If the Google patent is truly for something that is already known then it should not have been issued. I did not read the whole patent. Patents always have to be for something new that is not yet known by others: http://www.uspto.gov/patents/basics.jsp#novelty. If what Google patent was already known (and I'm not saying that it is because I didn't read the whole thing) then it can be challenged and overturned in court. It can be challenged in federal district court, appealed to the Federal Circuit, and appealed to the Supreme Court.
It is not true that if Google doesn't patent it, a troll will. A technique that is well known, such as MapReduce, is the property of the general public and is unpatentable. Any technology that has been sold or in use for over a year is unpatentable.
Cyc is a controversial project in the AI community, and I'm glad that you brought it up. I don't think anyone yet knows how to use a database of commonsense facts, which is what Cyc is (though limited - the open source version only has a few hundred thousand facts) and which is one thing NELL could create. However, researchers continue to think about ways that an AI could use knowledge of the real world. There are numerous publications based on Cyc: http://www.opencyc.org/cyc/technology/pubs.
It's unfortunate that the Chinese government continues to spy on the population. Many text messages that people send are quite personal.
Text message technology actually makes it easier to spy on people because you can just filter for words like "democracy" rather than actually having to pay an operator to listen to people's phone conversations. Many human rights activists in China had previously reported having their phones tapped.
Particularly interesting is the following clause indicating that your electronic devices can be searched even if the officer has no good reason to think you are 'suspicious.'
5.1.2. In the course of a border search, with or without individualized suspicion, an Officer may examine electronic devices and may review and analyze information encountered at the border, subject to the requirements herein and applicable law.
I suppose this is the same as the right of officers to open everyone's bags, without any need of proving suspicion. This is quite an invasion of people's privacy and property, yet realistically I don't think this will endanger attorney-client privilege or trade secrets. The officer searching you probably searches thousands of people a day. It's not like he's going to go through your data files and memorize all the important business/legal documents and then report them to your competitors. The policy document indicates that all electronic searches take place in your presence and with a supervisor present.
According to CNN, Tiananmen is the proper spelling and Tienanmen is a misspelling that is not properly censored due to technical errors. Apparently those errors have not been fixed since 2006.
http://money.cnn.com/2006/01/30/technology/browser0130/index.htm
Thanks for the link though. Impressive number of tank man pictures.
I hope Google does provide uncensored search, even if for just a few hours.
As other people have pointed out any sort of computer device must store and send data. (1) Any modern computer will have a framebuffer that stores the image temporarily. (2) The signal has to be sent from the scanners to the display, so someone could open the machine and could redirect the signal to go somewhere else like a network.
It makes more sense to ask how easy it is to store and send data. There isn't enough information to determine this.
"The official declined to say whether activating test mode requires additional hardware, software or simply additional knowledge of how the machines operate."
It would make a big difference whether the device has a hard drive or can write to portable media like a flash drive or writable CD. It would also make a big difference whether the device has the hardware to connect to networks (e.g. a NIC). We might ask similar questions about the device's software.
Without knowing the answer to those questions, it is hard to say whether the device is posing any danger in excess of that which is unavoidable by nature of its being a computer.
Final thought: It might be more sensible to secure the devices by relying on encryption rather than trying to preventing "storage" and "transmission." For example, one could encrypt the signal from the scanners as it was passed to the display, store the encrypted bits in the framebuffer, and only decipher the signal as it was passed to the display. The encryption mechanism would need to be a blackbox hardware device so the keys couldn't be snooped.
I agree that Jetpack will make it easier for developers to create apps and will also likely result in safer apps that don't fail as often. However, this is only apparent in hindsight, now that we realize writing add-ons with HTML/CSS/JS type technologies is probably smart. The fact is that Firefox has a significant number of extremely useful applications that might go beyond what is possible to implement with Jetpack. My business uses some Firefox extensions that are absolutely critical to us. I don't mind if Mozilla goes to Jetpack, but I think that they should keep support for traditional extensions. If they get rid of extensions, they will hurt a lot of people. Going to "Jetpack only" would make more sense if they were starting from a clean slate, but currently I think they have a responsibility to existing users.
Slashdot Mirror
If you click on the "patent applications" side it goes to http://appft.uspto.gov/ instead.
If the Google patent is truly for something that is already known then it should not have been issued. I did not read the whole patent. Patents always have to be for something new that is not yet known by others: http://www.uspto.gov/patents/basics.jsp#novelty. If what Google patent was already known (and I'm not saying that it is because I didn't read the whole thing) then it can be challenged and overturned in court. It can be challenged in federal district court, appealed to the Federal Circuit, and appealed to the Supreme Court.
http://www.uspto.gov/patents/basics.jsp#novelty
All documents at http://patft.uspto.gov/ are issued patents.
It is not true that if Google doesn't patent it, a troll will. A technique that is well known, such as MapReduce, is the property of the general public and is unpatentable. Any technology that has been sold or in use for over a year is unpatentable.
Cyc is a controversial project in the AI community, and I'm glad that you brought it up. I don't think anyone yet knows how to use a database of commonsense facts, which is what Cyc is (though limited - the open source version only has a few hundred thousand facts) and which is one thing NELL could create. However, researchers continue to think about ways that an AI could use knowledge of the real world. There are numerous publications based on Cyc: http://www.opencyc.org/cyc/technology/pubs.
It's unfortunate that the Chinese government continues to spy on the population. Many text messages that people send are quite personal.
Text message technology actually makes it easier to spy on people because you can just filter for words like "democracy" rather than actually having to pay an operator to listen to people's phone conversations. Many human rights activists in China had previously reported having their phones tapped.
Particularly interesting is the following clause indicating that your electronic devices can be searched even if the officer has no good reason to think you are 'suspicious.'
5.1.2. In the course of a border search, with or without individualized suspicion, an Officer may examine electronic devices and may review and analyze information encountered at the border, subject to the requirements herein and applicable law.
I suppose this is the same as the right of officers to open everyone's bags, without any need of proving suspicion. This is quite an invasion of people's privacy and property, yet realistically I don't think this will endanger attorney-client privilege or trade secrets. The officer searching you probably searches thousands of people a day. It's not like he's going to go through your data files and memorize all the important business/legal documents and then report them to your competitors. The policy document indicates that all electronic searches take place in your presence and with a supervisor present.
According to CNN, Tiananmen is the proper spelling and Tienanmen is a misspelling that is not properly censored due to technical errors. Apparently those errors have not been fixed since 2006. http://money.cnn.com/2006/01/30/technology/browser0130/index.htm Thanks for the link though. Impressive number of tank man pictures. I hope Google does provide uncensored search, even if for just a few hours.
Note that Adi Shamir is the "S" in RSA. The other two authors are from the Weizmann Institute and Hebrew University. Though the paper is not public yet, these guys seem like genuine crypto all stars. http://www.emergentchaos.com/archives/2010/01/another_week_another_gsm.html
As other people have pointed out any sort of computer device must store and send data. (1) Any modern computer will have a framebuffer that stores the image temporarily. (2) The signal has to be sent from the scanners to the display, so someone could open the machine and could redirect the signal to go somewhere else like a network.
It makes more sense to ask how easy it is to store and send data. There isn't enough information to determine this.
"The official declined to say whether activating test mode requires additional hardware, software or simply additional knowledge of how the machines operate."
It would make a big difference whether the device has a hard drive or can write to portable media like a flash drive or writable CD. It would also make a big difference whether the device has the hardware to connect to networks (e.g. a NIC). We might ask similar questions about the device's software.
Without knowing the answer to those questions, it is hard to say whether the device is posing any danger in excess of that which is unavoidable by nature of its being a computer.
Final thought: It might be more sensible to secure the devices by relying on encryption rather than trying to preventing "storage" and "transmission." For example, one could encrypt the signal from the scanners as it was passed to the display, store the encrypted bits in the framebuffer, and only decipher the signal as it was passed to the display. The encryption mechanism would need to be a blackbox hardware device so the keys couldn't be snooped.
I agree that Jetpack will make it easier for developers to create apps and will also likely result in safer apps that don't fail as often. However, this is only apparent in hindsight, now that we realize writing add-ons with HTML/CSS/JS type technologies is probably smart. The fact is that Firefox has a significant number of extremely useful applications that might go beyond what is possible to implement with Jetpack. My business uses some Firefox extensions that are absolutely critical to us. I don't mind if Mozilla goes to Jetpack, but I think that they should keep support for traditional extensions. If they get rid of extensions, they will hurt a lot of people. Going to "Jetpack only" would make more sense if they were starting from a clean slate, but currently I think they have a responsibility to existing users.