Couldn't the same be said for Google? Isn't Google the default search engine for Firefox, Chrome, Safari, and Opera? Doesn't Google toolbar come pre-installed on some machines? Isn't Google the default search engine on every iPhone and Android device?
I find Bing maps to be much better than Google's. At least for my area, bing has higher resolution maps, and the Bird's eye view is a nifty feature: view and location from any angle. I also thing Bing maps has better transitions for zooming. Zoom in real far on Google maps, then zoom out very fast. Your old position will be a small square in a sea of gray, where the new images haven't loaded yet. On Bing maps you get more transitions as you zoom out.
I'm actually using satellite images for part of my research, and I chose Bing's over Google's for just this reason.
Also, what exactly is wrong with Bing's results? Generally, I don't think I've found any deficiencies from using it. If anything, I've been finding more link farms at the top of Google results lately.
You're paying for OS X, for an aluminum unibody, for an awesome keyboard and high-res screen.
Care to explain my HP Envy, with aluminum/magnesium body, backlit keyboard, slot loading dvd, 128gb SSD, Radeon HD 5650, i5, 1600x900 screen?
Oh, and I bought it for $980 ($1400 before rebate. I miss you 30% BCB). The closest Macbook configuration (at the time) cost almost $3000, and couldn't even match in some specs (like the video card). Is OSX worth almost $1600? And people scoff at the price of Windows 7.
Supported resolutions: 1280 by 800 (native), 1152 by 720, 1024 by 640, and 800 by 500 pixels at 16:10 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio stretched; 720 by 480 pixels at 3:2 aspect ratio; 720 by 480 pixels at 3:2 aspect ratio stretched
15-Inch
Supported resolutions: 1440 by 900 (native), 1280 by 800, 1152 by 720, 1024 by 640, and 800 by 500 pixels at 16:10 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio stretched; 720 by 480 pixels at 3:2 aspect ratio; 720 by 480 pixels at 3:2 aspect ratio stretched
17-inch
Supported resolutions: 1920 by 1200 (native), 1680 by 1050, 1280 by 800, 1152 by 720, 1024 by 640, and 800 by 500 pixels at 16:10 aspect ratio; 1280 by 1024 pixels at 5:4 aspect ratio; 1280 by 1024 pixels at 5:4 aspect ratio stretched; 1600 by 1200, 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio; 1600 by 1200, 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio stretched; 720 by 480 pixels at 3:2 aspect ratio; 720 by 480 pixels at 3:2 aspect ratio stretched
the kinect isn't revolutionary because it holds a very small percentage of the market.
What market? I don't know a robotics lab without at least a couple kinects, or a robotics hobbyist, who doesn't have one or isn't planning to get one.
As many have posted, it is being used in robotic research right now. It has the potential to revolutionize that field. It just hasn't had time to do that, yet.
I do robotics research, and I can uneqivocally tell you it has already changed the course of robotics in a way we never thought possible. On of the biggest problems in robotics, and even in computer science in general, is reproducibility. Every lab has their own sensors, their own robot platforms, and their own software to run their robots. If I develop an algorithm for detecting pole features in my lab, there's no guarantee you can get it to work on your robot without a lot of work, if at all.
ROS is one part of the equation, as it provides a common software basis for robotics development. This is great, and in itself is revolutionary, but we're still not there unless labs can share a common sensor to capture data with. The kinect is this sensor, and it has allowed robotics researches to develop algorithms, publish them, and reproduce them in a way never before seen in robotics.
Even further, hobbyists can download the very same cutting-edge algorithms being developed in universities and use them on their robots. Maybe they can even improve them! The Kinect sensor, along with ROS, is taking robotics development out of the universities and putting it into the hands of actual users. This is not a small feat, and I feel like you're marginalizing it without really understanding what the implications are.
Do you really believe that future cars will have a kinect sitting in the grill for accident avoidance?
Actually they're using sensors from Velodyne, which are doing to LIDARs what the Kinect is doing to 3D sensors. It used to be that you needed an array of 2D LIDARS to create a 3D image. These could cost upwards of $100,000, where prone to failure, time consuming to create, and one of a kind. Then came Velodyne with their $70,000 3D LIDAR now being used on any serious autonomous vehicle. Of all the cars that finished the DARPA urban challenge, only one didn't use a Velodyne. Even Google's autonomous car has one. Now Velodyne released a new model for $20,000 the size of a coffee mug.
Yes in 2005 we could create a 3D image using lidars. That was 5 years ago, and at the time we couldn't get a car to drive through the desert. Now, because of Velodyne and the ubiquity of their sensors, cars are driving themselves through crowded streets.
Was the Velodyne revolutionary? Absolutely. Was it brand new? No, we had 3D sensors on our cars before. But it was smaller, cheaper, and easier to integrate, which is exactly what the kinect is.
First, I enjoy following your blog immensely. You always bring new stuff to my attention, and I appreciate that.
You are correct, even my lab bought Kinects as soon as we could. Actually, I personally bought one first, and after I showed my advisor what I was doing, he went and bought some more immediately.
I think the main value of the Kinect is in standardization. DARPA sees standardization as a problem for robotic manipulator research, which is why they're providing a standard base for their ARM competition. I see the same problem in sensing, where every lab has their own sensor suite. If lab1 comes up with a great algorithm for mapping, but it requires special data as input, lab2 can't replicate lab1's work without spending possibly upwards of $10,000. Now everyone has the same sensor, we have the same software, and I can download another lab's research and have it working on my robot in an hour. This is amazing, and unprecedented. ROS did it with software, and now Kinect is doing it with hardware.
I think for the near and intermediate future, the Kinect will maintain its utility, as robots are nowhere near ubiquity. A Kinect is safe for now on your health care robot, because there's probably ever going to be only one in a room at a time. This is the way I see things panning out: Kinect makes robots cheaper and more useful, so they find their way out of academia into industry and the home. As the market for robots grows, current sensor manufacturers increase production, R&D, etc. to meet demand. When the day comes that you need two health care robots in the hospital, the swissranger or its offspring might cost $500.
At any rate, it's an exciting time to be developing robots.
Yugo didn't decrease the cost of an automobile 10 fold while simultaneously increasing the quality. It isn't JUST that the Kinect is so cheap. I can buy an sharp IR range finder for $15, 10 times cheaper than even the Kinect. But in terms of resolution, it's not 24 times better, like the Kinect is compared to the IFM sensors I'm currently using. If Yugo cost $300 and got 100mpg, then we'd be talking about revolutionary.
A perfect example of this happening before is sitting right in front of you. Computers used to take up entire rooms. These had niche application in business, government, and academia, but the computer revolution didn't happen until they were cheap enough for anyone to afford. Not only that but as they became cheaper they also became better!
In terms of your definition, Kinect meets this perfectly. It has had a major sudden aspect on an aspect of human endeavor, specifically robotics. Society has changed in that anyone can now afford to integrate what used to be expensive data into their robotics project. It's even further revolutionizing academic research, which you claim has been doing this all along, and therefore should be no big deal. We now all have the same sensors. I can develop an algorithm in my lab, publish it on the internet, and anyone in another research lab can download it and use it without worrying about configuration, compatibility, differences between sensors. Not only that, any hobbyist can download my state of the art algorithm and use it on their personal robotics project. This isn't something that might happen one day, it's.
I am not doubting that the Kinect is changing the game. I just question whether it is revolutionary or not.
Kinect is changing the game but it's not revolutionary? What's your definition of a revolution then? Before the kinect, it cost me close to $10,000 for a good 3d point cloud data. If I had more room on my robot, I might put a Hokuyo LIDAR on a pivot but that still put me back 6 grand. Today I use industrial sensors from IFM, re-purposed for Robotics. They cost about $1500, and only provide 50x64 pixels of range data, as compared to the Kinect's 320x240.
So the cheapest feasible sensor I can buy costs $1500. So here comes Microsoft. They're selling a sensors 10 times cheaper with 24 time the resolution. Now any old schmuck can buy this and test their idea for a new image segmentation algorithm. This has NEVER been possible before.
So yeah, Kinect is changing the game. That's the definition of a revolution. Just because it was done in a lab before by Ph.D.s after 10s of thousands of dollars of time, effort, and equipment doesn't diminish it. If a company started selling robot cars to the public, that would be revolutionary too, even though we can do that in the lab (for $1,000,000+).
And Microsoft can't get all the credit; none of this would be possible without ROS and the amazing Point Cloud Library. This is a second component of the kinect revolution, which, in itself is revolutionary.
I love the kinect; I've done some great stuff in my robotics research with it already. It's a great sensor for testing out algorithms because of the high definition of the data, but it's next to useless as a long term solution for mobile robotics due to the nature of structure light sensors; the dot pattern projected by the IR camera can be easily interfered with by other kinects.
While there has been one example of two cameras working orthogonally, I can't see it expanding much more beyond that. To use more than a couple, you'd have to time the sensors to work together, or something more ingenious. Regardless, right now they're great in the lab, but the state of mobile robotics is still such that good sensors cost >$10,000.
It literally has cell phone hardware inside of it. This is a fact, not opinion. It says nothing about the relative merit of the iPad. I own one, it's a fine device. Seriously, chill the fuck out.
Also, while Apple does heavily advertise, they don't have to pay for a super bowl add to reach millions of customers. When the iPad was announced it was on every local news channel in america (at 6:00, 12:00, and 11:00), national news, local radio, every freaking blog on the Internet, newspapers, late night talk shows, forums, etc. If you consume some sort of media, you heard about the iPad. How many local news outlets are reporting on the XOOM?
No one was really predicting a phone with a 10" screen. 256MB RAM, 1GHz single core processor, no cameras, 1024x768 display, no ports. When people were guessing 800-1000 they were really envisioning something more along the lines of a OSX Tablet, rather than an iOS tablet.
I mean the only real world example. If Google results only show up in Bing in the contrived situation where the only signal comes from user generated data, and it's remarkably strong, Google has nothing to complain about. Even in their test only a small fraction of those terms they designed made it into Bing's results. They illustrated an extreme corner case unlikely to occur in the real world, and they can't point to any real concrete example.
The first one is to exchange geography. As robots move around, they build maps of their environment.
This is exactly what we're working on in my lab. High definition 3D LIDARS are very expensive (~70,000) and also very large. They'll fit on a robot car, like Google's, but not on something smaller. But what if Google car saved its expensive high-def maps to a network, where any robot could access them. Then, a small robot with a cheap laser could download the maps for localization and path planning. The problem is scan matching between the two laser maps in a process known as sensor fusion.
But essentially what this does is create a Google Maps for robots. They log on to a network to get directions and maps just as a human logs on to google. The difference is, they can do this on the fly, as they will be networked.
Bing uses this data to provide 'top results' that it obviously values above those provided by its own algorithm.
In fact, Google's own experiment proves just the opposite. They created search -> result mappings guaranteed to have zero signal from every Bing algorithm except the toolbar. Of the fake mappings they generated, only a small fraction made it into Bing results. This only illustrates how the algorithm behaves in an extreme corner case. Google is not able to point to any legitimate search results where it's clear that Bing obtained those from Google. Even the only example they do point to, "torsoraphy," is questionable, since it's just as, if not more likely that could have been combed from users on Wikipedia.
Essentially Bing's defense (as outlined in the article) goes like this:
Bing is monitoring users who opted in to send Bing data. They are watching their activity on any site, and not specifically Google.
The search signal generated by users does not dominate, unless it's the only signal (as Google tried to ensure it would be) it will have more weight, but not absolute. Even Google's test showed this to be true, as only a fraction of their honeypot terms made it to the other side.
Less frequent seach terms (the example given is pontneddfechan) Bing's results are relevant, unique, and ordered differently from Google's. Google's tests reveal the very special case where 0 signal comes from other sources.
What's the BFD in the end? Google alleges Bing is stealing results, but only shows one concrete example of this (tarsorrhaphy), which can be easily accounted for by crawling Wikipedia, which seems much more likely.
In addition, if this were not a major signal in their ranking, they'd likely stop using it to get away from the controversy. The fact that they're trying to dance around the issue rather than removing the signal proves that a major source of their search relevancy is Google search results.
And yet only a fraction of Google's injected terms made it into Bing's results. If it were such a major part, all of them would have. All this does is show that given sparse information from other indicators, and a very strong indication from customer feedback, Bing will take into account customer feedback.
It's also irrelevant. Whether the users know or not, whether they gave permission or not, Bing is still receiving *and using* direct search data from another search engine! The click-through is monitoring two pieces of data and then providing that back to Bing - the search term entered into Google, and the page the user goes to after the results page is returned.
No, this is very relevant. The piece of data Microsoft is interested in is the user's selection based on a search term. They aren't interested that it appeared on Google's search results, they aren't interested where it ranked in Google's terms. They just care that given a search term, the user found this link relevant enough to click on.
The reason this is relevant is because the USER owns this data, not Google. And the USER agreed to report this information to Microsoft. You might say that the mapping between search term and link is owned by Google or somehow Google's property, but in this day, this mapping is trivial to generate. Every engine out there is crawling and indexing, and for the most part, the mapping between a search term and the set of search results is generated by matching words. That is, if I search for "Hammer", the set of sites returned will be those containing the words "Hammer".
The difficulty is ordering these results. If I search "Hammer" do I want hammer manufacturers, hammer resellers, or MC Hammer? This is a hard problem compared to generating the mapping. Therefore, the real value of the data Microsoft collects is not the list of sites, but the site selected amongst a list of sites, and that again, that is NOT Google's data.
If their own web crawling and search data was halfway competent it would have identified that the search terms didn't match the results in any meaningful way
Wrong, Google gamed Bing's algorithm to break it. When Google did this "experiment" they installed Bing Toolbar. When they did this, Microsoft said "Hey, we would like to see how you use the internet to improve Bing. We would like your permission to use your searches to do this." to which Google replied "Sure thing. We'll send you our preferences and you use them to make Bing better." Google then proceeded to send the data they promised to send, although it was fake. What they sent was a mapping between adzxcsafas -> rim.com or any other random phrase with a nonsensical result. On receipt of over 20 instances of this pairing, Bing eventually said "It looks like people who are searching for "adzxcsafas" find the site 'im.com' useful. This makes no sense to us, but obviously the data says otherwise so we'll go ahead and add it."
The fact that Google engineers were gaming the algorithm is very very important, because aside from Google engineers purposefully injecting dummy data, this method would help reorder searches in an already established mapping. That is, it is an extremely remote case that customer generated data would create a mapping between search and site. What Google illustrated is a fringe behavior of the data collection algorithm, that when presented with extremely sparse information Bing trusts user preferences for searches over their own algorithms, which I feel is a fine assumption to make. Further, Google seeded Bing with other 100 fake mappings and only a fraction of these appeared in Bing's results, meaning there this data is being injected into some decision process. This implies Bing is not just combing Google for results, but instead applying heuristics to customer-generated data to decide rankings. The former is reprehensible, the latter is a perfectly legitimate tool.
Either way, they are relying on the results of other search engines to bolster or ammend their own.
Pointing an algorithm to Google.com and telling it to figure out how they rank their results is wrong, yes. Again, that is not what is going on here. First, the customer data is a small part in a large decision process for reorderi
Slashdot Mirror
Couldn't the same be said for Google? Isn't Google the default search engine for Firefox, Chrome, Safari, and Opera? Doesn't Google toolbar come pre-installed on some machines? Isn't Google the default search engine on every iPhone and Android device?
I find Bing maps to be much better than Google's. At least for my area, bing has higher resolution maps, and the Bird's eye view is a nifty feature: view and location from any angle. I also thing Bing maps has better transitions for zooming. Zoom in real far on Google maps, then zoom out very fast. Your old position will be a small square in a sea of gray, where the new images haven't loaded yet. On Bing maps you get more transitions as you zoom out.
I'm actually using satellite images for part of my research, and I chose Bing's over Google's for just this reason.
Also, what exactly is wrong with Bing's results? Generally, I don't think I've found any deficiencies from using it. If anything, I've been finding more link farms at the top of Google results lately.
I don't think acronyms can be defined recursively.
You're paying for OS X, for an aluminum unibody, for an awesome keyboard and high-res screen.
Care to explain my HP Envy, with aluminum/magnesium body, backlit keyboard, slot loading dvd, 128gb SSD, Radeon HD 5650, i5, 1600x900 screen?
Oh, and I bought it for $980 ($1400 before rebate. I miss you 30% BCB). The closest Macbook configuration (at the time) cost almost $3000, and couldn't even match in some specs (like the video card). Is OSX worth almost $1600? And people scoff at the price of Windows 7.
I guess to be fair, Microsoft does have a windows icon.... of a broken, battered window pane. http://a.fsdn.com/sd/topics/windows_64.png
Supported resolutions: 1280 by 800 (native), 1152 by 720, 1024 by 640, and 800 by 500 pixels at 16:10 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio stretched; 720 by 480 pixels at 3:2 aspect ratio; 720 by 480 pixels at 3:2 aspect ratio stretched
15-Inch
Supported resolutions: 1440 by 900 (native), 1280 by 800, 1152 by 720, 1024 by 640, and 800 by 500 pixels at 16:10 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio stretched; 720 by 480 pixels at 3:2 aspect ratio; 720 by 480 pixels at 3:2 aspect ratio stretched
17-inch
Supported resolutions: 1920 by 1200 (native), 1680 by 1050, 1280 by 800, 1152 by 720, 1024 by 640, and 800 by 500 pixels at 16:10 aspect ratio; 1280 by 1024 pixels at 5:4 aspect ratio; 1280 by 1024 pixels at 5:4 aspect ratio stretched; 1600 by 1200, 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio; 1600 by 1200, 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio stretched; 720 by 480 pixels at 3:2 aspect ratio; 720 by 480 pixels at 3:2 aspect ratio stretched
http://www.apple.com/macbookpro/specs.html
They have the Apple logo, iOS logo, iPhone, and Macbook. Why does apple get so many special Slashdot icons?
Microsoft pays dividends.
I think he meant "preeminent." Still a pretty bad mistake.
the kinect isn't revolutionary because it holds a very small percentage of the market.
What market? I don't know a robotics lab without at least a couple kinects, or a robotics hobbyist, who doesn't have one or isn't planning to get one.
As many have posted, it is being used in robotic research right now. It has the potential to revolutionize that field. It just hasn't had time to do that, yet.
I do robotics research, and I can uneqivocally tell you it has already changed the course of robotics in a way we never thought possible. On of the biggest problems in robotics, and even in computer science in general, is reproducibility. Every lab has their own sensors, their own robot platforms, and their own software to run their robots. If I develop an algorithm for detecting pole features in my lab, there's no guarantee you can get it to work on your robot without a lot of work, if at all.
ROS is one part of the equation, as it provides a common software basis for robotics development. This is great, and in itself is revolutionary, but we're still not there unless labs can share a common sensor to capture data with. The kinect is this sensor, and it has allowed robotics researches to develop algorithms, publish them, and reproduce them in a way never before seen in robotics.
Even further, hobbyists can download the very same cutting-edge algorithms being developed in universities and use them on their robots. Maybe they can even improve them! The Kinect sensor, along with ROS, is taking robotics development out of the universities and putting it into the hands of actual users. This is not a small feat, and I feel like you're marginalizing it without really understanding what the implications are.
Do you really believe that future cars will have a kinect sitting in the grill for accident avoidance?
Actually they're using sensors from Velodyne, which are doing to LIDARs what the Kinect is doing to 3D sensors. It used to be that you needed an array of 2D LIDARS to create a 3D image. These could cost upwards of $100,000, where prone to failure, time consuming to create, and one of a kind. Then came Velodyne with their $70,000 3D LIDAR now being used on any serious autonomous vehicle. Of all the cars that finished the DARPA urban challenge, only one didn't use a Velodyne. Even Google's autonomous car has one. Now Velodyne released a new model for $20,000 the size of a coffee mug.
Yes in 2005 we could create a 3D image using lidars. That was 5 years ago, and at the time we couldn't get a car to drive through the desert. Now, because of Velodyne and the ubiquity of their sensors, cars are driving themselves through crowded streets.
Was the Velodyne revolutionary? Absolutely. Was it brand new? No, we had 3D sensors on our cars before. But it was smaller, cheaper, and easier to integrate, which is exactly what the kinect is.
First, I enjoy following your blog immensely. You always bring new stuff to my attention, and I appreciate that.
You are correct, even my lab bought Kinects as soon as we could. Actually, I personally bought one first, and after I showed my advisor what I was doing, he went and bought some more immediately.
I think the main value of the Kinect is in standardization. DARPA sees standardization as a problem for robotic manipulator research, which is why they're providing a standard base for their ARM competition. I see the same problem in sensing, where every lab has their own sensor suite. If lab1 comes up with a great algorithm for mapping, but it requires special data as input, lab2 can't replicate lab1's work without spending possibly upwards of $10,000. Now everyone has the same sensor, we have the same software, and I can download another lab's research and have it working on my robot in an hour. This is amazing, and unprecedented. ROS did it with software, and now Kinect is doing it with hardware.
I think for the near and intermediate future, the Kinect will maintain its utility, as robots are nowhere near ubiquity. A Kinect is safe for now on your health care robot, because there's probably ever going to be only one in a room at a time. This is the way I see things panning out: Kinect makes robots cheaper and more useful, so they find their way out of academia into industry and the home. As the market for robots grows, current sensor manufacturers increase production, R&D, etc. to meet demand. When the day comes that you need two health care robots in the hospital, the swissranger or its offspring might cost $500.
At any rate, it's an exciting time to be developing robots.
So by that definition, a Yugo is revolutionary.
Yugo didn't decrease the cost of an automobile 10 fold while simultaneously increasing the quality. It isn't JUST that the Kinect is so cheap. I can buy an sharp IR range finder for $15, 10 times cheaper than even the Kinect. But in terms of resolution, it's not 24 times better, like the Kinect is compared to the IFM sensors I'm currently using. If Yugo cost $300 and got 100mpg, then we'd be talking about revolutionary.
A perfect example of this happening before is sitting right in front of you. Computers used to take up entire rooms. These had niche application in business, government, and academia, but the computer revolution didn't happen until they were cheap enough for anyone to afford. Not only that but as they became cheaper they also became better!
In terms of your definition, Kinect meets this perfectly. It has had a major sudden aspect on an aspect of human endeavor, specifically robotics. Society has changed in that anyone can now afford to integrate what used to be expensive data into their robotics project. It's even further revolutionizing academic research, which you claim has been doing this all along, and therefore should be no big deal. We now all have the same sensors. I can develop an algorithm in my lab, publish it on the internet, and anyone in another research lab can download it and use it without worrying about configuration, compatibility, differences between sensors. Not only that, any hobbyist can download my state of the art algorithm and use it on their personal robotics project. This isn't something that might happen one day, it's .
I am not doubting that the Kinect is changing the game. I just question whether it is revolutionary or not.
Kinect is changing the game but it's not revolutionary? What's your definition of a revolution then? Before the kinect, it cost me close to $10,000 for a good 3d point cloud data. If I had more room on my robot, I might put a Hokuyo LIDAR on a pivot but that still put me back 6 grand. Today I use industrial sensors from IFM, re-purposed for Robotics. They cost about $1500, and only provide 50x64 pixels of range data, as compared to the Kinect's 320x240.
So the cheapest feasible sensor I can buy costs $1500. So here comes Microsoft. They're selling a sensors 10 times cheaper with 24 time the resolution. Now any old schmuck can buy this and test their idea for a new image segmentation algorithm. This has NEVER been possible before.
So yeah, Kinect is changing the game. That's the definition of a revolution. Just because it was done in a lab before by Ph.D.s after 10s of thousands of dollars of time, effort, and equipment doesn't diminish it. If a company started selling robot cars to the public, that would be revolutionary too, even though we can do that in the lab (for $1,000,000+).
And Microsoft can't get all the credit; none of this would be possible without ROS and the amazing Point Cloud Library. This is a second component of the kinect revolution, which, in itself is revolutionary.
I love the kinect; I've done some great stuff in my robotics research with it already. It's a great sensor for testing out algorithms because of the high definition of the data, but it's next to useless as a long term solution for mobile robotics due to the nature of structure light sensors; the dot pattern projected by the IR camera can be easily interfered with by other kinects.
While there has been one example of two cameras working orthogonally, I can't see it expanding much more beyond that. To use more than a couple, you'd have to time the sensors to work together, or something more ingenious. Regardless, right now they're great in the lab, but the state of mobile robotics is still such that good sensors cost >$10,000.
It literally has cell phone hardware inside of it. This is a fact, not opinion. It says nothing about the relative merit of the iPad. I own one, it's a fine device. Seriously, chill the fuck out.
Also, while Apple does heavily advertise, they don't have to pay for a super bowl add to reach millions of customers. When the iPad was announced it was on every local news channel in america (at 6:00, 12:00, and 11:00), national news, local radio, every freaking blog on the Internet, newspapers, late night talk shows, forums, etc. If you consume some sort of media, you heard about the iPad. How many local news outlets are reporting on the XOOM?
No one was really predicting a phone with a 10" screen. 256MB RAM, 1GHz single core processor, no cameras, 1024x768 display, no ports. When people were guessing 800-1000 they were really envisioning something more along the lines of a OSX Tablet, rather than an iOS tablet.
The only example? you didn't read much...
I mean the only real world example. If Google results only show up in Bing in the contrived situation where the only signal comes from user generated data, and it's remarkably strong, Google has nothing to complain about. Even in their test only a small fraction of those terms they designed made it into Bing's results. They illustrated an extreme corner case unlikely to occur in the real world, and they can't point to any real concrete example.
The first one is to exchange geography. As robots move around, they build maps of their environment.
This is exactly what we're working on in my lab. High definition 3D LIDARS are very expensive (~70,000) and also very large. They'll fit on a robot car, like Google's, but not on something smaller. But what if Google car saved its expensive high-def maps to a network, where any robot could access them. Then, a small robot with a cheap laser could download the maps for localization and path planning. The problem is scan matching between the two laser maps in a process known as sensor fusion.
But essentially what this does is create a Google Maps for robots. They log on to a network to get directions and maps just as a human logs on to google. The difference is, they can do this on the fly, as they will be networked.
A search for "torsorophy" on Wikipedia suggests the correct spelling.
Bing uses this data to provide 'top results' that it obviously values above those provided by its own algorithm.
In fact, Google's own experiment proves just the opposite. They created search -> result mappings guaranteed to have zero signal from every Bing algorithm except the toolbar. Of the fake mappings they generated, only a small fraction made it into Bing results. This only illustrates how the algorithm behaves in an extreme corner case. Google is not able to point to any legitimate search results where it's clear that Bing obtained those from Google. Even the only example they do point to, "torsoraphy," is questionable, since it's just as, if not more likely that could have been combed from users on Wikipedia.
I think this article says everything that needs to be said on the issue:
http://searchengineland.com/bing-why-googles-wrong-in-its-accusations-63279
Essentially Bing's defense (as outlined in the article) goes like this:
In addition, if this were not a major signal in their ranking, they'd likely stop using it to get away from the controversy. The fact that they're trying to dance around the issue rather than removing the signal proves that a major source of their search relevancy is Google search results.
And yet only a fraction of Google's injected terms made it into Bing's results. If it were such a major part, all of them would have. All this does is show that given sparse information from other indicators, and a very strong indication from customer feedback, Bing will take into account customer feedback.
It's also irrelevant. Whether the users know or not, whether they gave permission or not, Bing is still receiving *and using* direct search data from another search engine! The click-through is monitoring two pieces of data and then providing that back to Bing - the search term entered into Google, and the page the user goes to after the results page is returned.
No, this is very relevant. The piece of data Microsoft is interested in is the user's selection based on a search term. They aren't interested that it appeared on Google's search results, they aren't interested where it ranked in Google's terms. They just care that given a search term, the user found this link relevant enough to click on.
The reason this is relevant is because the USER owns this data, not Google. And the USER agreed to report this information to Microsoft. You might say that the mapping between search term and link is owned by Google or somehow Google's property, but in this day, this mapping is trivial to generate. Every engine out there is crawling and indexing, and for the most part, the mapping between a search term and the set of search results is generated by matching words. That is, if I search for "Hammer", the set of sites returned will be those containing the words "Hammer".
The difficulty is ordering these results. If I search "Hammer" do I want hammer manufacturers, hammer resellers, or MC Hammer? This is a hard problem compared to generating the mapping. Therefore, the real value of the data Microsoft collects is not the list of sites, but the site selected amongst a list of sites, and that again, that is NOT Google's data.
If their own web crawling and search data was halfway competent it would have identified that the search terms didn't match the results in any meaningful way
Wrong, Google gamed Bing's algorithm to break it. When Google did this "experiment" they installed Bing Toolbar. When they did this, Microsoft said "Hey, we would like to see how you use the internet to improve Bing. We would like your permission to use your searches to do this." to which Google replied "Sure thing. We'll send you our preferences and you use them to make Bing better." Google then proceeded to send the data they promised to send, although it was fake. What they sent was a mapping between adzxcsafas -> rim.com or any other random phrase with a nonsensical result. On receipt of over 20 instances of this pairing, Bing eventually said "It looks like people who are searching for "adzxcsafas" find the site 'im.com' useful. This makes no sense to us, but obviously the data says otherwise so we'll go ahead and add it."
The fact that Google engineers were gaming the algorithm is very very important, because aside from Google engineers purposefully injecting dummy data, this method would help reorder searches in an already established mapping. That is, it is an extremely remote case that customer generated data would create a mapping between search and site. What Google illustrated is a fringe behavior of the data collection algorithm, that when presented with extremely sparse information Bing trusts user preferences for searches over their own algorithms, which I feel is a fine assumption to make. Further, Google seeded Bing with other 100 fake mappings and only a fraction of these appeared in Bing's results, meaning there this data is being injected into some decision process. This implies Bing is not just combing Google for results, but instead applying heuristics to customer-generated data to decide rankings. The former is reprehensible, the latter is a perfectly legitimate tool.
Either way, they are relying on the results of other search engines to bolster or ammend their own.
Pointing an algorithm to Google.com and telling it to figure out how they rank their results is wrong, yes. Again, that is not what is going on here. First, the customer data is a small part in a large decision process for reorderi