Why do people think this invention is so interesting?
It cleanses the surface of the skin of bacteria. There are already a huge number of cheap ways to do the same thing (alcohol, hydrogen peroxide, uv lights).
So is this cool just because it takes batteries and looks like light sabre jr?
When one says Lisp, they can mean a range of languages, including (most likely) one or both of: Common Lisp, which was designed by committee and has everything including the kitchen sink built in, and Scheme, which is much cleaner and smaller.
Python is more like Scheme and its philosophy than Common Lisp.
If you are in the SF area and very well versed in Java, C, or AJAX-like techniques, and looking to work on really interesting problems, let me know. Bonus points if you're a hacker (in the traditional sense).
Ah, but is it a diaeresis or an umlaut? If the former, it indicates a that the marked vowel is pronounced separately from the preceding vowel (noël, coöperate). If the latter, it modifies the sound of the vowel.
--Pat "diaeresis -- the little mark with the terrib le name"
Parent post: Must you invent events in order to push your point? The writer was not arrested, DNA profiled, and he didn't incur a police record.
Perhaps you should read the article. They took a DNA sample from they guy's cheek, and his solicitor (lawyer) told him that it was unlikely that the police would expunge the arrest from his record.
So, they do have his DNA on file and they have created a police record for him. Lovely.
This is a nice hack, but as a do-it-yourself sentry, well, it'll keep you very safe from intruders wearing bright white clothing. From the video, it looks like their software looks for the brightest region, and then aims at the center of that region.
Yes, I am aware that mice, as a system, are similar to humans, as are pigs, monkeys, and rabbits. However, it is also common to find huge differences in the way a gene or drug works in one system and another -- that life-saving drug that works on cells in vitro or in a non-human system often has different effects or side-effects in humans. So non-human systems are a good start, but they are never close enough to guarantee success. If they were, we wouldn't have human patient trials of drugs before general release.
A similar argument is true for genetic research. We use animals because, among other things, they are cheaper and good enough to begin with. But they are still far from the endpoint -- how the same gene is expressed and has an effect on the human body.
The article summary says "The trick... will be to find ways of getting the life-enhancing results of Klotho while avoiding the drawbacks"
No, the trick will be finding whether what they did with the mice applied to humans. Suppressing the side effects they found in mice is nothing by comparison.
While mice are similar in some ways to people, they are also rather different. What extends the life of lab mice might, in humans: a) have no effect, b) cause humans to sprout extra limbs, c) live longe and prosper, or d) none of the above.
And it's going to take a long time before they can try these experiments on humans.
Much of Google's older Usenet archive content was originally collected by DejaNews. At the time Deja went under, I and the other Archive folks were interested in getting it before it went to the great bit bucket in the sky. But I believe Google made a better or quicker offer for the data.
For what it's worth, the Internet Archive has at least at one point in its history collected Usenet posts. This isn't in the Wayback machine, though.
No, we do not collect or archive chat systems or personal email messages that have not been posted to Usenet bulletin boards or publicly accessible online message boards.
Yes, and this is something the Economist reporter would like to hear about. It's the question "yes, we know you're making money now, but how to you intend to make even more money in the future?"
Let's say Google has won the on-line ad business. How will they get more revenue? They can: 1) get more companies to advertise, 2) charge more to existing customers, or 3) introduce new services that make money and have high and growing earnings.
Bringing out the Chief Food Officer answers none of these questions. It is this that the Economist takes issue with. They expect a company to tell its stockholders something about the company intends to continue its growth. "We have great food" is not an answer.
The parent poster says "The answer... is that it doesn't matter at all when google will start delivering on that stock price to a long-term investor, and that they don't give a shit about day-traders gaming their stock."
The issue the Economist focused on, the price to earnings ratio, is one used by long-term investors to judge the relative worth of one investment over another. Specifically, P/E is used to predict the long-term performance of an investment.
If the price is low, and earnings are high, and the company has a history of high earnings, and is likely to continue to have high earnings, then it is likely that the company's stock is a good long term investment at the current price.
Google's extraordinarily high P/E ratio means that, at its current price, it must grow earnings astronomically (grow, not just maintain, and certainly not lose money) over the long-term in order to earn a good rate of return at the current price, relative to other stocks.
As another poster points out, hitting an Internet Archive page can result in requests to the original server. For instance, if the page contains Javascript that itself refers to a URL, the Archive may fail to rewrite the URL to point to the archive, leaving a reference to the original site.
A user pulling up this page on the Archive would generate a request to the original server, which would then show up in the original server's log.
However, the Internet Archive does use robots.txt as an access control mechanism. Part of IA's story is that, if you don't want to be crawled, you can opt out via robots.txt
If it turns out that they were not complying with their own policy (I do not know if this is the case) it could weaken their argument that they should be exempt from being sued for copyright infringement.
As I said in another post, they are basically getting a patent on a similarity function.
Imagine someone invents a new Go algorithm that, when implemented, beats Go masters. I think any CS researcher or Go player would say is a major breakthrough.
Now say this algorithm is a pattern matcher, a static evaluation function. Given the current board, the algorithm chooses the next move.
Following your reasoning, this would be a pattern matcher, and so could not be novel. Is this correct?
This is the standard way that patents are worded. To decode the patent to get to the underlying algorithm, you generally focus on the "preferred implementation" wording.
As other technology people have said, patents are a bit like Cobol -- a language of their own -- but the courts read them differently than a normal person, and give particular weight to the preferred implementation parts and much less to the "and everything else, including the kitchen sink" wording that surrounds this.
Actually, I was familiar with the credit card fraud work. I was working in that area in the mid-90s. And I'd say that fraud detection is different than recommending products. The systems of the era that I was personally familiar with using supervised learning techniques where large batches of bad (fraud) and good (non-fraud) examples were used to train up a learning system.
There may have been work on statistical detection of anomalies without prior measures of what was anomalous, but even if this existed (and I never heard of work in this area), there's no guarantee that an algorithm that detects credit card fraud will be good at recommending products as things one would want to purchase.
If I were a patent examiner, or even an expert witness, I'd say that taking a CC fraud algorithm and getting it to make good purchase recommendations in real time on a shopping site would be a novel application.
Slashdot Mirror
Why do people think this invention is so interesting?
It cleanses the surface of the skin of bacteria. There are already a huge number of cheap ways to do the same thing (alcohol, hydrogen peroxide, uv lights).
So is this cool just because it takes batteries and looks like light sabre jr?
--Pat
When one says Lisp, they can mean a range of languages, including (most likely) one or both of: Common Lisp, which was designed by committee and has everything including the kitchen sink built in, and Scheme, which is much cleaner and smaller.
Python is more like Scheme and its philosophy than Common Lisp.
--Pat
--Pat
--Pat
--Pat
--Pat "diaeresis -- the little mark with the terrib le name"
--Pat
Perhaps you should read the article. They took a DNA sample from they guy's cheek, and his solicitor (lawyer) told him that it was unlikely that the police would expunge the arrest from his record.
So, they do have his DNA on file and they have created a police record for him. Lovely.
--Pat
It's a cool hack.
--Pat
Tell me more about this laser-irradiated Borat.
--Pat
A similar argument is true for genetic research. We use animals because, among other things, they are cheaper and good enough to begin with. But they are still far from the endpoint -- how the same gene is expressed and has an effect on the human body.
--Pat
--Pat
No, the trick will be finding whether what they did with the mice applied to humans. Suppressing the side effects they found in mice is nothing by comparison.
While mice are similar in some ways to people, they are also rather different. What extends the life of lab mice might, in humans: a) have no effect, b) cause humans to sprout extra limbs, c) live longe and prosper, or d) none of the above.
And it's going to take a long time before they can try these experiments on humans.
--Pat
Quad toastage!
first thing in the morning.
--Pat
For what it's worth, the Internet Archive has at least at one point in its history collected Usenet posts. This isn't in the Wayback machine, though.
http://www.archive.org/about/faqs.php#The_Wayback_ Machine
--Pat / zippy@cs.brandeis.edu / blog / pics
Let's say Google has won the on-line ad business. How will they get more revenue? They can: 1) get more companies to advertise, 2) charge more to existing customers, or 3) introduce new services that make money and have high and growing earnings.
Bringing out the Chief Food Officer answers none of these questions. It is this that the Economist takes issue with. They expect a company to tell its stockholders something about the company intends to continue its growth. "We have great food" is not an answer.
--Pat / zippy@cs.brandeis.edu
Google's extraordinarily high P/E ratio means that, at its current price, it must grow earnings astronomically (grow, not just maintain, and certainly not lose money) over the long-term in order to earn a good rate of return at the current price, relative to other stocks.
--Pat / zippy@cs.brandeis.edu
A user pulling up this page on the Archive would generate a request to the original server, which would then show up in the original server's log.
--Pat / zippy@cs.brandeis.edu
If it turns out that they were not complying with their own policy (I do not know if this is the case) it could weaken their argument that they should be exempt from being sued for copyright infringement.
--Pat / zippy@cs.brandeis.edu
And if you could demonstrate this novel application of mud bricks, I'm sure we'd all be impressed. Here's a helpful link.
--Pat / zippy@cs.brandeis.edu
Imagine someone invents a new Go algorithm that, when implemented, beats Go masters. I think any CS researcher or Go player would say is a major breakthrough.
Now say this algorithm is a pattern matcher, a static evaluation function. Given the current board, the algorithm chooses the next move.
Following your reasoning, this would be a pattern matcher, and so could not be novel. Is this correct?
--Pat / zippy@cs.brandeis.edu
As other technology people have said, patents are a bit like Cobol -- a language of their own -- but the courts read them differently than a normal person, and give particular weight to the preferred implementation parts and much less to the "and everything else, including the kitchen sink" wording that surrounds this.
--Pat / zippy@cs.brandeis.edu
--Pat / zippy@cs.brandeis.edu
--Pat / zippy@cs.brandeis.edu
There may have been work on statistical detection of anomalies without prior measures of what was anomalous, but even if this existed (and I never heard of work in this area), there's no guarantee that an algorithm that detects credit card fraud will be good at recommending products as things one would want to purchase.
If I were a patent examiner, or even an expert witness, I'd say that taking a CC fraud algorithm and getting it to make good purchase recommendations in real time on a shopping site would be a novel application.
--Pat / zippy@cs.brandeis.edu