Quick question... why have you designed your company's intranet to use an element (canvas) from a standard (HTML5) that isn't agreed upon, and probably never will be ?
So that in 10 years, when it's already out of date, and all browsers have defined their own incompatible versions of freeform rendering, you can blame Microsoft for sucking ?
So what exactly is the point in installing a faster browser (as a plugin) into a browser you think is insecurely sandboxed.
I know what will happen, just counting the days... the first exploit for the Chrome I-Frame will come out, Microsoft will deny liability because the user is running a plugin they said was insecure, and Google will blame Microsoft for not habing a properly sandboxed environment for plugins to tun in.
End result, Google gets their users surfing preference data for "not evil" purposes, Microsoft takes the blame, user still gets shafted, and no support from either side.
So something made by the world's biggest ad seller, that will force your homepage to google.com, and silently report back every URL you are visiting to google.com is "free" ???
It's Google Toolbar in a bigger window, don't kid yourself.
The (optimal) AI model is unique in the sense that it has no parameters which could be adjusted to the actual environment in which it is used
Yes, this makes perfect sense, however, the algorithms thus far presented have been subject to many *human defined* parameters to optimize the compression. It's hardly unsupervised AI "learning".
The kind of thing described above is more akin to Kohonen Networks, which when trained on specific inputs (with no pre-specified outputs), form a map or representation of the data (which may or may not turn out to be a more compressed version).
Then it follows that the trained network when presented with similar data will form similar maps, and possibly, given a specific map can reform the original data (in a lossy way I'd assume).
When a genetic algoritm can (unsupervised) form a representation of the data, AND it happens to also be the most compact representation, then that will be AI.
First thing Hutter needs to do is refine his summary table, and remove the column named "Compression Options". Any algorithm that has "compression options" clearly does not meet the criteria of "a model with no parameters".
No, the difference being the symptoms don't affect you, but as you are still a carrier, you manage to go to work without loss of pay, and spread it to all the poor bastards who weren't vaccinated, costing them time and wages.
Vaccinations really only work for selfish motives, unless everyone, and I mean everyone, has been vaccinated against the same disease.
Look I'm getting really tired of arguing with you, as you seem to have some special axe to grind on this.
I'd suggest you look at the title of the page to start with...
"50'000 Prize for Compressing Human Knowledge"
Then the opening paragraph...
"Being able to compress well is closely related to intelligence as explained below. While intelligence is a slippery concept, file sizes are hard numbers. Wikipedia is an extensive snapshot of Human Knowledge. If you can compress the first 100MB of Wikipedia better than your predecessors, your (de)compressor likely has to be smart(er). The intention of this prize is to encourage development of intelligent compressors/programs."
Now reading futher, there is the underlying assumption that to simulate AI, we must be able to find the simplest representation of the data, ergo the parallel with compression.
But I'd argue, not being a PhD, just someone with a different opinion, that the way information is stored in the brain is not necessarily in the most compressed or compact form. There is plenty to suggest that information storage is redundant, not stored in one specific place, and not even in a form that we would consider structured... most certainly it is lossy in certain aspects, and lossless in other aspects.
Being able to compress knowledge into it's simplest repesentation does not equal AI, simply because we have no certainty of how *real* intelligence is represented across those billions of neurons and trillions of interconnected synapses.
English corpus fine... but the point is that the wiki data text set *isn't* a standard English Text, it's a form of specialised wikipedia markup language (think XML) with abbrevations, wiki code numbers and dates.
A lot of the so called "optimizations" have been achieved by identifying the structure of that *specific* markup to make extra gains in compression.
It's like making a compressor that works specifically well on Windows PE exe file structure, and expecting it to do the same thing on a jpeg or plain text file in the English Language.
Nothwithstanding that, I guess the think that put me off was the whole "compression == AI" angle that Hutter tried to put on things, when the results gained have been patently NON AI inspired... unless that russian guy is actually a robot ?
The Hutter Prize is a nonsense... there's only been one russian competing in it since 2006, and he's won the princely sum of about 6,700 Euros. Bit of a far cry from a million bucks.
Also, the prize structure is flawed, in that it penalizes the very people who are achieving the best work. The first person to achieve an improvement in data compression gets a chunk of the prize money. Then someone else who comes in later and manages to compress it a bit further (arguably a harder task to make any gains), only gets a chunk of the *remaining* prize money.
And as it is patently obvious that the compression algorithms are NOT general purpose, but specifically tuned / optimized to the data set in question (a 100MB chunk of wiki data), it is probably going to be useless for any other data set. At least the Netflix development was performed against an oracle, and will probably improve their systems, if not by a full 10%, but somewhere pretty damn close.
In Soviet Russia, Alexander Ratushnyak compresses you ! (Sorry, just couldn't resist)
According to Sony Ericsson, this will allow you to 'play your music and answer phone calls just by inserting the buds into your ear or taking them out'
So you can listen to music with both earphones, but only listen to a telephone conversation piped through one earphone as you had to remove the other to pause the music ? And what happens if the person calling you puts you on one of those annoying musical hold things ?
It's bad enough they sneak in rootkits, now they're sneaking in paradoxes.
Judging by the typical clickthru rates for banner ads (some fraction of 1%), most people will never click an ad and it would make no difference if the ads were blocked or not, and the minority who do already know what they are looking for and will happily tolerate the ads anyway if it helps them find a cheaper flight, or cheaper viagra, or whatever.
So saying that people who make a concious effort to block ads will kill the revenue is nonsense... they would never have clicked on the ad anyway. And those who don't block ads are exactly the people the advertisers want to target.
You know, I decided to pay for Sky Cable TV here in the Philippines for the exact same reason... no commercials.
Now the bastards started sneaking them in as "sponsored by", "supported by", and "in association with" links, before, during and after every damn show and intermission. And it's not just *one* sponsor like CNN does, it's at least 5 for popular primetime shows.
I much like the British BBC for their high quality documentaries. The making of these involves lots of research, putting people on the ground in conflict zones, undercover operations, etc. I would not mind paying a small amount each time I watch one of these documentaries, just to support putting them together. Surely I'm not alone there.
If you live in the UK and own a television receiver, you already *do* pay a small amount (about 30 pence a day in licence fees), whether you watch it or not.
Slashdot Mirror
Quick question ... why have you designed your company's intranet to use an element (canvas) from a standard (HTML5) that isn't agreed upon, and probably never will be ?
So that in 10 years, when it's already out of date, and all browsers have defined their own incompatible versions of freeform rendering, you can blame Microsoft for sucking ?
So what exactly is the point in installing a faster browser (as a plugin) into a browser you think is insecurely sandboxed.
I know what will happen, just counting the days ... the first exploit for the Chrome I-Frame will come out, Microsoft will deny liability because the user is running a plugin they said was insecure, and Google will blame Microsoft for not habing a properly sandboxed environment for plugins to tun in.
End result, Google gets their users surfing preference data for "not evil" purposes, Microsoft takes the blame, user still gets shafted, and no support from either side.
Something free
So something made by the world's biggest ad seller, that will force your homepage to google.com, and silently report back every URL you are visiting to google.com is "free" ???
It's Google Toolbar in a bigger window, don't kid yourself.
Okay, one more shot before I sleep ...
The (optimal) AI model is unique in the sense that it has no parameters which could be adjusted to the actual environment in which it is used
Yes, this makes perfect sense, however, the algorithms thus far presented have been subject to many *human defined* parameters to optimize the compression. It's hardly unsupervised AI "learning".
The kind of thing described above is more akin to Kohonen Networks, which when trained on specific inputs (with no pre-specified outputs), form a map or representation of the data (which may or may not turn out to be a more compressed version).
Then it follows that the trained network when presented with similar data will form similar maps, and possibly, given a specific map can reform the original data (in a lossy way I'd assume).
When a genetic algoritm can (unsupervised) form a representation of the data, AND it happens to also be the most compact representation, then that will be AI.
First thing Hutter needs to do is refine his summary table, and remove the column named "Compression Options". Any algorithm that has "compression options" clearly does not meet the criteria of "a model with no parameters".
No, the difference being the symptoms don't affect you, but as you are still a carrier, you manage to go to work without loss of pay, and spread it to all the poor bastards who weren't vaccinated, costing them time and wages.
Vaccinations really only work for selfish motives, unless everyone, and I mean everyone, has been vaccinated against the same disease.
Look I'm getting really tired of arguing with you, as you seem to have some special axe to grind on this.
I'd suggest you look at the title of the page to start with ...
"50'000 Prize for Compressing Human Knowledge"
Then the opening paragraph ...
"Being able to compress well is closely related to intelligence as explained below. While intelligence is a slippery concept, file sizes are hard numbers. Wikipedia is an extensive snapshot of Human Knowledge. If you can compress the first 100MB of Wikipedia better than your predecessors, your (de)compressor likely has to be smart(er). The intention of this prize is to encourage development of intelligent compressors/programs."
Now reading futher, there is the underlying assumption that to simulate AI, we must be able to find the simplest representation of the data, ergo the parallel with compression.
But I'd argue, not being a PhD, just someone with a different opinion, that the way information is stored in the brain is not necessarily in the most compressed or compact form. There is plenty to suggest that information storage is redundant, not stored in one specific place, and not even in a form that we would consider structured ... most certainly it is lossy in certain aspects, and lossless in other aspects.
Being able to compress knowledge into it's simplest repesentation does not equal AI, simply because we have no certainty of how *real* intelligence is represented across those billions of neurons and trillions of interconnected synapses.
Anyone who's been exposed to the ramblings of James Harris, even for a short time, will quickly discover a nasty taste in their mouths.
Hence the sad emoticon.
English corpus fine ... but the point is that the wiki data text set *isn't* a standard English Text, it's a form of specialised wikipedia markup language (think XML) with abbrevations, wiki code numbers and dates.
A lot of the so called "optimizations" have been achieved by identifying the structure of that *specific* markup to make extra gains in compression.
It's like making a compressor that works specifically well on Windows PE exe file structure, and expecting it to do the same thing on a jpeg or plain text file in the English Language.
Nothwithstanding that, I guess the think that put me off was the whole "compression == AI" angle that Hutter tried to put on things, when the results gained have been patently NON AI inspired ... unless that russian guy is actually a robot ?
The Hutter Prize is a nonsense ... there's only been one russian competing in it since 2006, and he's won the princely sum of about 6,700 Euros. Bit of a far cry from a million bucks.
Also, the prize structure is flawed, in that it penalizes the very people who are achieving the best work. The first person to achieve an improvement in data compression gets a chunk of the prize money. Then someone else who comes in later and manages to compress it a bit further (arguably a harder task to make any gains), only gets a chunk of the *remaining* prize money.
And as it is patently obvious that the compression algorithms are NOT general purpose, but specifically tuned / optimized to the data set in question (a 100MB chunk of wiki data), it is probably going to be useless for any other data set. At least the Netflix development was performed against an oracle, and will probably improve their systems, if not by a full 10%, but somewhere pretty damn close.
In Soviet Russia, Alexander Ratushnyak compresses you !
(Sorry, just couldn't resist)
Go back to trolling sci.math, James :-(
Usually they just find "evidence of WMDs" and then bomb your ass into oblivion.
whereas with the recipe, you simply have a list of instructions on how to get to the finished product
Remove plastic cover.
Microwave for 90 seconds.
Doesn't take *that* much culinary skill, does it ?
This conversation is giving me a headache. Anyone got an aspirin ?
The blinking was actually them transmitting in morse code "turn the bloody music down".
electric universe theory
Archimedes Plutonium, is that you and your theory, again ???
According to Sony Ericsson, this will allow you to 'play your music and answer phone calls just by inserting the buds into your ear or taking them out'
So you can listen to music with both earphones, but only listen to a telephone conversation piped through one earphone as you had to remove the other to pause the music ? And what happens if the person calling you puts you on one of those annoying musical hold things ?
It's bad enough they sneak in rootkits, now they're sneaking in paradoxes.
Nice try, you must be new here, but seeing as you posted AC, who knows ?
I think what you meant to say was "In Soviet Russia, COBOL writes you !"
7. It's another antiquated dinosaur ?
You should have used a wheelbarrow.
(Sorry, throwback to an 80's comic featuring Buster Gonad and his Unfeasibly Large Testicles).
The number of people in the world who can code in it ?
(Assuming of course that 2003 is a base-4 number).
SELECT * FROM TABLE INTO CRAPPY_MICROSOFT_CONTROL_STRUCTURE ?
This is what happens when you employ MCSEs.
(Filter error: Don't use so many caps. It's like YELLING ... no it's called SQL, you Slashdot f***wit coders).
Nope I don't agree there ...
Judging by the typical clickthru rates for banner ads (some fraction of 1%), most people will never click an ad and it would make no difference if the ads were blocked or not, and the minority who do already know what they are looking for and will happily tolerate the ads anyway if it helps them find a cheaper flight, or cheaper viagra, or whatever.
So saying that people who make a concious effort to block ads will kill the revenue is nonsense ... they would never have clicked on the ad anyway. And those who don't block ads are exactly the people the advertisers want to target.
You know, I decided to pay for Sky Cable TV here in the Philippines for the exact same reason ... no commercials.
Now the bastards started sneaking them in as "sponsored by", "supported by", and "in association with" links, before, during and after every damn show and intermission. And it's not just *one* sponsor like CNN does, it's at least 5 for popular primetime shows.
I much like the British BBC for their high quality documentaries. The making of these involves lots of research, putting people on the ground in conflict zones, undercover operations, etc. I would not mind paying a small amount each time I watch one of these documentaries, just to support putting them together. Surely I'm not alone there.
If you live in the UK and own a television receiver, you already *do* pay a small amount (about 30 pence a day in licence fees), whether you watch it or not.
are supposed to be well written, complete and verified
Well that leaves Murdoch out in the cold anyway.
You wouldn't have found out about Watergate or similar cases by Twitter
Yes, I can just imagine The Sun's version ... "Tricky Dicky bares all on Page 3".