Along a similar vein... Here's something else Shannon (I believe it was him) introduced in his original 1948 paper. It's how one calculates the minimum description length given only the probability distribution of the RV (independent of the channel it's being sent on.)
The Entropy of a random variable is the minimum number of bits needed to encode all information about said variable, and it can be found by calculating
-[Sum over all i] Pi*log(1/Pi)
where Pi is the probability that the variable takes on the ith value. E.G. if an arbitary letter in a text message has a 1/4th probability of being an "a", then P1 = 1/4.
One of the main problems with this value, is that it causes the average codeword length of an ideal coding scheme to often not be an integer, which is why optimal coding schemes only approach the ideal (the entropy), as has been stated earlier. (That is, of course, unless the probabilities are inverse powers of the # of letters in your alphabet - for binary codes this means that probabilities are 1/2, 1/4, 1/8th, etc.)
It is known, however, how far off a real code gets from the entropy. The average code length of an optimal code always lies within one bit of the entropy.
That is, Entropy(X)
This is using a scheme that encodes each character individually, however, and can be made more accurate coding in block lenghts (this is the idea behind arithmetic coding, which somebody mentioned eariler).
The main idea behind Arithmetic Coding is this: the probablility difference between a long string of likely characters and a long string of unlikely characters is greater than the probability difference of just the lone likely character and the lone unlikely character. That is, instead of encoding each letter of your message alphabet individually, encode chunks of your alphabet in one shot.
This has been shown to get better compression, that is, you're able to divide the previous inequality by the encoded block length n, and get:
Entropy(X)/n
So you can probably guess that increasing the block length (n) gets you closer and closer to the entropy.
Keep in mind all of my post assumed an error-free channel.
A foreign company is getting government help so it can produce at costs better than their comparative advantages. This distorts a market in which producers produced in line with their respective comparative advantages.
In order to bring the price of the foreign good back in line with the comparative advantange of the foreign producer, the importing government imposes a countervailing duty. When implemented correctly a countervailing duty does two things: It restores the market to it's ideal state, and the foreign country's citizens pay some of the local government taxes.
Think about it. Korean taxpayers pay their government money that goes to a korean corporation. That corporation pays export duties to the US (forcing them to compete in line with their comparative advantages). Those export duties pay for governmental costs in the US. End result - Koreans pay US taxes.
The Korean citizens should be complaining to their governmen.
I currently am a student worker in my University's ITS dept. To give you an idea about what kind of university, out of many many thousands, we're #31 (with whatever rating scheme is used). Frankly, only 2-3 people who work here in ITS are computer science majors. Most of the people are just savvy users that can learn a new app in a day instead of a year. This lends one to think that a changeover would be just a weeks worth or so of new-learning for ITS staff, but that's not true.
On the various computer classroom computers, there are at least 10 pieces of windows-only software that professors use to teach classes, and some professors (not the CS dept. interestingly enough) write their own apps for their classes.
I don't mean to speculate on whether a complete switch over would be a Good Thing (even though I think the end result would be worth it, at least in the sciences), just to show that the dependency on whatever system is in place is often rather strong.
Also, there is some handi-capped accessible software that is installed on some of the computers that we are required by law to provide, and it happens to be windows-only. (It's called ZoomText, and it's use is rather self-explanatory).
Liberties are somewhat better, if memory serves (don't have any statistics on me, though). Tolerance of varying ethnicities is much better, and gun-related crime is drastically lower than in the US, even though 7/10ths of their households own firearms
the mental de-sensatizing helps a lot with longevity on your first time. (I remember a friend not knowing her new 25yr old boyfriend was a virgin when she popped his cherry because he lasted a while. He of course had loads of porn on his machine).
(aside from the poor impersonation of an element for a name)
If MS goes ahead with Palladium, I'll be keeping my eye out for the first virus to fool the OS into rejecting every app, regardless of signature. Perfect DOS attack. Can't do anything but reinstall from the installation media, if your DRM bios will let you that is...
This is the main problem with this idea, the owner of the computer no longer has root access to their own machine with this DRM (emphasis on management) software. Ultimately, regardless of the level of knowledge, you cannot change certain settings/fix certain problems because of the micro-micro-kernel that's playing in the backgroung with read and write protection.
The main reason the most popular part of the metaverse was popular was because of something sound and text couldn't provide. The Black Sun was a sucess because of the insanely detailed coding of facial expressions. See we are a species that naturally communicates on both a concious (specific string of words) and unconcious ('body language') level. IM and email only communicate on a concious level. Subsequently IM and email discussions often seem hollow or thin unless you know the person writing them and can accurately guess at their probable facial expressions, etc.
(time to make this ontopic)
Until substantial efforts in facial detail are put in, there won't be much mass acceptance of these types of systems.
The following is slightly offtopic, so feel free to ignore it. This hints at the usefulness of emotes in MMOG's; it allows a certain level of body langauge for all those who have more friends online than "IRL". Similarly interesting is that many a-social MMOG players seem more comfortable communicating online. This could be because of a lack of sub-concious communication skills.. (of course this is all wild speculation)
You know those lines on solar panels? It could be that a single-layer solar lense loses energy from converting wavelengths orthogonal to those lines in to energy.
It could be that this new technology is doubly layered, letting through all the wavelengths orthogonal to the top layer down to the bottom layer; each layer turning in to energy the wavelength orientation that's most efficient.
We agree that an invention benefits the society most when it is in the public domain -- anyone can use and benefit from it. However, we agree that in order for individuals in a capitalist society to have incentive to invent, they must be able to capitalize on their invention. Hence the "limited times" in the constitution for an inventor to profit from his mind.
For the greatest benefit of the society at large, we want the "limited times" to grant just enough incentive to the inventors to invent at high rates (my idea would be to have the copyright term be a function of the average amount of time taken to invent something). One can assume rather assuredly that the length of a copyright should most certainly not be as long as a generation, otherwise entire generations would never know the free access to the idea.
As is, the terms are something like life+50 years. Life plus 50 years?? look at it like this: people who were born after Mickey Mouse was copyrighted and have died since then (there's a lot of them, 1920's-) never benefitted from any of Disney's creations in the public domain. Does this benefit society as a whole, or the corporate monopolies who own the copyright?
Your comment is humorous in intention, but it sheds light on a problem. "I only get exclusive rights for 90 years?!? I don't think that's enough. OTOH, 15 years from now, the term will be increased to 110 years. So yeah, I guess doing this work is worth it, after all."
for a nice, stable economy, one wants nice stable copyright laws so inventors don't have things to worry about things like that at all. They should be worrying about one thing only, their invention. As is, companies that want to maintain their monopoly are paying to have the copyright terms extended, throwing monkey-wrenches in to the planning of anyone other than those who are controlling the changes (read: not legislators).
This benefits the companies, sure, but these actions shed light on a key principle when talking about length of a copyright.
Re:[OT] Oh my gosh! A female!
on
ALICE vs. ALICE
·
· Score: 1
What made the original post so offensive was that the poster appeared serious in his belief that females are intellectually inferior to males.
this is what the original post said:
When talking to a woman they can generally fake it and pretend to be smart, but then when they talk to each other it degenerates into talk about fashion, shoes, who's the bigger slut etc..
Ever notice this?
now maybe it's just me, but as I read that it registered as too outrageous to be true (I mean, c'mon, no sensible/sane person actually things all females are degenerates). What the comment did is remind me of the few females I know who do fit the description (they may even be movie characters, adding to the lack-of-seriousness of the original post), and I laughed at the memory of them.
The post made a generalization, and all generalizations are false ('cept this one and all that). Subsequently the post's material is not true in all cases, and is probably meant as either humor, satire, or sarcasm (meant to degrade the idea that all women are degenerates). Pick one. None of them are serious insults directed at women.
MS doesn't have niether competition nor federal mandates preventing computers from being restricted.
that comes out to: microsoft NOT-has NOT(competition AND mandates preventing computers from being restricted).
so they don't have neither.
that means they have at least one of the two, i suppose we get to guess which
try: MS has neither competition nor federal mandates preventing computers from being restricted.
Slashdot Mirror
We American have never fought a war with UK so that why I don't know where it's at.
What? Me fail history and geography?
Nope, it's quite possible. We fought a war with them.
bah. damn HTML parsing my inequalities.
Along a similar vein... Here's something else Shannon (I believe it was him) introduced in his original 1948 paper. It's how one calculates the minimum description length given only the probability distribution of the RV (independent of the channel it's being sent on.)
The Entropy of a random variable is the minimum number of bits needed to encode all information about said variable, and it can be found by calculating
-[Sum over all i] Pi*log(1/Pi)
where Pi is the probability that the variable takes on the ith value. E.G. if an arbitary letter in a text message has a 1/4th probability of being an "a", then P1 = 1/4.
One of the main problems with this value, is that it causes the average codeword length of an ideal coding scheme to often not be an integer, which is why optimal coding schemes only approach the ideal (the entropy), as has been stated earlier. (That is, of course, unless the probabilities are inverse powers of the # of letters in your alphabet - for binary codes this means that probabilities are 1/2, 1/4, 1/8th, etc.)
It is known, however, how far off a real code gets from the entropy. The average code length of an optimal code always lies within one bit of the entropy.
That is, Entropy(X)
This is using a scheme that encodes each character individually, however, and can be made more accurate coding in block lenghts (this is the idea behind arithmetic coding, which somebody mentioned eariler).
The main idea behind Arithmetic Coding is this: the probablility difference between a long string of likely characters and a long string of unlikely characters is greater than the probability difference of just the lone likely character and the lone unlikely character. That is, instead of encoding each letter of your message alphabet individually, encode chunks of your alphabet in one shot.
This has been shown to get better compression, that is, you're able to divide the previous inequality by the encoded block length n, and get:
Entropy(X)/n
So you can probably guess that increasing the block length (n) gets you closer and closer to the entropy.
Keep in mind all of my post assumed an error-free channel.
We've reported on this before.
Perhaps it's just me, but doesn't that seem like a big ole "We're posting something for the second time".
(it may well be that this time there's an update, but the blurb doesn't mention it.)
Obligatory IMDB link
and it's time to start over new. And, hopefully get it right this time!
I see you haven't used many microsoft products before.
Anyone else read "India Plans Moon Mission" as "India is planning to drop their pants and moon everyone"?
Then again it is 3am at a LAN party and my brain is running on caffeine
(define (.sig) (cons 'my (list 'other 'car 'is 'a 'cdr))
.sig (cons 'my '(other car is a cdr))
writing the following will have the same effect, and be shorter (hooray for syntactic sugar!)
(define
A foreign company is getting government help so it can produce at costs better than their comparative advantages. This distorts a market in which producers produced in line with their respective comparative advantages.
In order to bring the price of the foreign good back in line with the comparative advantange of the foreign producer, the importing government imposes a countervailing duty. When implemented correctly a countervailing duty does two things: It restores the market to it's ideal state, and the foreign country's citizens pay some of the local government taxes.
Think about it. Korean taxpayers pay their government money that goes to a korean corporation. That corporation pays export duties to the US (forcing them to compete in line with their comparative advantages). Those export duties pay for governmental costs in the US. End result - Koreans pay US taxes.
The Korean citizens should be complaining to their governmen.
way to go, [name censored], the all nighters paid off!
I currently am a student worker in my University's ITS dept. To give you an idea about what kind of university, out of many many thousands, we're #31 (with whatever rating scheme is used). Frankly, only 2-3 people who work here in ITS are computer science majors. Most of the people are just savvy users that can learn a new app in a day instead of a year. This lends one to think that a changeover would be just a weeks worth or so of new-learning for ITS staff, but that's not true.
On the various computer classroom computers, there are at least 10 pieces of windows-only software that professors use to teach classes, and some professors (not the CS dept. interestingly enough) write their own apps for their classes.
I don't mean to speculate on whether a complete switch over would be a Good Thing (even though I think the end result would be worth it, at least in the sciences), just to show that the dependency on whatever system is in place is often rather strong.
Also, there is some handi-capped accessible software that is installed on some of the computers that we are required by law to provide, and it happens to be windows-only. (It's called ZoomText, and it's use is rather self-explanatory).
Symantec. Ghost. Console.
If anything goes wrong, hose the entire room.
Liberties are somewhat better, if memory serves (don't have any statistics on me, though). Tolerance of varying ethnicities is much better, and gun-related crime is drastically lower than in the US, even though 7/10ths of their households own firearms
"ding-dong the witch is dead, the wicked witch! the witch is dead!" (define dead declining)
...posters will be able to flame RIAA, hammer home the blandness of mainstream music offerings and blame Canada in a single post!
That's nice, but how am I going to say that in soviet Russia, posting flaming canadians hammer home the lameness of YOU!
Oi, moderators, this is not meant to insult the parent. Go easy with those flamebait mods.
the mental de-sensatizing helps a lot with longevity on your first time. (I remember a friend not knowing her new 25yr old boyfriend was a virgin when she popped his cherry because he lasted a while. He of course had loads of porn on his machine).
(aside from the poor impersonation of an element for a name)
If MS goes ahead with Palladium, I'll be keeping my eye out for the first virus to fool the OS into rejecting every app, regardless of signature. Perfect DOS attack. Can't do anything but reinstall from the installation media, if your DRM bios will let you that is...
This is the main problem with this idea, the owner of the computer no longer has root access to their own machine with this DRM (emphasis on management) software. Ultimately, regardless of the level of knowledge, you cannot change certain settings/fix certain problems because of the micro-micro-kernel that's playing in the backgroung with read and write protection.
The main reason the most popular part of the metaverse was popular was because of something sound and text couldn't provide.
The Black Sun was a sucess because of the insanely detailed coding of facial expressions. See we are a species that naturally communicates on both a concious (specific string of words) and unconcious ('body language') level. IM and email only communicate on a concious level. Subsequently IM and email discussions often seem hollow or thin unless you know the person writing them and can accurately guess at their probable facial expressions, etc.
(time to make this ontopic)
Until substantial efforts in facial detail are put in, there won't be much mass acceptance of these types of systems.
The following is slightly offtopic, so feel free to ignore it.
This hints at the usefulness of emotes in MMOG's; it allows a certain level of body langauge for all those who have more friends online than "IRL". Similarly interesting is that many a-social MMOG players seem more comfortable communicating online. This could be because of a lack of sub-concious communication skills.. (of course this is all wild speculation)
You know those lines on solar panels? It could be that a single-layer solar lense loses energy from converting wavelengths orthogonal to those lines in to energy.
It could be that this new technology is doubly layered, letting through all the wavelengths orthogonal to the top layer down to the bottom layer; each layer turning in to energy the wavelength orientation that's most efficient.
We agree that an invention benefits the society most when it is in the public domain -- anyone can use and benefit from it. However, we agree that in order for individuals in a capitalist society to have incentive to invent, they must be able to capitalize on their invention. Hence the "limited times" in the constitution for an inventor to profit from his mind.
For the greatest benefit of the society at large, we want the "limited times" to grant just enough incentive to the inventors to invent at high rates (my idea would be to have the copyright term be a function of the average amount of time taken to invent something). One can assume rather assuredly that the length of a copyright should most certainly not be as long as a generation, otherwise entire generations would never know the free access to the idea.
As is, the terms are something like life+50 years. Life plus 50 years?? look at it like this: people who were born after Mickey Mouse was copyrighted and have died since then (there's a lot of them, 1920's-) never benefitted from any of Disney's creations in the public domain. Does this benefit society as a whole, or the corporate monopolies who own the copyright?
Your comment is humorous in intention, but it sheds light on a problem.
"I only get exclusive rights for 90 years?!? I don't think that's enough. OTOH, 15 years from now, the term will be increased to 110 years. So yeah, I guess doing this work is worth it, after all."
for a nice, stable economy, one wants nice stable copyright laws so inventors don't have things to worry about things like that at all. They should be worrying about one thing only, their invention. As is, companies that want to maintain their monopoly are paying to have the copyright terms extended, throwing monkey-wrenches in to the planning of anyone other than those who are controlling the changes (read: not legislators).
This benefits the companies, sure, but these actions shed light on a key principle when talking about length of a copyright.
What made the original post so offensive was that the poster appeared serious in his belief that females are intellectually inferior to males.
this is what the original post said:
When talking to a woman they can generally fake it and pretend to be smart, but then when they talk to each other it degenerates into talk about fashion, shoes, who's the bigger slut etc.. Ever notice this?
now maybe it's just me, but as I read that it registered as too outrageous to be true (I mean, c'mon, no sensible/sane person actually things all females are degenerates). What the comment did is remind me of the few females I know who do fit the description (they may even be movie characters, adding to the lack-of-seriousness of the original post), and I laughed at the memory of them.
The post made a generalization, and all generalizations are false ('cept this one and all that). Subsequently the post's material is not true in all cases, and is probably meant as either humor, satire, or sarcasm (meant to degrade the idea that all women are degenerates). Pick one. None of them are serious insults directed at women.
ATM
MS doesn't have niether competition nor federal mandates preventing computers from being restricted.
that comes out to: microsoft NOT-has NOT(competition AND mandates preventing computers from being restricted).
so they don't have neither.
that means they have at least one of the two, i suppose we get to guess which
try: MS has neither competition nor federal mandates preventing computers from being restricted.
i was refuting the first argument with the second