing to waste time vandalising pages repeatedly despite their "work" getting repeatedly undone.
The problem is: What does it take to get rid of them? Are these simply people that happen to see a page and edit it, or are they people who'd be willing to go to more trouble? Do you require an e-mail address? I have a dozen or so that I've used as throwaway accounts for testing etc. over the years that wouldn't be easy to trace to me, and everyone knows how to sign up for free e-mail accounts. So would it make a difference?
Verifying identity online is hard unless you're willing to take what for a site like Wikipedia would be considered extreme measures.
One idea: Implement a reputation system. Let people vote on the quality of specific changes. Don't moderate changes from people with a low reputation, but place them in a review queue and get a group of volunteers to focus on reviewing edits from the lower reputation sources.
For anonymous users, apply the reputation per ip, c-net, and AS number (an id used for routing purposes of blocks of ip addresses). Yes, I know that users can easily get another ip, and that some isp's will have multiple AS numbers, but it ought to make it easier to spot cases of persistent abusers.
You might even specifically ad a header to an article status "This article was most recently edited by an anonymous contributors / a contributor with low reputation and has not yet been verified - click here to see recent changes", to highlight to people that they may want to be particularly careful. That would make it more attractive to register if you do good work, while allowing those that do to make edits. Remove the banner after a higher reputation user has "verified" it.
Of course reputation does not mean you have to be more right than others - which is why I think it's still a good idea to post all edits and only use a reputation system to warn readers and to prioritise reviews.
Sure, he had an erroneous (libelous?) biography published on Wikipedia. Fucking change it
Last I heard we don't have any working time machines. The main complaint seems to be that this mis-information was available for an extended amount of time - on the order of months. Presumably he's pissed off that lots of people may have seen it in the meantime and believed what it said, which is a quite understandable concern.
The right response to that is another matter, but to claim that just changing it is a solution is short sighted.
True, but it would only be relevant if there was anything of significance there - if one were to start listing everyone the police might have been considering as possible suspects at one time or another, then you'd have a pretty long list which would tell people nothing than that the investigation was thorough.
Presumably the police would have looked into anyone and everyone involved with the Kennedy family to various extents at one point or another, and I'm sure lots of theories were thrown around. So in and of itself it wouldn't be noteworthy.
And if he was at some point mentioned, and if one were to include that in his biography, then it would only be fair to mention more about it as well:
There are several sources that lists Siegenthaler as one of several aides Robert Kennedy used to check up on investigations into the JFK assassination. So if he was at any point being looked over by the police or security services, something RFK would undoubtably know about and see the results from, RFK must at very least have been confident that nothing was found.
Should we close down Slashdot while we're at it? Lots of incorrect statements gets published here all the time, including character assasinations.
The fact that not everything on Wikipedia is accurate is well worth noting, but it's a huge step from that to it not being useful.
I use Wikipedia a lot, and for the most part I'm happy with the material, but if I am going to depend on the result in some way, I would look for more controlled sources.
However that is true of practically any source - online or offline - that I'd use.
Wikipedia is certainly far above most online sources I'd turn to in terms of quality, and that's what matters to me.
I'd rather see people get more critical about all sources they use... I see more lies and inaccuracies in an average daily newspaper than the total of errors I've come across in Wikipedia so far.
They were threatening no such thing. They standardised on ODF and made it clear they'd be happy to work with anyone who - by the time the policy goes into force in 2007 - supports ODF in the appropriate way in their software.
That MS chose to present that as if they were being excluded is more about MS' fear of competition and the free market than about reality.
Look at it this way: every couple of years, there's another big legal push by activist groups to eliminate the porn or in some way subdue it. All of this would stop (or at least decrease significantly) if most porn were successfully channeled into a segment of the 'net that could easily be blocked by consumers who don't want it.
Why do you believe that?
A significant number of the people clamouring for restrictions on porn do it based on arguments centered around morality, degradation or violence. If they believe porn to be degrading, morally reprehensible or cause increased violent behavior, why would they stop just because they were given "protection" for themselves?
These people are in it for the long haul - if they can't get broad prohibitions, they'll chip away at the "problem" one bit at a time.
The fact that Ralph Yarro is involved and is claiming to have gotten positive feedback from relativel heavyweights like Orrin Hatch is more than enough reason to start worrying about it.
Except it's a slippery slope. If specific content is restricted to specific ports, then the next thing you see will be all kinds of pressure groups pushing for ISP's to filter that port.
The "business man" in question, Ralph Yarro, is the guy that used to run Canopy group (SCO's largest shareholder) until he was ousted after a battle with the Noorda family over control. Hardly the kind of guy you'd want involved in anything requiring a sliver of ethics...
While I agree that the post you replied to is a bit on the paranoid side (the anti-trust settlement won't help Microsoft against brand new violations of the Sherman act for instance), your reponse validates one of the points of the post you replied to: He was pointing out exactly that because the 360 is a toy, it is an ideal "testing ground" if (and it's a big "if") Microsoft is considering a platform that can only run Windows.
If Microsoft ever goes that far, they'll immediately be under assault from all sides, and so it's not something they could take the chance on doing prematurely - if their protection isn't sufficient they'd catch all the heat without any of the benefits. Testing the technology on a console - where this kind of thing is already more or less accepted - would be a great way of getting experience.
Also, by starting out with the 360, and possibly moving the concept on to more "PC like" platforms like PDA's, they'd be able to grow public acceptance of a locked down platform/
Personally I think Microsoft would have loved doing something like this to PC's if they could, but I don't think they're stupid enough to try.
It definitively leaks. Both on Linux and Windows I regularly have to restart Firefox - it regularly grows to more than half a GB before I kill it, and closing down browser windows doesn't help. And before anyone suggests I remove the flash plugin etc., my Linux box NOT have flash installed. I've followed every single tip I've seen for tweaking prefs, turning off caching etc., and it hasn't helped.
It's been that way as long as I've used Firefox and Mozilla.
My Linux box is continuously on, but I use the Windows version on my work laptop, and that gets restarted once a day. Despite that I can rarely use Firefox for a whole day without restarting it unless I want everything to freeze up when it eats so much memory the machine starts trashing badly.
So what you are saying is that because you have an inadequate power supply it doesn't work for you. I'd have no problem running a 30kW heater on my supply.
There are plenty of electrical instant heating systems that can handle larger applications. The main reason they aren't that popular is that electrical heaters are simple, and to increase volume heated you have to duplicate most of it anyway (the heating elements needs to increase in size), so there's not really much to gain from having a single central system. Having localised supplies also means that if a heater fails you still have hot water elsewhere in the house AND you only need cold water pipes.
Slashdot Mirror
Not even that is entirely new - the UK version of Messenger has had outbound calls via a partnership with BT for ages.
And what about clients that don't want or don't need the flash, javascript, css files, images or other things you think your page "depends on"?
What you are suggesting is a poor solution to something that isn't a real problem that would likely do more harm than good.
Profiling is just a form of benchmarking specifically intended to identify parts of code. When the code path taken is well understood and various parts can be exercised by changing parameters passed in, normal benchmarking techniques can easily be used to gather the same data as "proper" profiling.
This is and attempt at mandating filters that any parent can put in place by themselves. And nobody would be complaining if an ISP offered this as an opt-in service. Or for that matter if they were required to offer it as an opt-in service.
That raises the question of the motive. Why not just mandate it as an opt-in service? Or require ISP's to inform about the availability of filtering software? People can already filter - people who would like a filtered internet connection but don't have one is in that situation either because of cost, ignorance, or not being able to figure out how to install the software. Mandating ISP's to inform and to offer a server side opt-in alternative would solve those concerns.
So why opt-out?
The only reasonable explanation is that this guy doesn't trust parents to do what he thinks is right, so he wants to do it for them, but since he know he'd never get through a mandatory service, he's going for the next best thing: An opt-out solution that requires people to actively contact their ISP and ask for something that most will find embarrassing.
And as for your "protecting your kids against harm", even putting aside the discussion about whether or not porn is harmful: Kids who wants porn will get hold of it, it's that easy. Ever since my C64 days I remember how kids then would exchange floppy disks with badly pixelated, badly dithered, badly colored porn, a few pics per disk. The only thing filtering will do if your kids wants porn is drive them to either exchange CD's or USB keys, or into chatrooms or web forums that have avoided filtering, but where people will happily exchange files in private.
The only thing filtering will do is give you a false sense of security and teach your kids that they certainly can't come to you to talk about stuff they see that worries them, seeing as what they'd come across would be something they'd gotten hold off behind your back.
No, I'm not advocating shoving it in their faces or recommending it to them - I'm advocating sitting down with them and telling them about how there's stuff online they might find disturbing and how you'd prefer if they'd close it down and tell you about it if they see something that troubles them. Kids don't usually act like idiots unless you treat them as idiots.
You must be new here... It's tradition for someone on Slashdot to bash the French any opportunity they get, particularly insinuating they'll surrender whenever they get a chance.
They're being honest and telling him why they as a company have a stake in ODF, and why that should matter to Gov. Romney, and that's good. It may look like a sales pitch, but it's a sales pitch that targets the right "customer" very well - along with all the other reasons corporate spending in his state is a compelling addition.
Messing around with an external converter would piss me off, possibly enough that I'd write a patch to make open office automatically use it to get back to the current usability.
In other words, the functionality is there because it is what users want. And this is open source - you'd be able to take it out of the "official" version, but my bet is that practically any distro that includes Open Office would apply patches to include automatic MS Office filters.
Fighting your users is only an acceptable strategy if you're a monopolist and your users have nowhere they can reasonably go. It doesn't work if your users can just go elsewhere, because they will.
My suggestion was a way of prioritising review based on peoples opinions of the person making the edit. And as someone not intent on destruction I'd welcome having my contribution looked over by someone else so I wouldn't be worried about my anonymous edits got flagged as having been done by someone with low reputation.
If that someone else want to revert my change or edit it further, then so be it. If I don't agree with it I can always come back and make further changes.
This is nothing that doesn't already happen - people already review articles and either revert changes or make new changes all the time, after all that's what Wikipedia is all about -, however today close scrutiny is mainly reserved for high profile articles that people care about and specific users that other contributors have strong feelings about.
A reputation system would go a long way to ensure "neglected" pages gets attention when someone with a low reputation have edited them - whether that low reputation is because they actively ruin pages or just because they are new -, in addition to the frequent attention that highly controversial and/or popular pages get.
One additional suggestion, though, based on your message: It would be easy enough to keep a list of the voting patterns for registered users (if they want to), and match them against each other, so that in addition to the list of edits by low reputation users, you could also get a list of edits by users that have been given a low rating by you and other users with substantially matching voting patterns.
It's a bit of number crunching, but it's simple to implement and it would make the tool more useful for someone with non-mainstream viewpoints.
It's all about creating mechanisms for automating what a lot of Wikipedia contributors already do manually: Scrutinise pages more heavily if they've been edited by someone they don't trust or know to do good work. I know if I see a lot of anyomous edits I am going to be a lot more sceptical than if I see a lot of edits by a person I know and/or respect the work of based on past experience.
At the same time I know anonymous contributions can have just as much value, so I don't want to lose them as long as they receive the scrutiny needed to weed out the junk.
Of course you'd stand a better chance getting a settlement accepted if you admit wrongdoing, but often what the other party is after is mainly the punishment, and they couldn't care less if you admit doing anything wrong if you're willing to pay.
One reason for being prepared to take the punishment without accepting wrongdoing may be if you worry that being convicted may leave you open to lawsuits from other parties related to what you'd admit to.
And yes, I do believe most papers have more lies and inaccuracies than Wikipedia. For starters, just from looking and the regular contradictions between the 4-5 major papers I flick through on a daily basis, I can say for certain that it's bloody hard to find a reliable newspaper. THAT is why I still read them - I've yet to find one I can trust enough to stick to as my only source of news. Instead I read multiple sources with different viewpoints that I have a reasonable idea of the bias and strengths and weaknesses of, and get an idea of where there are controversies or inaccurate reporting.
In Wikipedia the same thing usually plays out in the form of repeated edits or discussions on the talk pages quite quickly, which will also tell me if the page has received enough attention for me to be able to reasonably rely on grave errors or intentional misleading stuff getting caught, as well as allow me to easily see a summary of what controversies there are around a subject.
Implementing a reputation system for Wikipedia would simply mean:
That's all. It's simple. It doesn't involve moderation. The worst case abuse of it is if someone manage to lower the reputation of an innocent party, and the worst consequence of that is more frequent reviews by other users. It makes it hard to avoid reviews by IP hopping, unless you happen to have access to lots of different ISP's (thanks to penalising the whole C-net and AS for bad behaviour), while protecting genuine users by weighting a registered user which builds up a history higher.
You need to keep in mind that you are dealing with people that HAVE the interest needed to put in the "effort" of registering new addressed - they are already willing to waste time vandalising pages repeatedly despite their "work" getting repeatedly undone.
Verifying identity online is hard unless you're willing to take what for a site like Wikipedia would be considered extreme measures.
One idea: Implement a reputation system. Let people vote on the quality of specific changes. Don't moderate changes from people with a low reputation, but place them in a review queue and get a group of volunteers to focus on reviewing edits from the lower reputation sources.
For anonymous users, apply the reputation per ip, c-net, and AS number (an id used for routing purposes of blocks of ip addresses). Yes, I know that users can easily get another ip, and that some isp's will have multiple AS numbers, but it ought to make it easier to spot cases of persistent abusers.
You might even specifically ad a header to an article status "This article was most recently edited by an anonymous contributors / a contributor with low reputation and has not yet been verified - click here to see recent changes", to highlight to people that they may want to be particularly careful. That would make it more attractive to register if you do good work, while allowing those that do to make edits. Remove the banner after a higher reputation user has "verified" it.
Of course reputation does not mean you have to be more right than others - which is why I think it's still a good idea to post all edits and only use a reputation system to warn readers and to prioritise reviews.
Last I heard we don't have any working time machines. The main complaint seems to be that this mis-information was available for an extended amount of time - on the order of months. Presumably he's pissed off that lots of people may have seen it in the meantime and believed what it said, which is a quite understandable concern.
The right response to that is another matter, but to claim that just changing it is a solution is short sighted.
Presumably the police would have looked into anyone and everyone involved with the Kennedy family to various extents at one point or another, and I'm sure lots of theories were thrown around. So in and of itself it wouldn't be noteworthy.
And if he was at some point mentioned, and if one were to include that in his biography, then it would only be fair to mention more about it as well:
There are several sources that lists Siegenthaler as one of several aides Robert Kennedy used to check up on investigations into the JFK assassination. So if he was at any point being looked over by the police or security services, something RFK would undoubtably know about and see the results from, RFK must at very least have been confident that nothing was found.
The fact that not everything on Wikipedia is accurate is well worth noting, but it's a huge step from that to it not being useful.
I use Wikipedia a lot, and for the most part I'm happy with the material, but if I am going to depend on the result in some way, I would look for more controlled sources.
However that is true of practically any source - online or offline - that I'd use.
Wikipedia is certainly far above most online sources I'd turn to in terms of quality, and that's what matters to me.
I'd rather see people get more critical about all sources they use... I see more lies and inaccuracies in an average daily newspaper than the total of errors I've come across in Wikipedia so far.
I haven't tried the final release, but as of RC2 it still leaked massively for me at least.
That MS chose to present that as if they were being excluded is more about MS' fear of competition and the free market than about reality.
Why do you believe that?
A significant number of the people clamouring for restrictions on porn do it based on arguments centered around morality, degradation or violence. If they believe porn to be degrading, morally reprehensible or cause increased violent behavior, why would they stop just because they were given "protection" for themselves?
These people are in it for the long haul - if they can't get broad prohibitions, they'll chip away at the "problem" one bit at a time.
The fact that Ralph Yarro is involved and is claiming to have gotten positive feedback from relativel heavyweights like Orrin Hatch is more than enough reason to start worrying about it.
Except it's a slippery slope. If specific content is restricted to specific ports, then the next thing you see will be all kinds of pressure groups pushing for ISP's to filter that port.
The "business man" in question, Ralph Yarro, is the guy that used to run Canopy group (SCO's largest shareholder) until he was ousted after a battle with the Noorda family over control. Hardly the kind of guy you'd want involved in anything requiring a sliver of ethics...
If Microsoft ever goes that far, they'll immediately be under assault from all sides, and so it's not something they could take the chance on doing prematurely - if their protection isn't sufficient they'd catch all the heat without any of the benefits. Testing the technology on a console - where this kind of thing is already more or less accepted - would be a great way of getting experience.
Also, by starting out with the 360, and possibly moving the concept on to more "PC like" platforms like PDA's, they'd be able to grow public acceptance of a locked down platform/
Personally I think Microsoft would have loved doing something like this to PC's if they could, but I don't think they're stupid enough to try.
It's been that way as long as I've used Firefox and Mozilla.
My Linux box is continuously on, but I use the Windows version on my work laptop, and that gets restarted once a day. Despite that I can rarely use Firefox for a whole day without restarting it unless I want everything to freeze up when it eats so much memory the machine starts trashing badly.
So what you are saying is that because you have an inadequate power supply it doesn't work for you. I'd have no problem running a 30kW heater on my supply.
There are plenty of electrical instant heating systems that can handle larger applications. The main reason they aren't that popular is that electrical heaters are simple, and to increase volume heated you have to duplicate most of it anyway (the heating elements needs to increase in size), so there's not really much to gain from having a single central system. Having localised supplies also means that if a heater fails you still have hot water elsewhere in the house AND you only need cold water pipes.