Google can't even hide behind the mantra of not being able to micromanage automatic indexing, because the ENTIRE WEB is copyrighted in some form or another.
IANAL, but if I'm not mistaken US copyright law has a specific exemption for caching, which allows search engines to to operate legally.
The Mozilla Firefox web browser has increased its total global usage to 8.5% according to Onestat.com, a leading web analytics company. While Microsoft's Internet Explorer still dominates the global market at 87%, it has lost 1.6% of it's users since November of last year.
I admit I'm a bit clueless here, but AFAIK climate modelling deals with overall changes at a high level e.g. "It's going to be colder in the north atlantic by about 2 degrees on average in a few decades" and the like.
As opposed to weather prediction which says "It will rain in this spot on this day"
Just because we can't predict the 'noise' in the short term doesn't mean we can't determine overall changes long term.
It's named eigenfaces because the underlying compression technique uses eigenvalues/vectors.
It's called PCA (principle components analysis) and works by finding an underlying low-dimensional representation that minimises the reconstruction error when you put everything back together.
So you find a new set of basis vectors to represent the data with, where your new set is smaller than your old set.
It turns out that the best basis used for the reconstruction is formed from the 'biggest' eigenvectors (those with the largest eigenvalues) of the covariance matrix of the data.
The simple answer is that there are far more users than there were testers. When more people look at a system, it's increasingly likely that someone will 'stumble' onto a bug.
Also, it's easier to find an exploit in an established system. Now we have lots of hardware and available WLAN access points to play about with. It's quite difficult to 'hack' a specification...
Depends how you define intelligent, I guess (and that's probably the toughest problem with AI).
The turing test only works in terms of 'weak' AI - that is that actions determine intelligence, and internal state doesn't matter.
Strong AI on the other hand, says that to be intelligent, you must not only act intelligent, but also be intelligent internally.
On the other hand, the self-preservation requirement doesn't really strike me as a facet of intelligence. A suicidal person might still have the intelligence to write a thoughtful note...
Actually, enigma was the german coding machine that Turing and others were trying to figure out so that they could intercept encrypted german transmitions.
That's what I was thinking when I first started using the beta gmail - not much that opera doesn't do...
...except for one small difference. I now have those cool features (searching, labels, etc.) wherever I am in the world, regardless of the client machine.
I wish I could take opera with me. I can't, so gmail is a 'Good Thing' for me.
Isn't that weird, considering that MS is the only one with the source??
Microsoft isn't the only company with access to MS source code.
They have quite a few security 'partners' who have source code but are under full NDA and can't go public with any exploits found without an MS go-ahead.
I know quite a few recent CS graduates that have moved straight into jobs that would be considered software engineering positions.
I think that while most those CS graduates moving into SE jobs only found a small amount of their course very relevant, what is probably important is that each of them seems to tell me about a different part.
A friend moving into a DB engineering post found that understanding DB query optimisation on a low level helped him out.
Another friend found skills learned in a compiler-writing course came in useful, and another found some AI stuff to be handy.
Maybe the point here is that a CS degree has to teach fundamentals, but also usually aims to give a grounding in a wide range of stuff that caters to as many students as possible.
I'm just about to graduate with an AI degree, and in my course, even basic neural nets, or simpler concepts like generalization/overfitting, dimension reduction or the bias/variance dilemma are usually explained in terms that require use of:
There are online methods using both the techniques you mention. The theory is usually a little more involved, so you're not likely to get a good tutorial from page 1 of google results.
I hear all too often from people in the field of machine learning who get their favourite solution (SVMs and NNs are the most common) and then they go hunting for a problem.
It might not be exactly the best technique, but if at the time it was the easiest to understand and use, and gave really good results, then the right decision was made.
Is that the difference between theory and practice right there?
Well done! You have successfully figured out that this system is not designed with you in mind!
In fact, I think you'll probably find yourself much happier with windows. You are running windows, aren't you? A little bird told me that it supports your scanner, TV card and your camera!
However, for some of us Linux (or Solaris, etc) is a better option:
Do I own your camera? no.
Do I own your TV card? no.
Do I have your digital camera? no.
Do I have 200 bucks handy to buy a second copy of windows for my second pc? no.
Do I really want to agree to licensing terms that allow MS to pay me 'friendly visits' every no and again? no.
etc.
MS is one of the worst examples to cite in this case.
For at least a year more than half of the MS product line was called.NET something.
Maybe on the Java branding issue Sun could learn from MS, but IANAMG ( I am not a marketing guy;-) )
Re:Cone of Silence? More like cone of annoyance.
on
Directed Sound
·
· Score: 1
I think that the problem with spam isn't just how irritating it is, but also how difficult it is to get rid of. We're bombarded by it every day.
This technology (in the short term at least)requires a physical presence of the 'spamming' hardware to point the sound at you. It could easily be spotted and sorted out on an individual case.
Slashdot Mirror
We also have a page of resources that includes links to two ASP.NET controls.
As someone who's using gmail right now in Opera 8.0 (with no apparent bugs), I have to disagree...
If you release how you do the ranking function, suddenly every web scrambler in the world screws up the rank
PageRank ranking function:
http://citeseer.ist.psu.edu/page98pagerank.html
Details on the implementation of PageRank:
http://citeseer.ist.psu.edu/brin98anatomy.html
Both of these papers are extremely outdated, but the PageRank ranking function is by no means a secret.
IANAL, but if I'm not mistaken US copyright law has a specific exemption for caching, which allows search engines to to operate legally.
In this case, yes...
http://www.overclockersclub.com/?read=0970358
I admit I'm a bit clueless here, but AFAIK climate modelling deals with overall changes at a high level e.g. "It's going to be colder in the north atlantic by about 2 degrees on average in a few decades" and the like.
As opposed to weather prediction which says "It will rain in this spot on this day"
Just because we can't predict the 'noise' in the short term doesn't mean we can't determine overall changes long term.
It's named eigenfaces because the underlying compression technique uses eigenvalues/vectors.
It's called PCA (principle components analysis) and works by finding an underlying low-dimensional representation that minimises the reconstruction error when you put everything back together.
So you find a new set of basis vectors to represent the data with, where your new set is smaller than your old set.
It turns out that the best basis used for the reconstruction is formed from the 'biggest' eigenvectors (those with the largest eigenvalues) of the covariance matrix of the data.
>> I'm sorry, I don't have a 'naive'. Is it some sort of American device?
> naive adj. 1. Lacking worldly experience and understanding
That has to be the most ironic thing I've read in ages.
btw check out the meaning of "your", vladmir...
Also, it's easier to find an exploit in an established system. Now we have lots of hardware and available WLAN access points to play about with. It's quite difficult to 'hack' a specification...
The turing test only works in terms of 'weak' AI - that is that actions determine intelligence, and internal state doesn't matter.
Strong AI on the other hand, says that to be intelligent, you must not only act intelligent, but also be intelligent internally.
On the other hand, the self-preservation requirement doesn't really strike me as a facet of intelligence. A suicidal person might still have the intelligence to write a thoughtful note...
This page has a description of the machine.
Turing didn't invent the machine. The germans did.
Equivalence to a turing machine is used in lots of CS proofs even today.
The turing test is also still considered one of the fundamental challenges of 'weak' Artificial Intelligence.
...except for one small difference. I now have those cool features (searching, labels, etc.) wherever I am in the world, regardless of the client machine.
I wish I could take opera with me. I can't, so gmail is a 'Good Thing' for me.
Engineering something to be 'perfect' in some cases can be pretty easy.
Getting something with believable 'human' imperfections often turns out to be pretty hard...
Microsoft isn't the only company with access to MS source code.
They have quite a few security 'partners' who have source code but are under full NDA and can't go public with any exploits found without an MS go-ahead.
I know quite a few recent CS graduates that have moved straight into jobs that would be considered software engineering positions.
I think that while most those CS graduates moving into SE jobs only found a small amount of their course very relevant, what is probably important is that each of them seems to tell me about a different part.
A friend moving into a DB engineering post found that understanding DB query optimisation on a low level helped him out.
Another friend found skills learned in a compiler-writing course came in useful, and another found some AI stuff to be handy.
Maybe the point here is that a CS degree has to teach fundamentals, but also usually aims to give a grounding in a wide range of stuff that caters to as many students as possible.
I'm just about to graduate with an AI degree, and in my course, even basic neural nets, or simpler concepts like generalization/overfitting, dimension reduction or the bias/variance dilemma are usually explained in terms that require use of:
There are online methods using both the techniques you mention. The theory is usually a little more involved, so you're not likely to get a good tutorial from page 1 of google results.
Try MIT's open courseware (Machine Learning course) for some better explanations of this stuff, if you can handle the maths, ughhh.
I hear all too often from people in the field of machine learning who get their favourite solution (SVMs and NNs are the most common) and then they go hunting for a problem.
It might not be exactly the best technique, but if at the time it was the easiest to understand and use, and gave really good results, then the right decision was made.
Is that the difference between theory and practice right there?
In fact, I think you'll probably find yourself much happier with windows. You are running windows, aren't you? A little bird told me that it supports your scanner, TV card and your camera!
However, for some of us Linux (or Solaris, etc) is a better option:
Do I own your camera? no.
Do I own your TV card? no.
Do I have your digital camera? no.
Do I have 200 bucks handy to buy a second copy of windows for my second pc? no.
Do I really want to agree to licensing terms that allow MS to pay me 'friendly visits' every no and again? no.
etc.
For at least a year more than half of the MS product line was called .NET something.
Maybe on the Java branding issue Sun could learn from MS, but IANAMG ( I am not a marketing guy ;-) )
I think that the problem with spam isn't just how irritating it is, but also how difficult it is to get rid of. We're bombarded by it every day. This technology (in the short term at least)requires a physical presence of the 'spamming' hardware to point the sound at you. It could easily be spotted and sorted out on an individual case.