Re:Does anyone have an actual video of the demo?
on
Photosynth Demo
·
· Score: 1
That's hot. Glad to hear the photos were presented as is (except for the ones that were removed) and that the computer vision works well enough to get lots of good alignment through feature detection. Seeing photosynth at work for the BBC should be great (I showed photosynth to a friend at the New York Times and it sounds like they'll be jealous).
Re:Did the cat push your button?
on
Photosynth Demo
·
· Score: 0
I'll have you know I also changed the word "often" to "awesome", but yeah, sorry about that.
Thanks for the support. It's funny you should mention the voice over because I just showed it to my friend and he thought the voice over was great. Totally dead pan, no frills, just letting this totally awesome software speak for itself.
Not just released, but still awesome
on
Photosynth Demo
·
· Score: 0, Redundant
This system was demoed a while ago, I think at siggraph 2006. There are some videos on the original university of washington "PhotoTourism" research page. Also here's a repost of the same video on that page on youTube.
Also there's microsoft's photosynth page, which has the demo (it's been up for months). It seems to have some longer videos.
Non-newness and marketing hype aside, this software is frickin' awesome. It lets you view and tag photos organized in a 3D environment that reflects where the photos were taken. It should be particularly useful once cameras have GPS built in. It could be used by a single person, but it's definitely well suited to large collections such as flickr.
I imagine the reason the software is still in the demo phase is because it's very difficult to take a large number of photos and reliably figure out where they were all taken from. For the demo purposes, Microsoft probably hand corrected a lot of the placements. Even so, everyone I've shown this to thinks its awesome (even non-slashdot readers!).
Re:Does anyone have an actual video of the demo?
on
Photosynth Demo
·
· Score: 5, Informative
This system was demoed a while ago, I think at siggraph. There are some videos on the original university of washington
PhotoTourism page.. Also here's a repost of the video on
youTube.
Also there's microsoft's page, which has the demo (I don't think that's new either). It seems to have some longer videos
Non-newness and marketing hype aside, this software is frickin' awesome. It lets you view and tag photos organized in a 3D environment that reflects where the photos were taken. It should be particularly useful once cameras have GPS built in.
I imagine the reason the software is still in the demo phase is because it's very difficult to take a large number of photos and reliably figure out where they were all taken from. For the demo purposes, Microsoft probably hand corrected a lot of the placements. Even so, everyone I've shown this too thinks its often (even non-slashdot readers!)
More details can be found in the nature article (available to subscribers only)
here. There is also a news feature and editors summary. For those without a subscription:
The memory is nonvolatile (technically speaking), but bits decay in about an hour.
The technology is not at the point were word length is a concern. The researchers were reading and writing individual bits. About half of all bits were deemed defective (having an on/off ratio of less that 1.5)
As said above, bits last for about an hour. I'm sure they intend to improve this.
Bits were written using.2 second pulses of 1.5 volts and read using.2 volts. I doubt any attempt was made to optimize these quantities. Remember, voltages are not being applied directly to the molecules storing information, they are applied to sets of perpendicular nanowires. The electrodes controlling these nanowires might be the cause of the slow switching time.
There is no mention of power requirements in the article
As you can see, this technology is not coming to market any time soon. The research demonstrates that a grid of nanowires can control a layer of programmable molecules, yielding a very high storage densities. High defect rates, however, may require a large overhead (think of the size of the defect map required). More importantly, the nanowires that form the grid cannot yet be individually controlled. In this experiment, electrodes were used to turn on 2 to 4 nanowires at a time. To date, no reliable way of controlling many individual nanowires has been experimentally demonstrated.
Obviously it's a bummer to not be able to buy new music for a day, but with DRM, there's more at stake. On Chrismas, I wanted to transfer some songs and videos I already owned to my new iPod, but I was using my powerbook which I hadn't authorized to play the songs. Even though I had the files, iTunes wouldn't put unauthorized content on my new iPod. When I tried to authorize my computer, I couldn't because the iTunes music store was down.
Even if the iTunes store is too flooded to handle new purchases, Apple has an obligation to keep rights management up and running. It's very disheartening to lose access to content you made the deliberate choice to buy legally.
Free wifi + Voip = free cell phone service. It's that simple.
Sure we may have to pay some initial fee for a device, but I'm sure in 5 years we'll be able to get something that works for $20 or less. I guess some people will be willing to give you a phone for free, but I doubt that will be of interest to most consumers (and certainly not of most people reading this website). Most folks will get a nice looking device that works well with whatever electronics they already own (computer, stereo perhaps).
The state of Rhode Island is already in the process of putting up WiMax towers for the entire state. This isn't going to free for everyone to use initially (it's being marketed as something business will be able to pay for), but once it's up and running, people are going to push to make it open to the public. Whatever the cost, it'll be a lot less than individual households paying ISPs and phone companies, and eventually it'll be funded by taxes. Between WiMax and municiple WiFi, large chunks of the population will have free internet, and then we'll work on filling in the gaps.
Free phone service for all is a great thing, but I sure don't need Google to tell me it's on its way.
The story about a paralyzed man (Matthew Nagle) controling a computer with his brain is definitely not new. There was a very good story in Wired in March 2005, and much more recently, a piece on NPR's The Infinite Mind. According to the piece, Matthew has since had the implant removed, since the trial has ended. I believe at least one other trial is in progress.
As for bugs controlling stuff with their mind, here's a sciencenews article from 2000 about a lamprey (not actually a bug I guess) steering a computer-controlled robot for no good reason. I saw the original paper in Artificial Life at some point, and it was easily the most ridiculous scientific journal article I've ever seen.
The time to start monitoring is key. If quasars generate random bits at a high enough rate, it become infeasible for a third party to just start recording bits from some quasar then search for a particular one time pad.
A few years ago, I heard a talk by Micheal Rabin that explained how completely secure one time pads could be obtained from a satellite (or some other data source) that generated random numbers at a high enough rate (see New York Times Article here). It seems like a collection of quasars could play the roll of the satellite. Once again, the key to his approach was that the satellite generated too many random bits for them all to be recorded.
In his approach, traditional cryptographic techniques can be used initally decide when two parties should start sampling random bits to generate one time pad. That pad can then be used to decide on additional pads. If a third party intercepts every communication, they could potentially generate the pads themselves, but they would have to act very quickly, because once they failed to record bits from the satellite (or quasars) those bits would be gone forever.
Normally, if you intercept an encrypted communication, you can hold on to the message and attempt to break the decryption over time. With this approach, if you don't decrypt the intial communication right away, you've missed out on the one time pads, and thus the captured message is nothing more than random bits. It can never be decrypted.
Ron Rivest and Silvio Micali have looked closely at micropayments, and are even involved with a startup named Peppercoin.
It's been a while since I heard Professor Rivest (of RSA fame) speak, but part of his solution was to aggregate payments across customers. In his talk he explained how buyers could be given electronic money, only a fraction of which was actually valuable to a particular vendor. Suppose 9 out of 10 "e-dimes" would not be redeemed, but the 10th entitled the vendor to a dollar from the credit card company. The vendor would break even on average, but if they could test the money themselves, would use far fewer transactions.
Making the whole system secure relies on a bunch of cryptography. Specifically, it's important to not be able to forge money, not be able to test a bunch of money to see if it's valuable to a vendor, and not be able to give the same money to multiple vendors.
I imagine the Peppercorn is currently proposing a simpler solution that integrates better with existing technology. Either
way, when a payments are sufficiently large, the system could just defer to current technology, so the vendor is no worse off.
Biometrics can be pretty problematic in certain situations. Take one generated from a finger print. It works great if you can take someone's fingerprint on your own, secure equipment. It doesn't do much good though if someone just sends you their fingerprint data remotely. For remote transactions, a biometric may become no different than a really long credit card number. The key difference being that if someone steals your fingerprint, you damn sure won't be able to get a new one.
Preventing a simple number (even a long one) from allowing someone to steal your identity is the right idea.
A credit card with a chip, for instance, can respond to a query that validates the card. If a credit card company only allows each query/response to be used once, saved information becomes useless. For online purchases, software provided by the credit card company could play the roll of the card.
This is just one example, but the bottom line is there are lots of ways to prevent the information we provide companies from being useful in stealing our identity.
Also, a much simpler approach is to keep people as informed as possible. It should be easy to find out if your social security number has been used for anything. Credit card companies already allow you to recieve an e-mail if your card balance changes, and banks should do the same.
For a completely blind user-interface (which seems to be the much greater challenge), people seem too set on braille/tactile interfaces.
Braille is certainly a nice way to output text, but as someone posted, not all blind people know braille. Additionally, almost no sighted people know braille, which means a blind person couldn't use the computer with their sighted friend, nor could a sighted person help the blind person use their computer.
More elaborate tactile interfaces would have similar short-comings, and what's worse, they would rely on technology that doesn't really exist yet (or rather, is just starting to exist).
Also, though a text based interface is an easy solution, it's not a good one. There's no reason why your average blind person would like a text interface anymore than a sighted person. The interface should mimic the way a blind peson interacts with the world. They should have the ability to navigate in a space (perhaps with a mouse or 3D pointing device) and recieve feedback.
The feedback should be given both in a tactile fashion and through sound. Using a pointing device you could feel when you "hit" an icon, and then hear it's name.
Sound spacialization also offers a wealth of possiblities. Blind people would be quite capable of naviagating through a virtual space in which sounds and words suggested where things were. Ear-bud headphones along with the proper software could potentially do an excellent job of simulating 3D sound (check out a "binaural recording" to hear for yourself). Properly placed speakers would also work.
Obviously there is lots to consider, but a good 3D sound-based interface might be useful for sighted users as well, allowing for sophisticated control of devices without screens (I could imagine, for instance, an iPod shuffle that let you navigate through a much larger music collection by reading the names of bands)
Slashdot Mirror
That's hot. Glad to hear the photos were presented as is (except for the ones that were removed) and that the computer vision works well enough to get lots of good alignment through feature detection. Seeing photosynth at work for the BBC should be great (I showed photosynth to a friend at the New York Times and it sounds like they'll be jealous).
I'll have you know I also changed the word "often" to "awesome", but yeah, sorry about that.
Thanks for the support. It's funny you should mention the voice over because I just showed it to my friend and he thought the voice over was great. Totally dead pan, no frills, just letting this totally awesome software speak for itself.
This system was demoed a while ago, I think at siggraph 2006. There are some videos on the original university of washington "PhotoTourism" research page. Also here's a repost of the same video on that page on youTube.
Also there's microsoft's photosynth page, which has the demo (it's been up for months). It seems to have some longer videos.
Non-newness and marketing hype aside, this software is frickin' awesome. It lets you view and tag photos organized in a 3D environment that reflects where the photos were taken. It should be particularly useful once cameras have GPS built in. It could be used by a single person, but it's definitely well suited to large collections such as flickr. I imagine the reason the software is still in the demo phase is because it's very difficult to take a large number of photos and reliably figure out where they were all taken from. For the demo purposes, Microsoft probably hand corrected a lot of the placements. Even so, everyone I've shown this to thinks its awesome (even non-slashdot readers!).
This system was demoed a while ago, I think at siggraph. There are some videos on the original university of washington PhotoTourism page.. Also here's a repost of the video on youTube.
Also there's microsoft's page, which has the demo (I don't think that's new either). It seems to have some longer videos
Non-newness and marketing hype aside, this software is frickin' awesome. It lets you view and tag photos organized in a 3D environment that reflects where the photos were taken. It should be particularly useful once cameras have GPS built in.
I imagine the reason the software is still in the demo phase is because it's very difficult to take a large number of photos and reliably figure out where they were all taken from. For the demo purposes, Microsoft probably hand corrected a lot of the placements. Even so, everyone I've shown this too thinks its often (even non-slashdot readers!)
- The memory is nonvolatile (technically speaking), but bits decay in about an hour.
- The technology is not at the point were word length is a concern. The researchers were reading and writing individual bits. About half of all bits were deemed defective (having an on/off ratio of less that 1.5)
- As said above, bits last for about an hour. I'm sure they intend to improve this.
- Bits were written using
.2 second pulses of 1.5 volts and read using .2 volts. I doubt any attempt was made to optimize these quantities. Remember, voltages are not being applied directly to the molecules storing information, they are applied to sets of perpendicular nanowires. The electrodes controlling these nanowires might be the cause of the slow switching time.
- There is no mention of power requirements in the article
As you can see, this technology is not coming to market any time soon. The research demonstrates that a grid of nanowires can control a layer of programmable molecules, yielding a very high storage densities. High defect rates, however, may require a large overhead (think of the size of the defect map required). More importantly, the nanowires that form the grid cannot yet be individually controlled. In this experiment, electrodes were used to turn on 2 to 4 nanowires at a time. To date, no reliable way of controlling many individual nanowires has been experimentally demonstrated.Obviously it's a bummer to not be able to buy new music for a day, but with DRM, there's more at stake. On Chrismas, I wanted to transfer some songs and videos I already owned to my new iPod, but I was using my powerbook which I hadn't authorized to play the songs. Even though I had the files, iTunes wouldn't put unauthorized content on my new iPod. When I tried to authorize my computer, I couldn't because the iTunes music store was down.
Even if the iTunes store is too flooded to handle new purchases, Apple has an obligation to keep rights management up and running. It's very disheartening to lose access to content you made the deliberate choice to buy legally.
Free wifi + Voip = free cell phone service. It's that simple.
Sure we may have to pay some initial fee for a device, but I'm sure in 5 years we'll be able to get something that works for $20 or less. I guess some people will be willing to give you a phone for free, but I doubt that will be of interest to most consumers (and certainly not of most people reading this website). Most folks will get a nice looking device that works well with whatever electronics they already own (computer, stereo perhaps).
The state of Rhode Island is already in the process of putting up WiMax towers for the entire state. This isn't going to free for everyone to use initially (it's being marketed as something business will be able to pay for), but once it's up and running, people are going to push to make it open to the public. Whatever the cost, it'll be a lot less than individual households paying ISPs and phone companies, and eventually it'll be funded by taxes. Between WiMax and municiple WiFi, large chunks of the population will have free internet, and then we'll work on filling in the gaps.
Free phone service for all is a great thing, but I sure don't need Google to tell me it's on its way.The story about a paralyzed man (Matthew Nagle) controling a computer with his brain is definitely not new. There was a very good story in Wired in March 2005, and much more recently, a piece on NPR's The Infinite Mind. According to the piece, Matthew has since had the implant removed, since the trial has ended. I believe at least one other trial is in progress.
As for bugs controlling stuff with their mind, here's a sciencenews article from 2000 about a lamprey (not actually a bug I guess) steering a computer-controlled robot for no good reason. I saw the original paper in Artificial Life at some point, and it was easily the most ridiculous scientific journal article I've ever seen.
The time to start monitoring is key. If quasars generate random bits at a high enough rate, it become infeasible for a third party to just start recording bits from some quasar then search for a particular one time pad.
A few years ago, I heard a talk by Micheal Rabin that explained how completely secure one time pads could be obtained from a satellite (or some other data source) that generated random numbers at a high enough rate (see New York Times Article here). It seems like a collection of quasars could play the roll of the satellite. Once again, the key to his approach was that the satellite generated too many random bits for them all to be recorded.
In his approach, traditional cryptographic techniques can be used initally decide when two parties should start sampling random bits to generate one time pad. That pad can then be used to decide on additional pads. If a third party intercepts every communication, they could potentially generate the pads themselves, but they would have to act very quickly, because once they failed to record bits from the satellite (or quasars) those bits would be gone forever.
Normally, if you intercept an encrypted communication, you can hold on to the message and attempt to break the decryption over time. With this approach, if you don't decrypt the intial communication right away, you've missed out on the one time pads, and thus the captured message is nothing more than random bits. It can never be decrypted.
Ron Rivest and Silvio Micali have looked closely at micropayments, and are even involved with a startup named Peppercoin.
It's been a while since I heard Professor Rivest (of RSA fame) speak, but part of his solution was to aggregate payments across customers. In his talk he explained how buyers could be given electronic money, only a fraction of which was actually valuable to a particular vendor. Suppose 9 out of 10 "e-dimes" would not be redeemed, but the 10th entitled the vendor to a dollar from the credit card company. The vendor would break even on average, but if they could test the money themselves, would use far fewer transactions.
Making the whole system secure relies on a bunch of cryptography. Specifically, it's important to not be able to forge money, not be able to test a bunch of money to see if it's valuable to a vendor, and not be able to give the same money to multiple vendors.
I imagine the Peppercorn is currently proposing a simpler solution that integrates better with existing technology. Either way, when a payments are sufficiently large, the system could just defer to current technology, so the vendor is no worse off.
Biometrics can be pretty problematic in certain situations. Take one generated from a finger print. It works great if you can take someone's fingerprint on your own, secure equipment. It doesn't do much good though if someone just sends you their fingerprint data remotely. For remote transactions, a biometric may become no different than a really long credit card number. The key difference being that if someone steals your fingerprint, you damn sure won't be able to get a new one.
Preventing a simple number (even a long one) from allowing someone to steal your identity is the right idea.
A credit card with a chip, for instance, can respond to a query that validates the card. If a credit card company only allows each query/response to be used once, saved information becomes useless. For online purchases, software provided by the credit card company could play the roll of the card.
This is just one example, but the bottom line is there are lots of ways to prevent the information we provide companies from being useful in stealing our identity.
Also, a much simpler approach is to keep people as informed as possible. It should be easy to find out if your social security number has been used for anything. Credit card companies already allow you to recieve an e-mail if your card balance changes, and banks should do the same.
For a completely blind user-interface (which seems to be the much greater challenge), people seem too set on braille/tactile interfaces. Braille is certainly a nice way to output text, but as someone posted, not all blind people know braille. Additionally, almost no sighted people know braille, which means a blind person couldn't use the computer with their sighted friend, nor could a sighted person help the blind person use their computer. More elaborate tactile interfaces would have similar short-comings, and what's worse, they would rely on technology that doesn't really exist yet (or rather, is just starting to exist). Also, though a text based interface is an easy solution, it's not a good one. There's no reason why your average blind person would like a text interface anymore than a sighted person. The interface should mimic the way a blind peson interacts with the world. They should have the ability to navigate in a space (perhaps with a mouse or 3D pointing device) and recieve feedback. The feedback should be given both in a tactile fashion and through sound. Using a pointing device you could feel when you "hit" an icon, and then hear it's name. Sound spacialization also offers a wealth of possiblities. Blind people would be quite capable of naviagating through a virtual space in which sounds and words suggested where things were. Ear-bud headphones along with the proper software could potentially do an excellent job of simulating 3D sound (check out a "binaural recording" to hear for yourself). Properly placed speakers would also work. Obviously there is lots to consider, but a good 3D sound-based interface might be useful for sighted users as well, allowing for sophisticated control of devices without screens (I could imagine, for instance, an iPod shuffle that let you navigate through a much larger music collection by reading the names of bands)