The only problem I have is that what do you do with a device that will be 10x13, cannot bend, and costs $300?
You stuff it in your bag where all the classic paper based stuff is? And about the price, a while ago somebody did a little calculation, result was that a one year subscriptions to the NYTimes could get you a Kindle for free, as most of the money is spend on delivering the paper.
Judging from all the eInk stuff I have seen, not any time soon. The refresh times are still measured in seconds and still have that "invert the screen" thing going on. I have seen laser printers that could print faster then those eInk things can refresh.
However outside of eInk based displays there is PixelQi who have build the OLPC screen and are now building stuff normal computers. Judging from the OLPC display, they certainly have the potential to rival eInk based stuff. The OLPC display is 200dpi, sunlight readable and has refresh times that are perfectly fine for watching video, where the OLPC display fails is at indoor readability without the backlight, as you need lots of light to see something on the screen. The display is also not white, but gray, so it reads more like recycling paper. However its still a really good display for reading books on it and its already a few years old, so they had plenty of time to improve upon it.
While it would be true that it would give you a percentage of actual usage, it still wouldn't be a valid estimate of actual Linux market share, as a lot of desktop users have Linux, but don't use it as their primary OS, instead they dual boot into it. I wouldn't be surprised if the actual usage count is at 1%, while the install base is more around the 10%.
The problem is that this only works if the information is valuable enough for people to actually copy it. With by far the most information on the net, that just never happens. And even if you copied it, you don't have any way to check that its authentic and not a manipulated copy and neither do you have a way to keep your copy online for the general public, as DMCA takedown notices will make sure that the information can only be found in obscure places.
I think the biggest problem with the Internet is the lack of unalterable static publications. You don't publish a thing on the Internet and then move on. You publish it, apply some bug fixes later, alter this and that, move it over to a new framework years later and so on. Things on the Internet just aren't static enough for proper archival, webpages evolve over time and even if you would do daily backups of webpages, you would miss plenty things. Things might look different when there would be a way to directly access the databases driving the webpages, but most of the time thats not the case. Even popular "open" projects fail at this, Wikipedia for example has all the history of edits stored and viewable, but they don't have that for deleted articles, those are purged and unviewable by the general public.
As I keep pointing out on each IPv6 story, there will be little motivation to move to IPv6 until you can hit major sites, like cnn.com and slashdot.org, using nothing but IPv6 packets.
That is never gonna happen, because enabling IPv6 for major sites has zero advantages and will break plenty of stuff. With IPv6 it isn't a "you have it" or "you don't have it" thing, for lots of consumers its a "you kind of have it, but routing is pretty broken" thing, meaning half the servers out there will not be reachable over IPv6, even if both parties have IPv6. And since plenty of software out there does the "clever" thing to default to IPv6, this means servers that worked before on IPv4 will no longer work once they have a IPv6 AAAA record or will only after a long timeout. Result of all this is that nobody is using AAAA records for their main DNS names, i.e. you get an AAAA record for ipv6.google.com, but not www.google.com. This kind of connection trouble of course would quickly fade away when IPv6 is actually used instead of just an experimental toy thing, but for quite a while peoples only IPv6 access will be via broken 6to4 tunnels.
Or to look at it from a different angle, there simply is no need for IPv6 for major sites, as there are so few of them that they aren't running out of IPv4 addresses anytime soon, they simply pay premium as soon as they get really rare.
The only way I can see IPv6 moving forward is if you start with it on the client side (torrent, ip telephony, whatever), since client to client communication is what is truly broken in IPv4 due to all the NAT'ing and other hacks and using IPv6 could make things there better instead of worse, since the status quo with IPv4 is already completly awful. And of course we already ran out of IPv4 addresses on the client side a long time ago, which is why the whole NAT thing become necessary in the first place.
In the end I however wouldn't expect miracles, this isn't something that is going to move much forward by free market forces anytime soon, this is something that could need a serious government push to get momentum. Or if that doesn't happen, I guess we have to wait till China, India and all the other counties see benefits of IPv6, before somebody here realizes that it might be a good idea to start with it too.
Any kind of bitmap scaling is going to look ugly for small non-integral factors (like 1.5x), which are typical when you zoom. The only thing that would solve this is adoption of SVG.
An easy fix would be to provide highres images and then scale those down, instead of providing lowres images and then trying to scale them up, but for that to work without wasting bandwidth you would need an image format that supports something like mipmaps and allows to only download the lower res part. Given how long it takes for an image format to be adopted, I consider that however unlikely to happen in the near future.
For the most part, it's the problem with the way specific sites are designed, not with browsers themselves.
Given that close to all non-trivial pages break, I doubt it. I think there are two core problems, one is that image size continue to be given in pixel instead of 'em', as most browsers are pretty bad at scaling images, thus the page layout stays rather inflexible. The other problem is that CSS doesn't have a proper way to know the size of the content its formating, so every specification of size in the CSS makes assumption about the content that no longer hold true when the font size changes, leading to overflowing boxes and all kinds of misery. Which of course relates to the problem of CSS based layout being extremely hacky compared to past table based layouts. With table based layouts its completly trivial to place to boxes horizontally next to each other with zero risk of accidental overlap, with CSS based layouts on the other side its a complete crazy hack-fest of overlapping boxes, hard coded margins and a whole bunch of other illogical crap and that is even considered "good practice", *yuck*. I would wish that CSS would allow to define layout tablets in the CSS files, thus getting rid of all the crazy hackery and still having that semantic/layout separation.
I don't see how they could possibly do that while remaining conformant with HTML and CSS. If the latter requires text layout to produce overlapped text, then that's pretty much it.
Well, the whole point is that it shouldn't be conformant, as neither of those standards guarantees good readability. I want a button that toggles between standard conform layout and readable layout. Look at the Readability script, its truly awesome, it strips out all the navigation bars, advertisements and all other distracting crap, leaving only the raw article content of a webpage, well formated with proper margins, font size and all. Its not perfect, as it doesn't work with all webpages, but when it works it improves the readability of a webpage a hell of a lot. The whole crux with browsers is that they are only good for viewing webpages, not reading them.
Instead, how about educating the web designers to use proper reflowing layouts?
The only "proper" layout is that decided by the user.
I really couldn't care less. Webbrowser these days seem to try everything to get pixel perfect rendering done, yet utterly fail at producing good looking readable webpages when there is even the tiniest deviation from the default. Try browsing with a larger default font for example, 99% webpages break, some worse then other, but pretty much all of them break. On Slashdot for example the "Reply to This" button falls apart on other webpages you are confronted with overlapping text and other unusable crap. And before somebody mentions the zoom feature, Firefox under Linux doesn't doesn't do any filtering when scaling, so all graphics look complete shit when zoom is used, making zoom unusable. There is other stuff that is annoying, for example the lack of build in support for link tags introduces in HTML2, you can get support via a plugin, but it would be nice to have solid support for that feature out of the box, maybe webpages would then finally start using it. But the most annoying thing is probably the lack of alternative view modes, I would like to have a modes that do not conform to pixel perfect rendering, but instead focus on producing readable results, i.e. avoiding overlapping text, making sure that line-width isn't to large, hide the navigation bars and all that other stuff, yet all the browser offers is pixel perfect rendering and rendering with no style sheets at all, neither of which is very readable. Luckily there is Readability which helps a good bit with that, making sure line-width is proper and navbars are gone, but again, it would be nice to have such basic stuff build into the browser.
The obsession with pixel perfect rendering and the complete ignorance on readable results is truly annoying and goes against anything that was considered "good practice" in the good old days.
Whereas people like me are advocates of just scrapping the whole damn thing because the potential benefits of doing so are way more interesting
What would be these "potential benefits" compared to a moderate copyright law (aka. 10-20 year term, allow non-commercial use)? The only difference I see is that no copyright would make it trivial for cooperations to exploit the authors even more, by taking the authors work, slapping their company name on it, changing it to fit their need, distributing it via their channels and all that without the author ever seeing a dime. I am pretty sure that that would piss of quite a handful of artists and not exactly be a good way to increase creativity.
None, nada zilch. No exceptions, no excuses. Default policy "Deny all".
That would be a bad idea, as it would lead to a lot of annoying messages poping up, doing nothing but training the user to ignore them and click them away quickly, as they are clearly useless. If you want warnings and user confirmation you have to make sure that they happen very seldom so that the user will actually notice them. The way that the OLPC/Bitfrost does it is by defining security categories and then defining combinations that are forbidden/require user intervention, i.e. you can have Internet access without warning, but if you do that you can't get file access or access to the web camera. On the other side you can have webcam access, but then no Internet access. That way a lot of applications can work just fine without having a confirmation dialog, as the access they have can't do harm.
Yes, there is an assumption that the user is in control of their applications, and I, for one, like it.
The user isn't in control of its applications, he can start and stop them, everything that happens in between is up to the application and the user doesn't even have a proper way to verify what is happening, as there is no log kept (well, you could strace, but thats impractical when applied to all applications).
Right. Until you get the first user who wants to upload a picture with their secure browser. And then you get the second user who wants to open their favorite movie with mplayer, because WMP choked on it. And you don't get a third user, because your system is unusable.
As mentioned before, that would be absolutely no problem in a sandbox, as you would simply pass the data to the application directly or 'mount' the requested data into their namespace, no need for the application to have full system access. Need an example? How about:
% mplayer - Ever seen a serious Flash application? No? You know why? Because you can't do anything interesting in a sandbox.
There have been tons of decent games written in Flash, I don't see why a game should have more rights to mess up my system when it comes as.exe instead. You are seriously attacking the thing from the wrong viewpoint. Its not about all the weird edge cases that might be troublesome (no doubt there are some, even so you haven't yet really named one that isn't trivially solvable), its about the 99% of applications that would work just fine.
If the user is tricked into installing crap, clearly they shouldn't have root.
You don't need to be root to install crap, that works just fine as a normal user and can do just as much harm on your regular single user desktop system.
If they're bribed, why should the computer refuse
The whole point is that the computer shouldn't refuse, instead it should isolate the applications so that they can't do any serious harm, even if they are evil. The whole trouble today is that applications simply aren't isolated enough from each other. Each application has full access to all the files and devices that the user has access to, which is far more then is needs and a clear violation of 'least privilege' principle. Most applications would work just fine if they would only have access to a place for configuration files and their own data files and nothing else. They wouldn't even need to have raw file system access if they would want to save things outside of those locations, as load/save could be handled by the OS, i.e. instead of having the application present the load/save dialog, its the OS that displays the dialogs and reads the files, all the application gets to see is the data, not the filesystem, thus ensuring that it can never overwrite on its own.
Or to attack the thing from a different angle, what would you say if a Flash application had full system access? Kind of a bad idea, don't you think? Why would you want far less security when a random game comes as.exe instead of Flash app on a webpage, why shouldn't both use cases, which are pretty much identical, have the exact same tight security applied?
The other is user security. And you cannot solve that problem with technology.
You can't solve all user security issues with technology, but you can solve a hell of a lot then todays OSs try. The whole insecure password thing could be easily fixed with smardcards/usbstick and pincodes, by far most applications could be run in isolation not having full system access and the filesystem could provide versioning so that even if things would go wrong, you could roll the changes back and get a look of what happened in the first place. None of them are rocket science or especially expensive and all of them make the system *more* convenient to use, not less. This whole "blame the user" business is in large part just an excuse for bad engineering. Its like the whole phising thing, yeah, users shouldn't click on everything, yet I have yet to see a single mail from a larger company that actually signs there mails cryptographically, which would make automatic phising detection trivial without the user having to lift a finger.
Looking up trivial facts in a database and spitting them out is easy, and not particularly significant...
The significant part here is that the database queries will be in natural language instead of SQL and they will be kind of vague. Which is kind of a big deal, considering its one of those areas where pretty much all search engines these days fail at, i.e. you can't ask questions to Google and get an answer, instead you simply do full text search and get all pages matching, no matter if they have anything to do with your question or not.
I have some doubt about that. Look for example at the flightsim genre. That genre was pretty hardcore and popular a decade ago and is now pretty much completly dead, no release in years and nothing announced or the future. The few games still around that involve planes (AceCombat, Hawks) are an utter joke compared to a decent ten year old flightsim, as they don't have a drop of realism left. The thing to realize is that todays "hardcore" is already really soft and mass market oriented, compared to games of the past.
That said, I don't think that every game will be like Wii Sports in the future, as there is a big enough market share for Hollywood style over the top action games, but they will be or already are, rather watered down experiences lacking in depth and that trend will only continue, since it simply sells more copies.
This really isn't an either "this or that" thing, you really want both elements. Games that force you to do things because the story dictates it suck, but so do games that don't have any story at all.
What I as a gamer want are interactive worlds that I can explore without hitting an invisible wall and that have enough simulation in them that "story" can evolve on its own without every bit being prescripted. Prescripted events in fact are the biggest problem I have with games today, you can take most FPS and completly break them by not following the predefined route. Take Killzone2 or Call of Duty 4, what happens when you don't follow your team mates and instead turn around and explore the area? Absolutely nothing. The war just stops and waits for you to touch the next trigger, if you are not there, nothing happens, the area doesn't offer anything interesting to explore. Compare that with a Operation Flashpoint or a EF2000, those games feature dynamic missions without trigger spots, the war doesn't wait for you, it just continues even if you are not there. The events that evolved from those dynamic gameplay where *far* more memorable then anything I have ever seen in a scripted shooter, because those events where real, not the magic hand of the designer fucking with virtual reality.
Now that said, meaningful choices and believable characters are of course extremely important, but the way to reach that goal isn't by a completly linear path for the player to follow, its by creating a flexible enough virtual environment that allows meaningful interaction with the gameworld and thats exactly where most linear games fail at.
The thing with WoW is that while its a total time sink, its also at the same time a very non-demanding game. You don't really have to use your brain to make progress in WoW, you just have to walk around and click on things. There are no puzzles to overcome or insane tactics to master, its just a matter of clicking on stuff. It might get quite a bit more hardcore at the higher levels, but before a casual gamer reaches them he will have to sunk in hundreds of hours of very casual play into the game.
If you can believe vgchartz.com there are 51 million hardcore consoles (PS3, Xbox360) around, while "only" 49 million casual gamer consoles (Wii). Seems to me to be an even match, not exactly what I would call death of hardcore gaming.
They have established a totally free online encyclopedia. No ads. They have had to balance quality with quantity.
A lot of what they accomplished, they accomplished before they turned into an elitist club. Now they seem to be more focused to preserve what they have, then to broaden their scope, which of course pisses of all people that actually want to add to it.
They have established rules that define what is encyclopedic.
Well, yeah, there have to be some rules, nothing wrong with that in principle, the trouble is that some of the rules are idiotic and instead of giving an article a chance, they now delete every border case. On the positive side of things however the English Wikipedia is still lightyears better then the German one, I consider the German one pretty useless these days, because it almost never has the in depth information I am looking for.
Soon we no longer need actors and we just need digitized versions of them.
We will have digital version of them, but we still need actors. The whole human digitalization so far is in very large parts based on motion capture. If you don't have an actor, there isn't any data to render. Even completly hand animated stuff, like Pixar movies, still have a human voice actor for reference. So far the whole digital side of things has done *nothing* in reducing the need for actors, it has done quite the opposite, you now not only need actors, often you have multiple onces for different tasks in a single role, you also need graphic artists, modeler, programmer and all that stuff to drive the digital puppet.
One day we might reach a point where we can generate the actual acting, but we are still very far away from that point. There really hasn't even been much research into that area aside from really basic stuff (i.e. walk cycles and such). What you however don't need in the future is a specific actor, since you can just capture a similar one and transfer his movement onto the other one.
The only way to make a blacklist work is to not do one. Blacklists are completly pointless for child porn, because its *ALREADY* illegal in pretty much any country that happens to have Internet. If you find child porn you can take down the server and be done with it, leaving no easy way around it. On the other side if you block it via a stupid blacklist you make sure that it stays available for everybody actually interested in the topic, since the filters are trivial to work around.
A Youtube video of the Uwe Boll movie Seed is blocked via that list as well, no kidding. The ridiculous part of those list is that they are completly unneeded, when there is real child porn on a server it will be taken down very quickly in pretty much every country as soon as you either inform the webmaster or the authorities.
Slashdot Mirror
The only problem I have is that what do you do with a device that will be 10x13, cannot bend, and costs $300?
You stuff it in your bag where all the classic paper based stuff is? And about the price, a while ago somebody did a little calculation, result was that a one year subscriptions to the NYTimes could get you a Kindle for free, as most of the money is spend on delivering the paper.
Judging from all the eInk stuff I have seen, not any time soon. The refresh times are still measured in seconds and still have that "invert the screen" thing going on. I have seen laser printers that could print faster then those eInk things can refresh.
However outside of eInk based displays there is PixelQi who have build the OLPC screen and are now building stuff normal computers. Judging from the OLPC display, they certainly have the potential to rival eInk based stuff. The OLPC display is 200dpi, sunlight readable and has refresh times that are perfectly fine for watching video, where the OLPC display fails is at indoor readability without the backlight, as you need lots of light to see something on the screen. The display is also not white, but gray, so it reads more like recycling paper. However its still a really good display for reading books on it and its already a few years old, so they had plenty of time to improve upon it.
While it would be true that it would give you a percentage of actual usage, it still wouldn't be a valid estimate of actual Linux market share, as a lot of desktop users have Linux, but don't use it as their primary OS, instead they dual boot into it. I wouldn't be surprised if the actual usage count is at 1%, while the install base is more around the 10%.
The problem is that this only works if the information is valuable enough for people to actually copy it. With by far the most information on the net, that just never happens. And even if you copied it, you don't have any way to check that its authentic and not a manipulated copy and neither do you have a way to keep your copy online for the general public, as DMCA takedown notices will make sure that the information can only be found in obscure places.
I think the biggest problem with the Internet is the lack of unalterable static publications. You don't publish a thing on the Internet and then move on. You publish it, apply some bug fixes later, alter this and that, move it over to a new framework years later and so on. Things on the Internet just aren't static enough for proper archival, webpages evolve over time and even if you would do daily backups of webpages, you would miss plenty things. Things might look different when there would be a way to directly access the databases driving the webpages, but most of the time thats not the case. Even popular "open" projects fail at this, Wikipedia for example has all the history of edits stored and viewable, but they don't have that for deleted articles, those are purged and unviewable by the general public.
As I keep pointing out on each IPv6 story, there will be little motivation to move to IPv6 until you can hit major sites, like cnn.com and slashdot.org, using nothing but IPv6 packets.
That is never gonna happen, because enabling IPv6 for major sites has zero advantages and will break plenty of stuff. With IPv6 it isn't a "you have it" or "you don't have it" thing, for lots of consumers its a "you kind of have it, but routing is pretty broken" thing, meaning half the servers out there will not be reachable over IPv6, even if both parties have IPv6. And since plenty of software out there does the "clever" thing to default to IPv6, this means servers that worked before on IPv4 will no longer work once they have a IPv6 AAAA record or will only after a long timeout. Result of all this is that nobody is using AAAA records for their main DNS names, i.e. you get an AAAA record for ipv6.google.com, but not www.google.com. This kind of connection trouble of course would quickly fade away when IPv6 is actually used instead of just an experimental toy thing, but for quite a while peoples only IPv6 access will be via broken 6to4 tunnels.
Or to look at it from a different angle, there simply is no need for IPv6 for major sites, as there are so few of them that they aren't running out of IPv4 addresses anytime soon, they simply pay premium as soon as they get really rare.
The only way I can see IPv6 moving forward is if you start with it on the client side (torrent, ip telephony, whatever), since client to client communication is what is truly broken in IPv4 due to all the NAT'ing and other hacks and using IPv6 could make things there better instead of worse, since the status quo with IPv4 is already completly awful. And of course we already ran out of IPv4 addresses on the client side a long time ago, which is why the whole NAT thing become necessary in the first place.
In the end I however wouldn't expect miracles, this isn't something that is going to move much forward by free market forces anytime soon, this is something that could need a serious government push to get momentum. Or if that doesn't happen, I guess we have to wait till China, India and all the other counties see benefits of IPv6, before somebody here realizes that it might be a good idea to start with it too.
Any kind of bitmap scaling is going to look ugly for small non-integral factors (like 1.5x), which are typical when you zoom. The only thing that would solve this is adoption of SVG.
An easy fix would be to provide highres images and then scale those down, instead of providing lowres images and then trying to scale them up, but for that to work without wasting bandwidth you would need an image format that supports something like mipmaps and allows to only download the lower res part. Given how long it takes for an image format to be adopted, I consider that however unlikely to happen in the near future.
For the most part, it's the problem with the way specific sites are designed, not with browsers themselves.
Given that close to all non-trivial pages break, I doubt it. I think there are two core problems, one is that image size continue to be given in pixel instead of 'em', as most browsers are pretty bad at scaling images, thus the page layout stays rather inflexible. The other problem is that CSS doesn't have a proper way to know the size of the content its formating, so every specification of size in the CSS makes assumption about the content that no longer hold true when the font size changes, leading to overflowing boxes and all kinds of misery. Which of course relates to the problem of CSS based layout being extremely hacky compared to past table based layouts. With table based layouts its completly trivial to place to boxes horizontally next to each other with zero risk of accidental overlap, with CSS based layouts on the other side its a complete crazy hack-fest of overlapping boxes, hard coded margins and a whole bunch of other illogical crap and that is even considered "good practice", *yuck*. I would wish that CSS would allow to define layout tablets in the CSS files, thus getting rid of all the crazy hackery and still having that semantic/layout separation.
I don't see how they could possibly do that while remaining conformant with HTML and CSS. If the latter requires text layout to produce overlapped text, then that's pretty much it.
Well, the whole point is that it shouldn't be conformant, as neither of those standards guarantees good readability. I want a button that toggles between standard conform layout and readable layout. Look at the Readability script, its truly awesome, it strips out all the navigation bars, advertisements and all other distracting crap, leaving only the raw article content of a webpage, well formated with proper margins, font size and all. Its not perfect, as it doesn't work with all webpages, but when it works it improves the readability of a webpage a hell of a lot. The whole crux with browsers is that they are only good for viewing webpages, not reading them.
Instead, how about educating the web designers to use proper reflowing layouts?
The only "proper" layout is that decided by the user.
I really couldn't care less. Webbrowser these days seem to try everything to get pixel perfect rendering done, yet utterly fail at producing good looking readable webpages when there is even the tiniest deviation from the default. Try browsing with a larger default font for example, 99% webpages break, some worse then other, but pretty much all of them break. On Slashdot for example the "Reply to This" button falls apart on other webpages you are confronted with overlapping text and other unusable crap. And before somebody mentions the zoom feature, Firefox under Linux doesn't doesn't do any filtering when scaling, so all graphics look complete shit when zoom is used, making zoom unusable. There is other stuff that is annoying, for example the lack of build in support for link tags introduces in HTML2, you can get support via a plugin, but it would be nice to have solid support for that feature out of the box, maybe webpages would then finally start using it. But the most annoying thing is probably the lack of alternative view modes, I would like to have a modes that do not conform to pixel perfect rendering, but instead focus on producing readable results, i.e. avoiding overlapping text, making sure that line-width isn't to large, hide the navigation bars and all that other stuff, yet all the browser offers is pixel perfect rendering and rendering with no style sheets at all, neither of which is very readable. Luckily there is Readability which helps a good bit with that, making sure line-width is proper and navbars are gone, but again, it would be nice to have such basic stuff build into the browser.
The obsession with pixel perfect rendering and the complete ignorance on readable results is truly annoying and goes against anything that was considered "good practice" in the good old days.
Whereas people like me are advocates of just scrapping the whole damn thing because the potential benefits of doing so are way more interesting
What would be these "potential benefits" compared to a moderate copyright law (aka. 10-20 year term, allow non-commercial use)? The only difference I see is that no copyright would make it trivial for cooperations to exploit the authors even more, by taking the authors work, slapping their company name on it, changing it to fit their need, distributing it via their channels and all that without the author ever seeing a dime. I am pretty sure that that would piss of quite a handful of artists and not exactly be a good way to increase creativity.
Slashdot eating my <'s, that should have been of course:
% mplayer - < your_favorite_movie.avi
None, nada zilch. No exceptions, no excuses. Default policy "Deny all".
That would be a bad idea, as it would lead to a lot of annoying messages poping up, doing nothing but training the user to ignore them and click them away quickly, as they are clearly useless. If you want warnings and user confirmation you have to make sure that they happen very seldom so that the user will actually notice them. The way that the OLPC/Bitfrost does it is by defining security categories and then defining combinations that are forbidden/require user intervention, i.e. you can have Internet access without warning, but if you do that you can't get file access or access to the web camera. On the other side you can have webcam access, but then no Internet access. That way a lot of applications can work just fine without having a confirmation dialog, as the access they have can't do harm.
Yes, there is an assumption that the user is in control of their applications, and I, for one, like it.
The user isn't in control of its applications, he can start and stop them, everything that happens in between is up to the application and the user doesn't even have a proper way to verify what is happening, as there is no log kept (well, you could strace, but thats impractical when applied to all applications).
Right. Until you get the first user who wants to upload a picture with their secure browser. And then you get the second user who wants to open their favorite movie with mplayer, because WMP choked on it. And you don't get a third user, because your system is unusable.
As mentioned before, that would be absolutely no problem in a sandbox, as you would simply pass the data to the application directly or 'mount' the requested data into their namespace, no need for the application to have full system access. Need an example? How about:
% mplayer - Ever seen a serious Flash application? No? You know why? Because you can't do anything interesting in a sandbox.
There have been tons of decent games written in Flash, I don't see why a game should have more rights to mess up my system when it comes as .exe instead. You are seriously attacking the thing from the wrong viewpoint. Its not about all the weird edge cases that might be troublesome (no doubt there are some, even so you haven't yet really named one that isn't trivially solvable), its about the 99% of applications that would work just fine.
If the user is tricked into installing crap, clearly they shouldn't have root.
You don't need to be root to install crap, that works just fine as a normal user and can do just as much harm on your regular single user desktop system.
If they're bribed, why should the computer refuse
The whole point is that the computer shouldn't refuse, instead it should isolate the applications so that they can't do any serious harm, even if they are evil. The whole trouble today is that applications simply aren't isolated enough from each other. Each application has full access to all the files and devices that the user has access to, which is far more then is needs and a clear violation of 'least privilege' principle. Most applications would work just fine if they would only have access to a place for configuration files and their own data files and nothing else. They wouldn't even need to have raw file system access if they would want to save things outside of those locations, as load/save could be handled by the OS, i.e. instead of having the application present the load/save dialog, its the OS that displays the dialogs and reads the files, all the application gets to see is the data, not the filesystem, thus ensuring that it can never overwrite on its own.
Or to attack the thing from a different angle, what would you say if a Flash application had full system access? Kind of a bad idea, don't you think? Why would you want far less security when a random game comes as .exe instead of Flash app on a webpage, why shouldn't both use cases, which are pretty much identical, have the exact same tight security applied?
If the "owner" or "user" of the computer is tricked, bribed or forced to install such malware, what computer is there that will protect itself?
OLPC with Bitfrost will do exactly that just fine. Just because most other OSs don't even try to prevent those issues doesn't mean you can't.
The other is user security. And you cannot solve that problem with technology.
You can't solve all user security issues with technology, but you can solve a hell of a lot then todays OSs try. The whole insecure password thing could be easily fixed with smardcards/usbstick and pincodes, by far most applications could be run in isolation not having full system access and the filesystem could provide versioning so that even if things would go wrong, you could roll the changes back and get a look of what happened in the first place. None of them are rocket science or especially expensive and all of them make the system *more* convenient to use, not less. This whole "blame the user" business is in large part just an excuse for bad engineering. Its like the whole phising thing, yeah, users shouldn't click on everything, yet I have yet to see a single mail from a larger company that actually signs there mails cryptographically, which would make automatic phising detection trivial without the user having to lift a finger.
And yet the most common security issue is some good old buffer overflow that could trivially be fixed by using another language then C.
Looking up trivial facts in a database and spitting them out is easy, and not particularly significant...
The significant part here is that the database queries will be in natural language instead of SQL and they will be kind of vague. Which is kind of a big deal, considering its one of those areas where pretty much all search engines these days fail at, i.e. you can't ask questions to Google and get an answer, instead you simply do full text search and get all pages matching, no matter if they have anything to do with your question or not.
Both can and will survive,
I have some doubt about that. Look for example at the flightsim genre. That genre was pretty hardcore and popular a decade ago and is now pretty much completly dead, no release in years and nothing announced or the future. The few games still around that involve planes (AceCombat, Hawks) are an utter joke compared to a decent ten year old flightsim, as they don't have a drop of realism left. The thing to realize is that todays "hardcore" is already really soft and mass market oriented, compared to games of the past.
That said, I don't think that every game will be like Wii Sports in the future, as there is a big enough market share for Hollywood style over the top action games, but they will be or already are, rather watered down experiences lacking in depth and that trend will only continue, since it simply sells more copies.
This really isn't an either "this or that" thing, you really want both elements. Games that force you to do things because the story dictates it suck, but so do games that don't have any story at all.
What I as a gamer want are interactive worlds that I can explore without hitting an invisible wall and that have enough simulation in them that "story" can evolve on its own without every bit being prescripted. Prescripted events in fact are the biggest problem I have with games today, you can take most FPS and completly break them by not following the predefined route. Take Killzone2 or Call of Duty 4, what happens when you don't follow your team mates and instead turn around and explore the area? Absolutely nothing. The war just stops and waits for you to touch the next trigger, if you are not there, nothing happens, the area doesn't offer anything interesting to explore. Compare that with a Operation Flashpoint or a EF2000, those games feature dynamic missions without trigger spots, the war doesn't wait for you, it just continues even if you are not there. The events that evolved from those dynamic gameplay where *far* more memorable then anything I have ever seen in a scripted shooter, because those events where real, not the magic hand of the designer fucking with virtual reality.
Now that said, meaningful choices and believable characters are of course extremely important, but the way to reach that goal isn't by a completly linear path for the player to follow, its by creating a flexible enough virtual environment that allows meaningful interaction with the gameworld and thats exactly where most linear games fail at.
The thing with WoW is that while its a total time sink, its also at the same time a very non-demanding game. You don't really have to use your brain to make progress in WoW, you just have to walk around and click on things. There are no puzzles to overcome or insane tactics to master, its just a matter of clicking on stuff. It might get quite a bit more hardcore at the higher levels, but before a casual gamer reaches them he will have to sunk in hundreds of hours of very casual play into the game.
If you can believe vgchartz.com there are 51 million hardcore consoles (PS3, Xbox360) around, while "only" 49 million casual gamer consoles (Wii). Seems to me to be an even match, not exactly what I would call death of hardcore gaming.
They have established a totally free online encyclopedia. No ads. They have had to balance quality with quantity.
A lot of what they accomplished, they accomplished before they turned into an elitist club. Now they seem to be more focused to preserve what they have, then to broaden their scope, which of course pisses of all people that actually want to add to it.
They have established rules that define what is encyclopedic.
Well, yeah, there have to be some rules, nothing wrong with that in principle, the trouble is that some of the rules are idiotic and instead of giving an article a chance, they now delete every border case. On the positive side of things however the English Wikipedia is still lightyears better then the German one, I consider the German one pretty useless these days, because it almost never has the in depth information I am looking for.
Soon we no longer need actors and we just need digitized versions of them.
We will have digital version of them, but we still need actors. The whole human digitalization so far is in very large parts based on motion capture. If you don't have an actor, there isn't any data to render. Even completly hand animated stuff, like Pixar movies, still have a human voice actor for reference. So far the whole digital side of things has done *nothing* in reducing the need for actors, it has done quite the opposite, you now not only need actors, often you have multiple onces for different tasks in a single role, you also need graphic artists, modeler, programmer and all that stuff to drive the digital puppet.
One day we might reach a point where we can generate the actual acting, but we are still very far away from that point. There really hasn't even been much research into that area aside from really basic stuff (i.e. walk cycles and such). What you however don't need in the future is a specific actor, since you can just capture a similar one and transfer his movement onto the other one.
The only way to make a blacklist work is to not do one. Blacklists are completly pointless for child porn, because its *ALREADY* illegal in pretty much any country that happens to have Internet. If you find child porn you can take down the server and be done with it, leaving no easy way around it. On the other side if you block it via a stupid blacklist you make sure that it stays available for everybody actually interested in the topic, since the filters are trivial to work around.
A Youtube video of the Uwe Boll movie Seed is blocked via that list as well, no kidding. The ridiculous part of those list is that they are completly unneeded, when there is real child porn on a server it will be taken down very quickly in pretty much every country as soon as you either inform the webmaster or the authorities.