As John mentioned in his post here, these are not new ideas. I remember playing around with raytracing/casting of sparse-octree voxels for fun almost ten years ago, and as a quick search of the literature shows, I was quite late to the game:) What is cool is that he thinks that the gaming world is ready for them, and that he is going to try and push the hardware folks to make it happen.
One of the most fundamental properties of voxmaps is that the geometry and texture are defined hand-in-hand - they have the same resolution, and every point in the geometry has a unique texture. If you want this, then there are data structures like sparse octrees that store the data quite efficiently.
However, decoupling the geometry and texture opens the door for all sorts of tricks usefull in limited memory situations. It was these tricks that made realtime polygon graphics possible in the past. Things like low resolution polygons with high resolution texture maps, tiled/reused texture maps and layered decals, are all ways to cut down on the amount of data needed while still creating a decent looking scene.
However, as the amount of memory increases, these tricks are less necessary and less desirable. Artists want to be able to paint any part of a scene any way they want - and this is exactly what John has done in id Tech 5, their newest engine. After doing so he did some experimentation and found that storing this data in a sparse octree is even more memory efficient than the way he is doing it now, using pixmaps and polygons. If this approach were to work, artists would then have the same freedom in editing the geometry of the world as they do now with textures - the entire world would have geometry at the resolution of current texture maps with zero additional memory costs. That would be awesome.
For this to work though, you need to be able to render the data efficiently. Raycasting of sparse octrees is one of those embarrassingly parallel problems, and thus hardware acceleration for it is relatively easy. But they don't exist due to lack of market, and unfortunately graphics cards are not well suited for this, IIRC because GPUs mostly accelerate floating point calculations, while descending the sparse-octree uses a lot of integer bit-twiddling (I might be wrong about the reasons here). But with the memory-usage tradeoffs shifting in favor of voxmaps, GPU vendors looking to make their products better suited for general purpose High Performace Computing, and John Carmack pushing for it, this may be an idea whose time has come.
Voxmaps in general are great for breakable objects, because unlike the polygon/texture-map approach where only the surface has a texture, with voxmaps every voxel in the object can be defined (CAT scans are examples of voxel data). Furthermore, if you assign properties like tensile strength to the voxmap, you can calculate on demand how and where something will break, and have it be different depending on exactly what the forces on it are. All sorts of cool opportunities there. Of course, the unless you play some tricks to only load the voxels that are currently on the exterior, your memory use will also sky rocket. I can't say whether the particular implementation of sparce-octrees that he is thinking of (I have seen a couple) would be memory efficient for these sorts of destructable environments.
As far as collisions, bouncing, and the sort, voxmaps are just as good as polygons.
Where voxmaps are not well suited though is for bendable, stretchable, or otherwise deformable objects. Calculating the deformation on the fly is much easier with meshes. Even statically defined deformations (like an animated character) will use a ton more memory than the equivalent polygon/texture-map solution. The reason is that the geometry and texture are one and the same so each frame of the animation duplicates all the texture information, as opposed to just changing the polygon and keeping the texture-map the same. This is likely why he would like to continue using the current approach for animated / deformable objects.
Oops, I lost some of my post. Obviously, this is similar to a neuron, however only if the outputs of one device can be fed into the inputs of another, which doesn't appear to be the case since the outputs only have 4 states each, and the input has 4^16 possibilities. If they also had the opposite of this device which took multiple inputs and created an output, then they could chain them together in a neural-net configuration.
I'm trying to understand how something like this could be used to implement a turing-complete device, and I'm not seeing it. The examples given in the article make it sound more like a serial to parallel converter, not a computer.
Maybe in your area. But in the four school boards I have dealt with over the years only one person was a former teacher, and current teachers and administrators were prohibited from sitting on the board as it was a conflict of interest. The boards were mostly comprised of folks that were also on the boards of home-owners associations or various "Think of the Children" activist groups.
Getting beyond the school shootings scenario, the biggest problem at schools in our area is vandalism. Students sneak into the building, trash classrooms, equipment, the athletic field, etc. Now the DVR will record them, and if the alarm is triggered the police view the video feed to learn where they are in the school, how many there are, and if they are armed. This is a good point. The school I went to has had a handfull of breakins, resulting in tens of thousands of dollars of theft and damage. They have never caught any of the people involved. Part of that is probably just laziness on the part of the police, but still cameras would have been very nice there.
At the same time I would absolutely refuse to allow any school that my kids were attending to treat them like prisoners - I do not want our children to grow up thinking that Big Brother tactics are normal and acceptable.
Frankly, I don't see any need for a live system - the police aren't going to respond fast enough for it to matter anyway. A tape recording is good enough, and the schools should be strictly prohibited from operating it during the day, including making the tapes inadmissible for evidence in court or for school punishment.
Why are you blaming this on the teachers when all of the problems you mentioned are the result of policies set by the school board and inflexibly enforced by the administration. A fair number of teachers do overlook the stupid rules, and even if they don't it's not their fault that the punishment for them is ridiculously out of proportion.
Well, Brin does have the tenancy to present the two worlds in his book as a distopia where the just the government knows everything, and a utopia where everyone knows everything. He does point out some flaws in the later system, but in general it is a bit whitewashed.
Schneier is just pointing out that the transparency society is no utopia - that it is, by nature, worse than the system we have now. As to whether it is inevitable - you're right that Schneier failed to counter it, but frankly I also think that Brin failed to demonstrate it to begin with:)
I do know that if we decide it is inevitable, then it most assuredly will be. There is no reason to accept it's inevitability - doing so now does not put us in a better position to create the least offensive kind of transparent society, and could cause us to settle for unnecessary intrusions into our lives.
Anyway, you can't prove anything in science, so I don't know what kind of a "proof" you're looking for. That doesn't make sense to me. The science part is that our observations agree with the model of flat space and not the models of open or closed space. However, the the geometric models themselves, and the predictions of what spot size should be expected from different models of spacial geometry are entirely mathematical, and thus there must be a proof as to how they derived this prediction.
If you haven't read the book, basically the argument that Brin makes is that the complete loss of privacy is inevitable given technology, and thus we shouldn't delude ourselves in thinking we can preserve it, but rather embrace it and fight for transparency on both sides. I don't buy the inevitability argument, and whether he is right or not, the best course of action to preserve balance of power is the same - to fight to preserve privacy on our part, and to increase transparency in the government.
However, there are some more interesting arguments in the book. For example consider CCTV systems. Assuming that their installation is inevitable, he argues that we should fight to make the feeds were available to everyone not just the government. This would empower us to watch the government as much as government is watching us. However, the biggest opposition to this would not be from the government, but from citizens themselves who trust the authorities to watch them, but not their neighbors. This was the attitude he was trying to counter in his book.
Actually, IE 5.5 and 6 do have an exposed method for loading PNGs with an alpha channel - but for some inexplicable reason they didn't use that method when loading a normal image. Someone wrote a javascript filethat modifies the DOM to replace IMG tags pointing to PNGs with IE's proprietary method at load time.
Works like a charm, and allowed me migrate entirely off of GIF years ago.
Obviously, when I say market, I mean enormous money hole... At Google's next shareholder conference call:
Google: Good news! By increasing video quality and duration we've managed to double YouTube's profits over the last quarter. Investor: Wait, didn't YouTube have negative profits last quarter? Google: Ah yes, that would be the bad news.
AIR is a cross-platform development environment that also allows easy porting between desktop and web-based applications. Adobe is planning on creating webapp versions of their major desktop software, including photoshop, within the next 5-10 years. How are they going to do this and keep a manageable code base? You guessed it, they are porting them all to AIR. So Linux should get a native port of Photoshop when that effort is completed, whose "nativeness" is roughly equivalent to the "nativeness" of XUL-Runner applications like Thunderbird.
Here is one article on arstechnica that has a little more detail. I'm sure you can google for more.
If you are understaffed and on a shoe-string budget you shouldn't be looking for superstars. Anyone who is a superstar will be cost you a lot of money, but not everyone who costs a lot of money ends up being a superstar. That is a far more risky proposition for a cash-strapped company than hiring interns is.
"We get brilliant results from average people managing brilliant processes. We observe that our competitors often get average (or worse) results from brilliant people managing broken processes." - Toyota
believe the filamentary structure itself is an artifact of the universe's inflationary epoch Dammit. That was a pretty picture, but now when I look at it all I can see are universal stretch marks:P
The main problem is that it causes inconsistent behaviour. Thats not true in this case, because the base list of visited websites is also changing constantly. Therefore, a sorting algorithm that brings the most often visited to the top will have a higher amount of consistency, compared to other options such as sorting alphabetically or sorting by most recently added. The only option that will have a higher consistency is if it sorted by the least recently added URL, and never removed URLs, but that is also the least useful option. The current sorting algorithm is the best trade-off between presenting a stable list of options, and presenting a useful list of options.
Furthermore, Firefox and IE already bring the most recently used item to the top of the list - that hasn't changed in version 3. What has changed is whether it searches for URLs starting with "text" or URLs that contain "text" anywhere in them.
Try purchasing one copy of a book, re-printing it under a new binding with only the content you don't want to censor, then selling it in large quantity -- and see how long you can get away with it. For every DVD that Cleanflix sold, they purchased an original copy at full price. The issue the directors/producers had with them was that they were changing the movies, not that they weren't getting paid.
How do your examples about the government forcing censorship on a company have anything to do with people choosing to purchase edited version done by a third party?
And I'm sorry but there are no financial justifications for this - allowing a third party to release censored versions increases total sales and total profits, period. The only legitimate objection is to preserve the integrity of ones artistic vision.
/. kids are smart enough to know that transcoding decreases the sound quality, and burning to CD is a waste of money.
But judging from the other comments here, while they're self-righteous enough to bitch about DRM, they don't have the fucking backbone to just not buy DRM'ed music.
Slashdot Mirror
In my defense though, there isn't much of anything I'd want more than a trouser of rabid ferrets! Giggity.
As John mentioned in his post here, these are not new ideas. I remember playing around with raytracing/casting of sparse-octree voxels for fun almost ten years ago, and as a quick :) What is cool is that he thinks that the gaming world is ready for them, and that he is going to try and push the hardware folks to make it happen.
search of the literature shows, I was quite late to the game
One of the most fundamental properties of voxmaps is that the geometry and texture are defined hand-in-hand - they have the same resolution, and every point in the geometry has a unique texture. If you want this, then there are data structures like sparse octrees that store the data quite efficiently.
However, decoupling the geometry and texture opens the door for all sorts of tricks usefull in limited memory situations. It was these tricks that made realtime polygon graphics possible in the past. Things like low resolution polygons with high resolution texture maps, tiled/reused texture maps and layered decals, are all ways to cut down on the amount of data needed while still creating a decent looking scene.
However, as the amount of memory increases, these tricks are less necessary and less desirable. Artists want to be able to paint any part of a scene any way they want - and this is exactly what John has done in id Tech 5, their newest engine. After doing so he did some experimentation and found that storing this data in a sparse octree is even more memory efficient than the way he is doing it now, using pixmaps and polygons. If this approach were to work, artists would then have the same freedom in editing the geometry of the world as they do now with textures - the entire world would have geometry at the resolution of current texture maps with zero additional memory costs. That would be awesome.
For this to work though, you need to be able to render the data efficiently. Raycasting of sparse octrees is one of those embarrassingly parallel problems, and thus hardware acceleration for it is relatively easy. But they don't exist due to lack of market, and unfortunately graphics cards are not well suited for this, IIRC because GPUs mostly accelerate floating point calculations, while descending the sparse-octree uses a lot of integer bit-twiddling (I might be wrong about the reasons here). But with the memory-usage tradeoffs shifting in favor of voxmaps, GPU vendors looking to make their products better suited for general purpose High Performace Computing, and John Carmack pushing for it, this may be an idea whose time has come.
Voxmaps in general are great for breakable objects, because unlike the polygon/texture-map approach where only the surface has a texture, with voxmaps every voxel in the object can be defined (CAT scans are examples of voxel data). Furthermore, if you assign properties like tensile strength to the voxmap, you can calculate on demand how and where something will break, and have it be different depending on exactly what the forces on it are. All sorts of cool opportunities there. Of course, the unless you play some tricks to only load the voxels that are currently on the exterior, your memory use will also sky rocket. I can't say whether the particular implementation of sparce-octrees that he is thinking of (I have seen a couple) would be memory efficient for these sorts of destructable environments.
As far as collisions, bouncing, and the sort, voxmaps are just as good as polygons.
Where voxmaps are not well suited though is for bendable, stretchable, or otherwise deformable objects. Calculating the deformation on the fly is much easier with meshes. Even statically defined deformations (like an animated character) will use a ton more memory than the equivalent polygon/texture-map solution. The reason is that the geometry and texture are one and the same so each frame of the animation duplicates all the texture information, as opposed to just changing the polygon and keeping the texture-map the same. This is likely why he would like to continue using the current approach for animated / deformable objects.
Oops, I lost some of my post. Obviously, this is similar to a neuron, however only if the outputs of one device can be fed into the inputs of another, which doesn't appear to be the case since the outputs only have 4 states each, and the input has 4^16 possibilities. If they also had the opposite of this device which took multiple inputs and created an output, then they could chain them together in a neural-net configuration.
I'm trying to understand how something like this could be used to implement a turing-complete device, and I'm not seeing it. The examples given in the article make it sound more like a serial to parallel converter, not a computer.
Maybe in your area. But in the four school boards I have dealt with over the years only one person was a former teacher, and current teachers and administrators were prohibited from sitting on the board as it was a conflict of interest. The boards were mostly comprised of folks that were also on the boards of home-owners associations or various "Think of the Children" activist groups.
At the same time I would absolutely refuse to allow any school that my kids were attending to treat them like prisoners - I do not want our children to grow up thinking that Big Brother tactics are normal and acceptable.
Frankly, I don't see any need for a live system - the police aren't going to respond fast enough for it to matter anyway. A tape recording is good enough, and the schools should be strictly prohibited from operating it during the day, including making the tapes inadmissible for evidence in court or for school punishment.
Why are you blaming this on the teachers when all of the problems you mentioned are the result of policies set by the school board and inflexibly enforced by the administration. A fair number of teachers do overlook the stupid rules, and even if they don't it's not their fault that the punishment for them is ridiculously out of proportion.
Well, Brin does have the tenancy to present the two worlds in his book as a distopia where the just the government knows everything, and a utopia where everyone knows everything. He does point out some flaws in the later system, but in general it is a bit whitewashed.
:)
Schneier is just pointing out that the transparency society is no utopia - that it is, by nature, worse than the system we have now. As to whether it is inevitable - you're right that Schneier failed to counter it, but frankly I also think that Brin failed to demonstrate it to begin with
I do know that if we decide it is inevitable, then it most assuredly will be. There is no reason to accept it's inevitability - doing so now does not put us in a better position to create the least offensive kind of transparent society, and could cause us to settle for unnecessary intrusions into our lives.
If you haven't read the book, basically the argument that Brin makes is that the complete loss of privacy is inevitable given technology, and thus we shouldn't delude ourselves in thinking we can preserve it, but rather embrace it and fight for transparency on both sides. I don't buy the inevitability argument, and whether he is right or not, the best course of action to preserve balance of power is the same - to fight to preserve privacy on our part, and to increase transparency in the government.
However, there are some more interesting arguments in the book. For example consider CCTV systems. Assuming that their installation is inevitable, he argues that we should fight to make the feeds were available to everyone not just the government. This would empower us to watch the government as much as government is watching us. However, the biggest opposition to this would not be from the government, but from citizens themselves who trust the authorities to watch them, but not their neighbors. This was the attitude he was trying to counter in his book.
Actually, IE 5.5 and 6 do have an exposed method for loading PNGs with an alpha channel - but for some inexplicable reason they didn't use that method when loading a normal image. Someone wrote a javascript filethat modifies the DOM to replace IMG tags pointing to PNGs with IE's proprietary method at load time.
Works like a charm, and allowed me migrate entirely off of GIF years ago.
Why do I get the feeling that this was not just any bunny
Google: Good news! By increasing video quality and duration we've managed to double YouTube's profits over the last quarter.
Investor: Wait, didn't YouTube have negative profits last quarter?
Google: Ah yes, that would be the bad news.
AIR is a cross-platform development environment that also allows easy porting between desktop and web-based applications. Adobe is planning on creating webapp versions of their major desktop software, including photoshop, within the next 5-10 years. How are they going to do this and keep a manageable code base? You guessed it, they are porting them all to AIR. So Linux should get a native port of Photoshop when that effort is completed, whose "nativeness" is roughly equivalent to the "nativeness" of XUL-Runner applications like Thunderbird.
Here is one article on arstechnica that has a little more detail. I'm sure you can google for more.
If you are understaffed and on a shoe-string budget you shouldn't be looking for superstars. Anyone who is a superstar will be cost you a lot of money, but not everyone who costs a lot of money ends up being a superstar. That is a far more risky proposition for a cash-strapped company than hiring interns is.
"We get brilliant results from average people managing brilliant processes. We observe that our competitors often get average (or worse) results from brilliant people managing broken processes." - Toyota
Furthermore, Firefox and IE already bring the most recently used item to the top of the list - that hasn't changed in version 3. What has changed is whether it searches for URLs starting with "text" or URLs that contain "text" anywhere in them.
How do your examples about the government forcing censorship on a company have anything to do with people choosing to purchase edited version done by a third party?
And I'm sorry but there are no financial justifications for this - allowing a third party to release censored versions increases total sales and total profits, period. The only legitimate objection is to preserve the integrity of ones artistic vision.
Is MS only charging licensing fees for encoders and not decoders (ie PS3)?
Or perhaps, Very Large High Lucidity Array. The first telescope to peer into the realm of the gods.
/. kids are smart enough to know that transcoding decreases the sound quality, and burning to CD is a waste of money.
But judging from the other comments here, while they're self-righteous enough to bitch about DRM, they don't have the fucking backbone to just not buy DRM'ed music.