I saw a lecture in '99 by Cassidy Curtis, the TD (technical director, for the rest of you) on David Gainey's "Fishing" among other remarkable shorts.
He spoke about the need for perl and the like to glue together the bits in a pipeline process. He would take the z-buffer for a frame and take the cross product with the gradient, use the vector to drag a pen stroke which decayed with distance, and wound up with a beautiful squiggly outline of the original object. Then you flatten out the colors in the original, overlay your lines, and they look like loosely inked color barriers in a comic strip.
Anyway, the point was that each of these was a separate utility, and you needed the scripts to lay them all end to end.
The USPTO has an impossible mandate and insufficient funds. I hear a lot of cries for reform, and not a lot of specific suggestions.
Here's one off the top of my head: the patent office should get a percentage of the proceeds of patent licensing fees, and use them to hire real domain experts.
A dicier proposition: patent trials should involve peer review. Open to a lot of abuse from peers, though. No worse than the current "the truth is whatever twelve hand-picked idiots vote for" system.
This is image-based rendering. In image-based rendering, there is no 3D geometry, in the sense of vertices and polygons. What you have is an image of a subject from a number of angles, and you interpolate to get the angles in between. Remember Quicktime VR? (My graphics wonk credentials include unadvisably early adoption of QTVR.) It's the first reference in this article. Regular SIGGRAPH attendees will be used to having their jaws dropped by new image-based rendering advancements every year. This is definitely a field to watch, but what's presented here looks like an incremental advance more than a breakthrough.
The article is a bit unclear, but it sounds like the new wrinkle here is that you use a set of microlenses to capture all the light coming through a plane...as if you had a solid wall of cameras. Then you have an output device which can take advantage of all the information in each of those source images (actually source video in this case). The point is a flat device which will give the illusion of 3D (including parallax) without an intervening plane or glasses. If I'm reading this correctly, it sounds like the effect will be very similar to a plane holograph, where you can walk around in front of the display and look at it from different angles, but if you go to far the image breaks, and of course you can't go behind it or rotate the object. So saying you can look at it from "arbitrary angles" is a bit thick.
Anyway, this is a very separate thing from those systems that take a number of photographs and reconstruct the geometry, or those other systems that use range-finding lasers or the like to actually measure the geometry.
In short, no application for Quake here, so it's surprising it got funded in the first place.:)
(Um...now watch them come out with image-rendered Quake...)
NOTE -- All committee ballots shall contain the following patent statement:
"If you are aware of any patents involved in this ballot, check this box and notify the
committee, citing the patent numbers."
The problem here is that while it sat on the JEDEC, RAMBUS had applied for but had not yet been awarded the patent. Everyone agress that RAMBUS didn't try to influence the standard setting process. The article also suggests that the outcome would have been the same even if the patents had been in place earlier, since the patents are so hard to circumvent (that's what you're going for when you write a patent). So the argument is that they should have disclosed the fact (presumably, somehow, without giving any indication of the contents) of the patent application.
Any business might reasonably be reluctant to do that. Setting the specifics of the RAMBUS case aside, if you're a company with a patent application in progress, you don't want anyone to know what the patent is about, because you don't have any protection for your technology. After all, the patent might be rejected, and then secrecy is your only protection.
As I am about the billionth person to point out, the problem is in the USPTO. Patent examiners come from the lower ranks of engineering, and patent trial juries are hopelessly overmatched by the issues they are presented with. However, there must be some system in place which allows someone to invest money in research with some expectation of making back that money. Otherwise you have the kid's soccer game model of technology, one company innovates and is immediately swarmed by as everyone clusters around the ball, and the winner is often the one who has the most to spend on marketing because they didn't have to do any engineering.
Insurance lobbyists outspend every other industry. They must be greasing the wrong palms.
It's nice to see that every once in a while, instead of taking the industry lobbyists' money and doing what they want, our "elected" representatives will have a fit of conscience, and take the lobbyist's money and not do what they want.
In the past I've humbly suggested that instead of giving money to the insurance companies who give it to congress which passes laws which require that we give more money to the insurance companies, we all just send our money directly to congress, in an unmarked shoebox. But in light of this new development, I suppose I'll have to reconsider.
"Great, so I'll put you down for two abdominizers and one 'Super thighs in 30 days the Steve Jobs way'?"
"Yes...OK..."
"Great, I assume you'll want to use the address and credit card we have on file."
"...OK...yes."
"We also have the Jeff Bezos Power Diet Plan, boxed set, signed, for $49.95."
"...I see...Yes...OK..."
My practice for years has been to simply set the phone down gently and let them talk on. Occasionally, if I walk by the phone a few minutes later and they're still talking, I'll pick up the receiver and say "Go on," or, "I'm sorry, I didn't catch that last part, could you go over that again," and set the phone back down. Sometimes they catch on quickly, of course, but some callers will go through their entire script without noticing I'm not there.
The point is not to piss off the telemarketer, that's just fortuitous. The point is to take up as much time on a fruitless call as possible.
Telemarketers' business models depend on their getting through the negative calls in as little time as possible. That is, they *depend* on us snarling and hanging up on them. If instead, the custom were to chat with them indefinitely, the business would become unprofitable, because they couldn't cycle through the negative calls quickly enough to get to a profitable margin of positives. In a polite society, telemarketing doesn't work.
It may be cheaper for them to give you a promotion and raise than to replace your boss on the open market. An added bonus is that they can fill your position with someone less qualified and cheaper, since you will still be around to pick up the slack. Given the choice between money and reason, management always goes for the green.
Ask for top dollar and the unqualified right to pick your successor. If they go for it, take the job, then use the new title to go job hunting. Remember, these aren't smart guys.
Knowing little about digital signal processing but slightly more about patents, I can point out that the "ominous words" at the end of the patent are little more than boilerplate legalese; you will see almost this exact formulation in many patents. It means simply that if they give specific examples in the description of the patent, that doesn't mean that the claims are implied to be referencing just those specific examples.
Read the abstract and description to understand the patent, but the claims are what are legally binding...this is the part where overly vague or inclusive language is truly damaging.
Yes, but you pay them. They work for you.
This kind of unilateral action may adhere to the boilerplate of the service agreement, but there is a difference between legal and ethical. In order to determine if the action is ethical, one must consider the motive...and what is the motive here? To avoid the suits brought by the owners of the information encoded in the MP3s? To reduce the disk space and bandwidth usage of their customers?
In either case, the option to vote with one's feet remains...in fact, it could be considered a civic duty to deny patronage to businesses which operate in violation of civic trust, regardless of the legality of their actions.
Slashdot Mirror
I saw a lecture in '99 by Cassidy Curtis, the TD (technical director, for the rest of you) on David Gainey's "Fishing" among other remarkable shorts.
He spoke about the need for perl and the like to glue together the bits in a pipeline process. He would take the z-buffer for a frame and take the cross product with the gradient, use the vector to drag a pen stroke which decayed with distance, and wound up with a beautiful squiggly outline of the original object. Then you flatten out the colors in the original, overlay your lines, and they look like loosely inked color barriers in a comic strip.
Anyway, the point was that each of these was a separate utility, and you needed the scripts to lay them all end to end.
The USPTO has an impossible mandate and insufficient funds. I hear a lot of cries for reform, and not a lot of specific suggestions. Here's one off the top of my head: the patent office should get a percentage of the proceeds of patent licensing fees, and use them to hire real domain experts. A dicier proposition: patent trials should involve peer review. Open to a lot of abuse from peers, though. No worse than the current "the truth is whatever twelve hand-picked idiots vote for" system.
This is image-based rendering. In image-based rendering, there is no 3D geometry, in the sense of vertices and polygons. What you have is an image of a subject from a number of angles, and you interpolate to get the angles in between. Remember Quicktime VR? (My graphics wonk credentials include unadvisably early adoption of QTVR.) It's the first reference in this article. Regular SIGGRAPH attendees will be used to having their jaws dropped by new image-based rendering advancements every year. This is definitely a field to watch, but what's presented here looks like an incremental advance more than a breakthrough. The article is a bit unclear, but it sounds like the new wrinkle here is that you use a set of microlenses to capture all the light coming through a plane...as if you had a solid wall of cameras. Then you have an output device which can take advantage of all the information in each of those source images (actually source video in this case). The point is a flat device which will give the illusion of 3D (including parallax) without an intervening plane or glasses. If I'm reading this correctly, it sounds like the effect will be very similar to a plane holograph, where you can walk around in front of the display and look at it from different angles, but if you go to far the image breaks, and of course you can't go behind it or rotate the object. So saying you can look at it from "arbitrary angles" is a bit thick. Anyway, this is a very separate thing from those systems that take a number of photographs and reconstruct the geometry, or those other systems that use range-finding lasers or the like to actually measure the geometry. In short, no application for Quake here, so it's surprising it got funded in the first place. :)
(Um...now watch them come out with image-rendered Quake...)
Usefulness, obviousness, and absense of prior art. Is there a clear application, is it trivial, and has someone else already discovered it?
Unfortunately the process pits domain experts and highly paid lawyers against underpaid patent examiners with an impossible mandate.
So would somebody please suggest a workable alternative, perhaps akin to peer review?
Not everyone at once, now...
From the JEDEC manual:
NOTE -- All committee ballots shall contain the following patent statement:
"If you are aware of any patents involved in this ballot, check this box and notify the
committee, citing the patent numbers."
The problem here is that while it sat on the JEDEC, RAMBUS had applied for but had not yet been awarded the patent. Everyone agress that RAMBUS didn't try to influence the standard setting process. The article also suggests that the outcome would have been the same even if the patents had been in place earlier, since the patents are so hard to circumvent (that's what you're going for when you write a patent). So the argument is that they should have disclosed the fact (presumably, somehow, without giving any indication of the contents) of the patent application.
Any business might reasonably be reluctant to do that. Setting the specifics of the RAMBUS case aside, if you're a company with a patent application in progress, you don't want anyone to know what the patent is about, because you don't have any protection for your technology. After all, the patent might be rejected, and then secrecy is your only protection.
As I am about the billionth person to point out, the problem is in the USPTO. Patent examiners come from the lower ranks of engineering, and patent trial juries are hopelessly overmatched by the issues they are presented with. However, there must be some system in place which allows someone to invest money in research with some expectation of making back that money. Otherwise you have the kid's soccer game model of technology, one company innovates and is immediately swarmed by as everyone clusters around the ball, and the winner is often the one who has the most to spend on marketing because they didn't have to do any engineering.
It's nice to see that every once in a while, instead of taking the industry lobbyists' money and doing what they want, our "elected" representatives will have a fit of conscience, and take the lobbyist's money and not do what they want.
In the past I've humbly suggested that instead of giving money to the insurance companies who give it to congress which passes laws which require that we give more money to the insurance companies, we all just send our money directly to congress, in an unmarked shoebox. But in light of this new development, I suppose I'll have to reconsider.
Purple: Lockheed Martin, Boeing, Raytheon and TRW.
Orange: The rest of the planet.
Looks like we lose again!
What about:
"Great, so I'll put you down for two abdominizers and one 'Super thighs in 30 days the Steve Jobs way'?"
"Yes...OK..."
"Great, I assume you'll want to use the address and credit card we have on file."
"...OK...yes."
"We also have the Jeff Bezos Power Diet Plan, boxed set, signed, for $49.95."
"...I see...Yes...OK..."
My practice for years has been to simply set the phone down gently and let them talk on. Occasionally, if I walk by the phone a few minutes later and they're still talking, I'll pick up the receiver and say "Go on," or, "I'm sorry, I didn't catch that last part, could you go over that again," and set the phone back down. Sometimes they catch on quickly, of course, but some callers will go through their entire script without noticing I'm not there.
The point is not to piss off the telemarketer, that's just fortuitous. The point is to take up as much time on a fruitless call as possible.
Telemarketers' business models depend on their getting through the negative calls in as little time as possible. That is, they *depend* on us snarling and hanging up on them. If instead, the custom were to chat with them indefinitely, the business would become unprofitable, because they couldn't cycle through the negative calls quickly enough to get to a profitable margin of positives. In a polite society, telemarketing doesn't work.
Probably "The Jigsaw Man", by Larry Niven, 1967.
What do I win?
It may be cheaper for them to give you a promotion and raise than to replace your boss on the open market. An added bonus is that they can fill your position with someone less qualified and cheaper, since you will still be around to pick up the slack. Given the choice between money and reason, management always goes for the green.
Ask for top dollar and the unqualified right to pick your successor. If they go for it, take the job, then use the new title to go job hunting. Remember, these aren't smart guys.
You are confused, this is Slashdot, not Science. Please address your concerns about the conduct of Science to the following address:
Science
PO Box 8473
Schenectady, NY 12301
I'm sure that you will be able to convince Science to mend its evil ways. If that doesn't work, though, try:
Capitalism
120B Broadway
NY, NY 10009
Good luck.
Will I be cursed as humor-impaired if I provide the blink constant: 0.33 seconds for the average human blink?
Thought so.
Knowing little about digital signal processing but slightly more about patents, I can point out that the "ominous words" at the end of the patent are little more than boilerplate legalese; you will see almost this exact formulation in many patents. It means simply that if they give specific examples in the description of the patent, that doesn't mean that the claims are implied to be referencing just those specific examples.
Read the abstract and description to understand the patent, but the claims are what are legally binding...this is the part where overly vague or inclusive language is truly damaging.
Yes, but you pay them. They work for you. This kind of unilateral action may adhere to the boilerplate of the service agreement, but there is a difference between legal and ethical. In order to determine if the action is ethical, one must consider the motive...and what is the motive here? To avoid the suits brought by the owners of the information encoded in the MP3s? To reduce the disk space and bandwidth usage of their customers? In either case, the option to vote with one's feet remains...in fact, it could be considered a civic duty to deny patronage to businesses which operate in violation of civic trust, regardless of the legality of their actions.