Now tell me, why should any sane person attempting to develop open solutions should use mono or any other technology aping (how apt) MS's closed, most likely patented implementations of any technology?
Good point! I'd better go patch out the MS Word support for OpenOffice.
There's also two other issues here: Some countries do not allow software patents. For the rest of us, there is still the question of "What's a sane alternative?" No matter where you go in the software industry, you'll be running into patents.
All that said, I do actually agree that it's maybe not the safest move, and that I would much rather start from scratch.
Oh, on a related note: Remember the whole GIF controversy? For quite a long time, the only reasonable alternative was to use JPEGs everywhere, because it was either GIF or JPEG. It took a long time for PNG to be widely supported enough to be a replacement for GIF, and various ways of animating PNGs aren't really officially standardized, and are certainly not commonly supported.
So, at a certain point, you have to ask yourself if you'll actually have a completely open replacement created by the time the patent runs out.
If you advocate the use open source operating systems, that's great and if you want to educate people about DRM and advocate against it, that's also great but stop thinking the first solution is a practical solution to the second problem.
It is far more an ideological solution than a practical solution.
But imagine, if you will, that some 20% of the market (being extremely optimistic) has switched to open source software because they actually care about having that kind of control. It seems incredibly unlikely to me that these people would allow any kind of DRM on their system, and that they would, in fact, take steps to prevent it.
There is also the issue that quite often "impossible" in software merely means "very, very hard", or even "statistically unlikely." (Example: cryptographic signatures.) So, you're right in that it's possible, but I do maintain that it would be harder -- Microsoft is helping implement DRM, and is certainly doing nothing to prevent it.
Oh, one more thing: It's quite possible that this debate is being confused by people who run a free OS, but not 100% free software. The latter is pretty much immune to DRM, as any DRM scheme would thus have to be implemented in free software.
The former becomes difficult to define. Am I running a Free OS, but with some proprietary software, if I use binary kernel modules? That would seem to suggest that anyone with an nVidia driver is already running a non-Free OS.
No. The reason why people celebrate is the basis for "reason for the season." Which is, of course, the birth of Christ.
Nope, we hadn't established that. (And round and round we go.
Let's look at that statement again:
The reason why people celebrate is the basis for "reason for the season."
I don't dispute that at all. I also don't see the relevance of redefining it that way. If I didn't already acknowledge the birth of Christ as the "reason for the season", why would I suddenly acknowledge it as the "reason why people celebrate"?
I'd be stunned if the government did anything to fix the web-media-doesn't-have-captions problem in anything less than a decade
That is the fault of the government, then, not the web media.
I also honestly fail to see how a telephone captioning service can put captions on a video screen.
It was an example of how the government can help with this sort of thing.
Such as, there's no way to transmit closed caption data over HDMI
So your TV won't be able to do the subtitles directly, but I don't really see the point of that. Games and movies have their own subtitles long before it hits the HDMI link.
I can't connect an HDMI upscaling DVD player to my HDTV and expect to see captions because of this lovely piece of stupidity, for example.
Out of curiosity, how is closed captioning done on DVDs, other than through subtitle tracks? (And more importantly, why are they done other than through subtitle tracks?)
A small company (Art Lebedev) scrapes together some money to produce the most expensive keyboard on the market (hell, who could reasonably justify $1500 for a keyboard?), and then decide to forgo the one market (Linux)
With that price point, why would they forgo a market if they don't have to? Is it really that hard to develop cross-platform USB drivers? (Hint: From what I understand, it is harder to go from Windows to OS X than from OS X to Linux.)
You want free GPL drivers to run your $1500 keyboard on Linux?
Surely $1500 should pay for those drivers, right?
Oh, by the way: I would easily pay that much for a good keyboard, if it weren't for the fact that I already have one. I develop software for a living, so, Linux geek or not, my keyboard is pretty central to my livelihood.
I like Apple's keyboards right now. I haven't really been able to find a faster (to type on), quieter keyboard yet. Easier to clean, too.
The problems I have are all related to the funny layout Apple's got. The "Super" key is where the alt key should be, so I have to swap those in a keymap -- which isn't working flawlessly, yet, and is a pain on my laptop, where the only way I know of messing with keys like that requires a reboot (or logout/login) to take effect. This means I can either have the Apple keyboard work, or the built-in keyboard, but not both at once. So, when switching keyboards, I pretty much have to reboot.
That's just the beginning. All the F-keys are slightly off, ending with F12 -- they've been shoved left to make room for an eject key (which all my OSes currently ignore -- eject which drive, now?) -- followed by F13, F14, and F15 instead of PrintScreen, ScrollLock, and Pause. At least they were nice enough to throw in an extra F16, F17, F18, and F19, where normal keyboards have capslock/numlock/etc LEDs. (This thing has a Capslock LED, on the capslock key, which is cool, but it has no LEDs for ScrollLock or NumLock.)
Instead of NumLock, there's Clear, which (finally, a break!) is the same keycode. But then, going clockwise around the numpad, there's =,/, *, -, +, and enter, where the standard is/, *, -, +, and enter. This is especially frustrating as the keypad plus is half the size it should be, and the minus takes up the other half of the same space.
The worst part, though, has to be Insert. Home, Pageup/Pagedown, Delete, and End are all where you expect them to be, but Insert? No, you get fn. And from what I understand, that fn actually does expose the fancy features that I see on those F-keys, like brightness, expose, dashboard, fastforward/rewind, play/pause, volume/mute. None of which really do anything right now.
Not that I would mind mapping them, but I basically have no option to put insert where I think it's supposed to go, and I do actually use it (or did).
I suppose it would be the perfect keyboard if I was on OS X, and actually, there's very little I physically dislike about it. Given enough time and patience, I'm sure I could remap everything except that fn key. The reason I'd want a keyboard which can physically show me the keymap is that I strongly doubt they'd be stupid enough to make a fixed fn key on that. It would also be especially cool if I could have the mapping stored in the keyboard itself, so that I don't have to teach my various OSes to flip between keymaps when I change keyboards.
It's been a meme for awhile now. You know, the pre-emptive "Watch the fanboys defend..." and "Imagine if (Microsoft|Sony|MPAA|Bush) did this, what a shitstorm there would be!"
Judging by the comments on this thread, there are a lot more people whining about fanboys than actual fanboys.
I've read your description of the "halting problem", which is fairly interesting, but it doesn't save you here.
Put simply, let's suppose you do prove that it's a simulation. You write a paper about it, and you publish it in a major scientific journal.
Fine, then the simulation notices. It subtly alters your results, inserting fnords (which really work, as they can directly control anyone's brain) into every published copy, and altering everyone's memory to suggest that your experiment had either failed utterly, or proved conclusively that the Universe was not a simulation.
That's actually more complex than they'd have to -- simply swoop in at the last second and change your results.
So, it's impossible to prove that the universe is not a simulation, because if it were a simulation, all "proof", in any form, is suspect. That's assuming the physics and math involved is sound.
Now, is it possible to prove that the universe is a simulation?
Depends on your definition of "simulation". After all, if you saw a character come out of the sky claiming to be the avatar of the Universe's programmer, that would be proof that either the Universe is a simulation, or that you are insane. But insanity, and dreams, could be described as a kind of simulation.
But I kind of doubt you could find any other proof. Wouldn't it always be possible to find another theory? For example: Suppose you claimed the movement of Mercury, being so unpredictable with regards to Newtonian physics, was "proof" that the Universe was a simulation. Well, it certainly proves Newton wrong, but we now have General Relativity.
So, in that sense, you can't poke holes in reality simply by finding something that doesn't make sense. You also have to have a corresponding theory which does make sense, and which makes testable predictions, and which is generally described mathematically.
That last part -- I don't have enough computer science theory to be sure, but I don't see how you can express "this is a simulation" mathematically.
One point which has been made is that the Classmate is direct competition. To rip off someone else's example:
OLPC discovers a particular area might be in the market.
Intel rep hears about this at the board meeting.
Intel rep phones home.
Intel parachutes in with free ClassmatePCs.
Essentially, Intel gets to rip off all of OLPC's market research, for lack of a better term.
That's ignoring, of course, the more tinfoil-hattish version: Classmate is Intel and Microsoft's response to OLPC, out of fear that an entire generation of children might grow up without them. Therefore, it's really got nothing to do with the children, and the design shows -- the XO is vastly superior for what it's designed to be.
Or, if you're too lazy, I'll spell it out for you: A blind bus driver is actually blind. A gay person may be slightly more likely to get AIDS, but not all gay people have it. And there are other "lifestyle choices" that are actual choices, and actually contribute a good deal more -- like drug use involving dirty needles, or swinging without adequate protection... (Yes, there are monogamous gays. Shocked?)
The keyboard is unusable. Maybe it works for kid-fingers, but don't expect any third-worlders to be cranking out code on these.
You're assuming kid-fingers won't be cranking out code.
No development tools.
Except for a magic key which shows you the source of the current app, lets you modify it right there...
Speed. Bad. No nukes will be designed with these babies.
Define "bad". I'll bet nukes were designed on far less powerful machines. And what were you doing with it that needed speed?
Wireless. Hopeless. It doesn't support a simple WPA connection out-of-the-box. Maybe there's something to the mesh idea, but without a flock of them, it's difficult to experiment.
Well, you answered your own point there. It's not like there's going to be tons of WPA-secured wifi networks in Uganda for these kids to hook up to.
Boot time. Horrible.
I'll take your word for that.
Screen. OK. It's kind of interesting the way it can shift from color to monochrome.
And that monochrome has 3x the resolution. And that it's viewable in bright sunlight. Neither of those seemed impressive for you?
Design. Toy-like. Yes, it's meant for children, but this pre-school look is dorky.
So what?
Yes, design is what they'll be complaining about when they get the first computer that their village has ever seen.
Sorry to be so negative, but a cheap cell phone could compete with this as far as functionality goes.
A cheap cell phone would have a smaller screen, smaller keyboard, not necessarily have a cell tower to connect to, absolutely not have mesh networking, nor the magic "source view" button or the social networking stuff... And of course, when the cell battery dies, they'll just plug it in, right? (Hint: XO has hand cranks designed for it.)
Now, a cheap cell phone could be designed to be like this... but there's still the problem of it being too small, and if you make it bigger, guess what? You've got the XO.
So far, your one point that I can't really argue with is boot time, and really, why does that matter?
Sounds to me like you had high expectations for a laptop for yourself. I was actually amazed at how much more usable it is to the actual kids it's aimed at than I was expecting.
With nspluginwrapper (configured automatically on OS install) Flash Player 9 works as expected in 64-bit firefox.
I wonder, can anyone with a 32-bit system benchmark this for me? I don't know if nspluginwrapper was the reason for the performance hit, but it was amazingly slow. Not noticeably, until you look at top. The same video on mplayer used 2% CPU at the very most, fullscreen, but with Flash, it was over 50%, windowed.
Oh, and my Firefox crashed when I did that. A LOT.
Sorry to be a dick, but how many people have to suggest this as ActiveX, and then get the concept of a sandbox hammered through their head? Silverlight offers no more access to the local machine than Java, or, arguably, JavaScript or Flash. (Arguably less -- Flash has webcam support, and I'm not sure Silverlight does.)
It has more in common with current iterations of Flash than with ActiveX.
1. if you become an addict, you screw up my life as well as tyour own, because now i have to feed and clothe a zombie
Or you could just let him die.
It's like requiring people to wear bicycle helmets. If I don't want to, and I bash my head in, that's my fault, and the government should not have to pay for my hospital bill, or my funeral. I'd much rather have choice than a nanny state safety net that limits choice.
however, substances that are both highly incapacitating AND highly addictive at the same time have negative effects so viral and devastating that the legalization of these substances (heroin, coke, meth) outweigh all the negatives of prohibition.
I'm sorry, but I'd rather have crackheads than the Mafia.
Oh, and coke? Guess what? It is still an ingredient in Coca-Cola, and coca tea is an amazing natural remedy for altitude. Seriously, if people had caffeine in the concentrations they have coke, wouldn't it be just as dangerous?
Drugs, by themselves, are not dangerous. They're just chemicals.
You seem to be ignorant of the fact that the specs for both Silverlight and Flash specs are available.
You seem to be ignorant of what the terms of the Flash spec are. I'm not sure about Silverlight, but you are not allowed to use the Flash spec to develop a competing client.
Dog slow also means absolutely nothing, when it comes to desktop apps.
In fact, there's two major areas in which this no longer matters: Places where the basic logic is in Javascript, and that basic logic is not CPU-intensive (sorry, your Excel spreadsheet does NOT count as CPU-intensive), and places where it's cheaper to throw CPU at the problem than it is to throw developers (Ruby on Rails, especially if you can cache the bigger things to a static server -- Rails on Amazon EC2 with S3 as a cache just rocks).
There are, of course, places where this does matter. I'd never write a video codec in Javascript, but fortunately, I don't have to -- in the case of Flash, Adobe already did that for me, probably in C or C++. HD-DVDs use Javascript, and Blu-Ray uses Java -- guess which one is faster? (My guess -- Blu-Ray is forcing Java to do the animation itself, while HD-DVD passes the physical animation off to hardware -- I just do foo.animateProperty(property, values, duration) and watch it go.)
And after all that, Javascript interpreters are getting faster.
Now, I actually agree with you. I can't wait for the Spidermonkey/Actionscript fusion, which will mean being able to code to a bytecode engine (so Javascript, probably also Ruby, Python, whatever else) which will be included in Firefox, and also in Flash, meaning it'll be in IE also. And I don't mind Silverlight either, because in theory, there's IronRuby, IronPython, and so on. I'd much rather have a better tradeoff than straight Javascript, and more options.
But you're wrong on two counts: You should care about dog slow languages, and Javascript will not always be that slow.
When version 2, or 3, or w/e, of Silverlight comes out will they be releasing the Linux version too? Or do I have to wait for some nice Open Source chaps to figure out how it works then implement it themselves?
That's kind of what they did for 1. Well, they got help from Microsoft, but the project itself is open source.
Compare that to Flash, where if Adobe decides not to upgrade Flash, you have to start implementing Gnash from scratch. That's kind of what happened.
Yes I do believe HTML is more open the Silverlight or Flash.
It absolutely is, and it absolutely is not as powerful. You can't make YouTube in HTML (without Flash or Silverlight).
And it's hard to choose, but so far, Silverlight is the lesser of two evils.
Consistent would be a good start. I'm not keen on the idea of it being done by studios, though, as I neither know nor care which studio does which flick.
Not really a better way around it -- and they seem to really be wanting you to care. Universal has a brand and everything.
How about some standard human-interface guidelines so it's consistent across movies?
Well, it's at least consistent within the studio, as I said. Universal movies all have menus that look and feel exactly the same, down to the graphics and sound. (The menu makes a sound when sliding in.)
Warner, on the other hand, has basically the same UI on almost all of them, but skinned differently. So you might have that white-to-silver crap, but at least the layout would be predictable.
my gripe was really about when one goes do something that should be simple, like selecting a single episode of a TV show instead of telling it to play all.
That part is going to be about the same on HD-DVDs as it was on DVDs, except that the menu is going to be available with the movie playing, and the resolution does actually make UI design a lot easier.
Also, I'm guessing partly because of the cost, it is, again, consistent. Every single Heroes disc had the same menu to select episodes, and every other Universal disc had the same menu to select scenes.
One more thing: Just to see if I could, I coded a particular scrolling menu with a key buffer, which also notices the user mashing and speeds up the scrolling animation at that point. It's not going to be as nice as an iPhone, or even a traditional scrollbar, but it's about the best we can do with a remote. (Holding the button down cannot work.)
However, I do freely admit that the platform makes it possible to absolutely torture you. We could make you sit through commercials, we could do five second long animations (or thirty seconds), could chain multiple animations for simple tasks, and could force you to register before even watching the movie. But I, personally, refuse to do that.
In many states, CapTel equipment is provided free or at a reduced rate to people with hearing loss.
In other words, get the government to pay for it.
it cannot be a windows-only solution.
Agreed. All of the technologies I listed will work on VLC and mplayer, on Linux and Mac, and on Windows, if you insist.
wheelchair users aren't charged for ramps, elevators, etc.
But not all buildings are wheelchair accessible.
And it's nice that some random fan production has captions.
Erm, look again. It is an A-list team, making a production exclusively (so far) for the Internet -- probably a direct result of the writer's strike. There are multiple fan groups making subtitles for this show. (Not full captioning, I'd assume, as in, no sound effects as text, but speech at least -- and in at least five or six languages so far.)
To be absolutely clear: The show is commercial, professional quality. The subs are done by fans -- fansubs, if you will.
Throwing me a total red herring that I have no interest in does not solve the problem.
Well, obviously not enough interest to even read the website. Sorry to have wasted your ridiculously precious time.
Um. Wow, I guess my point went completely over your head, because Christmas is evidence supporting my point.
Ditto. You don't see the Yule culture in Christmas?
(If you think that what I just said sounds reasonable, then you are beyond reason.)
No, I think what you said was pointless and offtopic.
What people currently care about is the only basis for "reason for the season"? I guess that circles back to the capitalism point. People currently care about getting and giving presents.
completely ignoring (and seemingly ignorant of) the fact that DRM can be implemented in ANY operating system with or without cooperation from the authors of the OS.
That part is not true.
At least on a Free(TM) OS, more Free than Linux currently is, it would not be possible to implement effective DRM, because the user would effectively have control over everything the application has access to. This means that, for instance, you could always run it in a virtual machine, record all traffic in and out of it (including to the pseudo random number generator library), and do a replay attack on it.
That's the more brute-force attack. The fact is, a rootkit should be much easier on Linux. Given the default policy of no root access and the sheer variety of kernels out there, there's simply far less that an app can be sure of about its environment, which makes it much more difficult to tell if that environment is "real" or "trusted". Most games which have been ported to Linux did not bother to port any of the CD-based copy protection, probably because they realized how insanely simple it would be for Linux people to implement an undetectable Daemontools.
With at least the major proprietary OSes, you'll first have to crack the DRM that's built-in to the OS -- convince it that it really is running on bare metal, or convince it to let you do that messing-with-the-IO trick.
So it doesn't completely solve the problem, but I do believe a free system is a lot more hostile, in practice and also in culture, to DRM.
I'm fully aware that Linux itself can have binary kernel modules, at which point, there's really no technological difference. But the cultural difference is important. Anyone switching to Linux is also going to be acutely aware of DRM, partly because things without DRM will work for them, and things with DRM won't (at least for now).
Amazon EC2, or S3 for static content. In fact, if you were fast enough, you could set up an HTTP redirect on your old server (assuming it can even handle that) to point to an S3 cache.
I'm sure there are plenty of others, though. Just telling what I know.
Slashdot Mirror
Good point! I'd better go patch out the MS Word support for OpenOffice.
There's also two other issues here: Some countries do not allow software patents. For the rest of us, there is still the question of "What's a sane alternative?" No matter where you go in the software industry, you'll be running into patents.
All that said, I do actually agree that it's maybe not the safest move, and that I would much rather start from scratch.
Oh, on a related note: Remember the whole GIF controversy? For quite a long time, the only reasonable alternative was to use JPEGs everywhere, because it was either GIF or JPEG. It took a long time for PNG to be widely supported enough to be a replacement for GIF, and various ways of animating PNGs aren't really officially standardized, and are certainly not commonly supported.
So, at a certain point, you have to ask yourself if you'll actually have a completely open replacement created by the time the patent runs out.
It is far more an ideological solution than a practical solution.
But imagine, if you will, that some 20% of the market (being extremely optimistic) has switched to open source software because they actually care about having that kind of control. It seems incredibly unlikely to me that these people would allow any kind of DRM on their system, and that they would, in fact, take steps to prevent it.
There is also the issue that quite often "impossible" in software merely means "very, very hard", or even "statistically unlikely." (Example: cryptographic signatures.) So, you're right in that it's possible, but I do maintain that it would be harder -- Microsoft is helping implement DRM, and is certainly doing nothing to prevent it.
Oh, one more thing: It's quite possible that this debate is being confused by people who run a free OS, but not 100% free software. The latter is pretty much immune to DRM, as any DRM scheme would thus have to be implemented in free software.
The former becomes difficult to define. Am I running a Free OS, but with some proprietary software, if I use binary kernel modules? That would seem to suggest that anyone with an nVidia driver is already running a non-Free OS.
Nope, we hadn't established that. (And round and round we go.
Let's look at that statement again:
I don't dispute that at all. I also don't see the relevance of redefining it that way. If I didn't already acknowledge the birth of Christ as the "reason for the season", why would I suddenly acknowledge it as the "reason why people celebrate"?
That is the fault of the government, then, not the web media.
It was an example of how the government can help with this sort of thing.
So your TV won't be able to do the subtitles directly, but I don't really see the point of that. Games and movies have their own subtitles long before it hits the HDMI link.
Out of curiosity, how is closed captioning done on DVDs, other than through subtitle tracks? (And more importantly, why are they done other than through subtitle tracks?)
With that price point, why would they forgo a market if they don't have to? Is it really that hard to develop cross-platform USB drivers? (Hint: From what I understand, it is harder to go from Windows to OS X than from OS X to Linux.)
Surely $1500 should pay for those drivers, right?
Oh, by the way: I would easily pay that much for a good keyboard, if it weren't for the fact that I already have one. I develop software for a living, so, Linux geek or not, my keyboard is pretty central to my livelihood.
I like Apple's keyboards right now. I haven't really been able to find a faster (to type on), quieter keyboard yet. Easier to clean, too.
/, *, -, +, and enter, where the standard is /, *, -, +, and enter. This is especially frustrating as the keypad plus is half the size it should be, and the minus takes up the other half of the same space.
The problems I have are all related to the funny layout Apple's got. The "Super" key is where the alt key should be, so I have to swap those in a keymap -- which isn't working flawlessly, yet, and is a pain on my laptop, where the only way I know of messing with keys like that requires a reboot (or logout/login) to take effect. This means I can either have the Apple keyboard work, or the built-in keyboard, but not both at once. So, when switching keyboards, I pretty much have to reboot.
That's just the beginning. All the F-keys are slightly off, ending with F12 -- they've been shoved left to make room for an eject key (which all my OSes currently ignore -- eject which drive, now?) -- followed by F13, F14, and F15 instead of PrintScreen, ScrollLock, and Pause. At least they were nice enough to throw in an extra F16, F17, F18, and F19, where normal keyboards have capslock/numlock/etc LEDs. (This thing has a Capslock LED, on the capslock key, which is cool, but it has no LEDs for ScrollLock or NumLock.)
Instead of NumLock, there's Clear, which (finally, a break!) is the same keycode. But then, going clockwise around the numpad, there's =,
The worst part, though, has to be Insert. Home, Pageup/Pagedown, Delete, and End are all where you expect them to be, but Insert? No, you get fn. And from what I understand, that fn actually does expose the fancy features that I see on those F-keys, like brightness, expose, dashboard, fastforward/rewind, play/pause, volume/mute. None of which really do anything right now.
Not that I would mind mapping them, but I basically have no option to put insert where I think it's supposed to go, and I do actually use it (or did).
I suppose it would be the perfect keyboard if I was on OS X, and actually, there's very little I physically dislike about it. Given enough time and patience, I'm sure I could remap everything except that fn key. The reason I'd want a keyboard which can physically show me the keymap is that I strongly doubt they'd be stupid enough to make a fixed fn key on that. It would also be especially cool if I could have the mapping stored in the keyboard itself, so that I don't have to teach my various OSes to flip between keymaps when I change keyboards.
It's been a meme for awhile now. You know, the pre-emptive "Watch the fanboys defend..." and "Imagine if (Microsoft|Sony|MPAA|Bush) did this, what a shitstorm there would be!"
Judging by the comments on this thread, there are a lot more people whining about fanboys than actual fanboys.
You'd think so, but actually, it wasn't Lisp.
I've read your description of the "halting problem", which is fairly interesting, but it doesn't save you here.
Put simply, let's suppose you do prove that it's a simulation. You write a paper about it, and you publish it in a major scientific journal.
Fine, then the simulation notices. It subtly alters your results, inserting fnords (which really work, as they can directly control anyone's brain) into every published copy, and altering everyone's memory to suggest that your experiment had either failed utterly, or proved conclusively that the Universe was not a simulation.
That's actually more complex than they'd have to -- simply swoop in at the last second and change your results.
So, it's impossible to prove that the universe is not a simulation, because if it were a simulation, all "proof", in any form, is suspect. That's assuming the physics and math involved is sound.
Now, is it possible to prove that the universe is a simulation?
Depends on your definition of "simulation". After all, if you saw a character come out of the sky claiming to be the avatar of the Universe's programmer, that would be proof that either the Universe is a simulation, or that you are insane. But insanity, and dreams, could be described as a kind of simulation.
But I kind of doubt you could find any other proof. Wouldn't it always be possible to find another theory? For example: Suppose you claimed the movement of Mercury, being so unpredictable with regards to Newtonian physics, was "proof" that the Universe was a simulation. Well, it certainly proves Newton wrong, but we now have General Relativity.
So, in that sense, you can't poke holes in reality simply by finding something that doesn't make sense. You also have to have a corresponding theory which does make sense, and which makes testable predictions, and which is generally described mathematically.
That last part -- I don't have enough computer science theory to be sure, but I don't see how you can express "this is a simulation" mathematically.
One point which has been made is that the Classmate is direct competition. To rip off someone else's example:
Essentially, Intel gets to rip off all of OLPC's market research, for lack of a better term.
That's ignoring, of course, the more tinfoil-hattish version: Classmate is Intel and Microsoft's response to OLPC, out of fear that an entire generation of children might grow up without them. Therefore, it's really got nothing to do with the children, and the design shows -- the XO is vastly superior for what it's designed to be.Wow. Go learn something about AIDS.
Or, if you're too lazy, I'll spell it out for you: A blind bus driver is actually blind. A gay person may be slightly more likely to get AIDS, but not all gay people have it. And there are other "lifestyle choices" that are actual choices, and actually contribute a good deal more -- like drug use involving dirty needles, or swinging without adequate protection... (Yes, there are monogamous gays. Shocked?)
I have to ask:
You're assuming kid-fingers won't be cranking out code.Except for a magic key which shows you the source of the current app, lets you modify it right there...
Define "bad". I'll bet nukes were designed on far less powerful machines. And what were you doing with it that needed speed?
Well, you answered your own point there. It's not like there's going to be tons of WPA-secured wifi networks in Uganda for these kids to hook up to.
Boot time. Horrible.I'll take your word for that.
And that monochrome has 3x the resolution. And that it's viewable in bright sunlight. Neither of those seemed impressive for you?
So what?
Yes, design is what they'll be complaining about when they get the first computer that their village has ever seen.
A cheap cell phone would have a smaller screen, smaller keyboard, not necessarily have a cell tower to connect to, absolutely not have mesh networking, nor the magic "source view" button or the social networking stuff... And of course, when the cell battery dies, they'll just plug it in, right? (Hint: XO has hand cranks designed for it.)
Now, a cheap cell phone could be designed to be like this... but there's still the problem of it being too small, and if you make it bigger, guess what? You've got the XO.
So far, your one point that I can't really argue with is boot time, and really, why does that matter?
Sounds to me like you had high expectations for a laptop for yourself. I was actually amazed at how much more usable it is to the actual kids it's aimed at than I was expecting.
I wonder, can anyone with a 32-bit system benchmark this for me? I don't know if nspluginwrapper was the reason for the performance hit, but it was amazingly slow. Not noticeably, until you look at top. The same video on mplayer used 2% CPU at the very most, fullscreen, but with Flash, it was over 50%, windowed.
Oh, and my Firefox crashed when I did that. A LOT.
Yes, you're totally right.
Except you're not.
Sorry to be a dick, but how many people have to suggest this as ActiveX, and then get the concept of a sandbox hammered through their head? Silverlight offers no more access to the local machine than Java, or, arguably, JavaScript or Flash. (Arguably less -- Flash has webcam support, and I'm not sure Silverlight does.)
It has more in common with current iterations of Flash than with ActiveX.
Or you could just let him die.
It's like requiring people to wear bicycle helmets. If I don't want to, and I bash my head in, that's my fault, and the government should not have to pay for my hospital bill, or my funeral. I'd much rather have choice than a nanny state safety net that limits choice.
I'm sorry, but I'd rather have crackheads than the Mafia.
Oh, and coke? Guess what? It is still an ingredient in Coca-Cola, and coca tea is an amazing natural remedy for altitude. Seriously, if people had caffeine in the concentrations they have coke, wouldn't it be just as dangerous?
Drugs, by themselves, are not dangerous. They're just chemicals.
You seem to be ignorant of what the terms of the Flash spec are. I'm not sure about Silverlight, but you are not allowed to use the Flash spec to develop a competing client.
Dog slow also means absolutely nothing, when it comes to desktop apps.
In fact, there's two major areas in which this no longer matters: Places where the basic logic is in Javascript, and that basic logic is not CPU-intensive (sorry, your Excel spreadsheet does NOT count as CPU-intensive), and places where it's cheaper to throw CPU at the problem than it is to throw developers (Ruby on Rails, especially if you can cache the bigger things to a static server -- Rails on Amazon EC2 with S3 as a cache just rocks).
There are, of course, places where this does matter. I'd never write a video codec in Javascript, but fortunately, I don't have to -- in the case of Flash, Adobe already did that for me, probably in C or C++. HD-DVDs use Javascript, and Blu-Ray uses Java -- guess which one is faster? (My guess -- Blu-Ray is forcing Java to do the animation itself, while HD-DVD passes the physical animation off to hardware -- I just do foo.animateProperty(property, values, duration) and watch it go.)
And after all that, Javascript interpreters are getting faster.
Now, I actually agree with you. I can't wait for the Spidermonkey/Actionscript fusion, which will mean being able to code to a bytecode engine (so Javascript, probably also Ruby, Python, whatever else) which will be included in Firefox, and also in Flash, meaning it'll be in IE also. And I don't mind Silverlight either, because in theory, there's IronRuby, IronPython, and so on. I'd much rather have a better tradeoff than straight Javascript, and more options.
But you're wrong on two counts: You should care about dog slow languages, and Javascript will not always be that slow.
Not really ready yet, but apparently there's source up, if you're interested.
That's kind of what they did for 1. Well, they got help from Microsoft, but the project itself is open source.
Compare that to Flash, where if Adobe decides not to upgrade Flash, you have to start implementing Gnash from scratch. That's kind of what happened.
It absolutely is, and it absolutely is not as powerful. You can't make YouTube in HTML (without Flash or Silverlight).
And it's hard to choose, but so far, Silverlight is the lesser of two evils.
Yes.
And holy shit, I'm amazed anyone knew it by name, even Anonymous Coward!
Not really a better way around it -- and they seem to really be wanting you to care. Universal has a brand and everything.
Well, it's at least consistent within the studio, as I said. Universal movies all have menus that look and feel exactly the same, down to the graphics and sound. (The menu makes a sound when sliding in.)
Warner, on the other hand, has basically the same UI on almost all of them, but skinned differently. So you might have that white-to-silver crap, but at least the layout would be predictable.
That part is going to be about the same on HD-DVDs as it was on DVDs, except that the menu is going to be available with the movie playing, and the resolution does actually make UI design a lot easier.
Also, I'm guessing partly because of the cost, it is, again, consistent. Every single Heroes disc had the same menu to select episodes, and every other Universal disc had the same menu to select scenes.
One more thing: Just to see if I could, I coded a particular scrolling menu with a key buffer, which also notices the user mashing and speeds up the scrolling animation at that point. It's not going to be as nice as an iPhone, or even a traditional scrollbar, but it's about the best we can do with a remote. (Holding the button down cannot work.)
However, I do freely admit that the platform makes it possible to absolutely torture you. We could make you sit through commercials, we could do five second long animations (or thirty seconds), could chain multiple animations for simple tasks, and could force you to register before even watching the movie. But I, personally, refuse to do that.
Did I say you should have to? No, actually, I said:
Now, from the Captioned Telephone website, we have:
In other words, get the government to pay for it.
Agreed. All of the technologies I listed will work on VLC and mplayer, on Linux and Mac, and on Windows, if you insist.
But not all buildings are wheelchair accessible.
Erm, look again. It is an A-list team, making a production exclusively (so far) for the Internet -- probably a direct result of the writer's strike. There are multiple fan groups making subtitles for this show. (Not full captioning, I'd assume, as in, no sound effects as text, but speech at least -- and in at least five or six languages so far.)
To be absolutely clear: The show is commercial, professional quality. The subs are done by fans -- fansubs, if you will.
Well, obviously not enough interest to even read the website. Sorry to have wasted your ridiculously precious time.
Try not being so bitter. It helps.
Ditto. You don't see the Yule culture in Christmas?
No, I think what you said was pointless and offtopic.
What people currently care about is the only basis for "reason for the season"? I guess that circles back to the capitalism point. People currently care about getting and giving presents.
That part is not true.
At least on a Free(TM) OS, more Free than Linux currently is, it would not be possible to implement effective DRM, because the user would effectively have control over everything the application has access to. This means that, for instance, you could always run it in a virtual machine, record all traffic in and out of it (including to the pseudo random number generator library), and do a replay attack on it.
That's the more brute-force attack. The fact is, a rootkit should be much easier on Linux. Given the default policy of no root access and the sheer variety of kernels out there, there's simply far less that an app can be sure of about its environment, which makes it much more difficult to tell if that environment is "real" or "trusted". Most games which have been ported to Linux did not bother to port any of the CD-based copy protection, probably because they realized how insanely simple it would be for Linux people to implement an undetectable Daemontools.
With at least the major proprietary OSes, you'll first have to crack the DRM that's built-in to the OS -- convince it that it really is running on bare metal, or convince it to let you do that messing-with-the-IO trick.
So it doesn't completely solve the problem, but I do believe a free system is a lot more hostile, in practice and also in culture, to DRM.
I'm fully aware that Linux itself can have binary kernel modules, at which point, there's really no technological difference. But the cultural difference is important. Anyone switching to Linux is also going to be acutely aware of DRM, partly because things without DRM will work for them, and things with DRM won't (at least for now).
Amazon EC2, or S3 for static content. In fact, if you were fast enough, you could set up an HTTP redirect on your old server (assuming it can even handle that) to point to an S3 cache.
I'm sure there are plenty of others, though. Just telling what I know.