Yeah, Its not a bell curve, you are right. However, the ppoint was not the particular shape a histogram would make from rolling die. The point was that the histogram would look the same if you measured every roll, or a random sample.
The thing to remember about statistics : They are very good at caclulating how an entire population will act.
The are very poor at determining how an individual will act.
Even things like "What books you will like, based on what other books you like" kinda stuff you see at amazon is pretty weak.
If 1/10 of of the people who like book A like book B, then that tells you not much about me, even if you know I like book A.
But if 1/10 of people who like book C like B, and D like B and E like B (etc) AND I like C,D, and E, then you have a 4/10 chance that I will like B. And that starts to be usefull.
Well, here is an example. I wrote a security camera program for a company once. They had a camera hooked up to a computer, and wanted it to save a GIF every time the picture changed (They had it pointed at a safe, they wanted to record every time someone accessed the safe)
We were taking pictures at 640x480 which is 288000 pixels. If I sampled as few as 50 pixels, and compared them to what the pixel was in the last frame, I got the same results as if I sampled 10,000 pixels, and the same as if I sampled every pixel.
It was very important that I pick my sample randomly, because obviously if I picked the first 50 pixels every time, and the motion was in the bottom of the frame, I would never see it.
Simmilarly, if 1/10 people in the US watches friends on Thursday nights, it doesn't matter if I pick 100 people, 1000 people, 10,000 people, or 100,000,000 people. 1/10 is 1/10.
Because the population is so large, and can be highly "pocketed" (people who live near eachother tend to watch the same stuff, because their friends tell them about it, or the local paper writes about it, or because someone local stars in it, or whatever) they need a VERY random, and very large (by statiscal measuring) sample.
Also, there is a wide variety of programming available, so if you only ask 100 people, you might get 1 person for each show so you ask 100000 people and then you notice that "Bob's Accordion World" only gets 1 viewer, and friends gets 10000.
In any case it works.
Here is a test you can do at home.
Take 2 dice. Roll them a few hundred times , and keep track of what you got each roll.
At the end, count how many ones you got, how many twos, how many threes etc.
You should end up with a nice bell curve, peaking around 6.
Now pick a random 50 rolls, and do the counts again. The bell curve should look almost identical, even though you only sampled a small percentage.
In this case, 50 is a large % of a few hundred rolls. But the graph would look the same if you rolled the dice 10,000,000 times. I just didn't think you would want to roll that many times:)
Uh. the DMCA specifically refers to circumventing access controls that prevent access to a copyrighted work. Since he doesn't get to acecss any "work" on your computer, it wouldn't be covered.
You could however set up an autoresponder to any ping or incoming email, that sends out a copyrighted work, and say the only way you could get in was by having a license, then he would be circumventing, and then you could maybe nail him:)
This is simmilar to the people putting haiku headers into email so you know it isnt spam. If a smapper copies the header, they are infringing the copyright
well, I totally disagree with this current lawsuit, but double jeopardy doesn't come into account.
Double Jeopordy refers to criminal action, all these anti trust suits are civil action (based on prior criminal acts tho)
Also, even if it was tried under criminal code, the suit is for a seperate market, which would presume seperate actions, which would be triable. (Just because you get off on a murder rap doesn't mean you can murder someone else for free).
Thirdly, Double Jeopordy is primarily a US thing (though much of europe has it as well. However, UK recently introduced legistlation weakening or canceling double jeopordy (Though I think it hasn't been signed yet, and may have been killed in process)
see : http://216.239.39.100/search?q=cache:JeAJHIKSZIQC: abcnews.go.com/sections/world/DailyNews/britain_ju stice020801.html+uk+double+jeopardy&hl=en&ie=UTF-8
The EU competition regulator is concerned that Microsoft might take over the market for mobile phone software, according to a senior adviser to the European Commission. "Mobile phones are a very important market in Europe. The Commission is worried that Microsoft might take advantage of the strength of Windows in order to dominate in mobile phones," he said
Uh, so they are suing because MS might do something? I forget what the legal term for this is, but I think this guy has a big stick up his ass.
Besides, Stinger has 0 market penetration right now.
Everyone is on just a big kick MSs ass because they are an easy target kick.
Actually, its not legal if the only intent is to damage the competitor, if its just a good marketing strategy, its not illegal.
Additionally, the fact that there is a subscription (XBox live) and accessories involved makes the case more complex. Companies can, will, and has argued successfully that dumping should be determined on the costs of everything the average buyer buys.
The dynamic you are talking about is NOT due to MS being a monopoly. It is because MS has more than one product line.
If Pepsi has a bad year, they still get to rely on taco bell, pizza hut, and doritos to make it through the year. Thats not a crime, its called diversification.
Note, you get to pay the taxes, even if you don't use cash. Using some made up cash would fall under barter rules, since at some point you either had to give them real cash, or something that someone else would pay real cash for.
Legal Tender is certainly a service provided for BY taxes, but is not in itself the source of tax revenue
Lion King was ripped from a Manga called Jungle Taitei. (Jungle Prince?)
Lead character : kimba. (To be fair, kimba/simba/timba and some others are african words for "cub" so its not like they ripped off a made up word)
However, some of the scenes were taken too, notable the cliff outcroping scene that is on all the lion king posters.
This was a huge controversy when lion king came out, Disney totally denied, said they had never even heard of the guys. Japanese guys said "We don't care, we take it as an hommage, and we ripped off bambi for our first picture anyway". Later, disney admited that several of the main animators were big fans, and had copies of the manga at their desks while making lion king.
This was also parodied in a great simpsons episode, with Lisa's mentor, (forget the first word) gums.
Gums was played by J.E. Jones, and after he died, they did this great thing where his ghost said a few words, then Darth vader said a few words, then mufasa said "Take care kimba.. I mean simba", and then JEJ said "This is CNN".
It was hilarious, and the number of people that would get the joke is really small.
commentary certainly allows quoting of the original. Just like a movie reviewer might show a still, or part of a scene.
But its not "quoting" if you put the whole thing. Can I quote your new book, in its entirety, and just add a few footnotes? Can I sell the result? Nope.
Note, I'm not saying this information shouldn't be out there. I'm just saying that a commentary or satire defense would quickly get squashed.
If he actually just quoted the document, or talked about it but didn't quote, then he would be much more in the clear.
Hrm, I think his defense on satire and commentary or non existant. Could I go to a movie, record the whole thing on my camcorder, and talk through it, saying what I thought, and then distribute the end product? Nope.
This also isnt satire, as he is making legitimate comments, not making jokes.
Now, if he re-wrote the document, but made everyhitng outrageous, that could be satire, but not the way it is now.
He is of course perfectly within his righs to publish commentary, but publishing the full text of the commented document I think is out of bounds (from a legal perspective)
There weren't tricks to get the descrepancy large. Sun wrote the original J2EE version. MS wrote the.net version. If Sun chose to write their code in an inefficient manner, (and you care about lines of code) then perhaps that tells you something about OTHER apps that sun writes?
This is going to sound long a snotty troll, but i don't mean it that way.
Perhaps the reason you are looking at replacing the system by the time upgrades come up, is that you didn't buy something powerfull enough to begin with. It could be that the sizing tools are taking that into account.
The pipe is [should be] metered because you are using up a % of it at any given time. There ARE things as bandwidth hogs (I am one of them). And If I am tying up 50% of the bandwidth for a day, I should pay for 50% of the bandwidth for the day.
Water is not finite, It will all get recycled eventually. However the amount of water that can be pumped at any given time, and the number of gallons per minute is finite. Simmilarly, the number of megabytes per second is finite.
Until you get to the backbone (and even then, as the recent DoS attack shows) bandwidth is most certainly finite.
Actually, a dynamically linked library DOES impose this restriction. The GPL was specifically modified to handle this. Peopel were taking GPL code, and just wrapping it into a library, so they could release their code under a different license. FSF got upset, and GPL was modified.
In the article, the authors refer to TCPIP. I think this is an excellent example. If the TCP-IP protocol were released under GPL, then anyone writing an application that used TCPIP in any fashion would have to be GPL as well. That means : No commercial web browsers. No commercial email programs. No commercial ANYTHING that does updates via the web. No operating systems (Whens the last time an OS got released without TCPIP support?)
Thats the point. The authors are not saying everyone shouldnt have access to the govt. funded code. They should. they are saying that govt funded code can be used in a non free app. This is correct.
As long as the original work created by the funding is free for everyone to use, the world is in a good spot.
The government invented velcro. Velcro is free for everyone to use. Should anything that uses velcro have to be free too?
(Question: Can you realease a spec under GPL, or just an implementation of the spec. If you can release a spec under the GPL, would the GPL just apply to extentions/modifcations of the spec, or anything that implemented the spec (or happened to comply with the spec)
Acutally, selective enforcement is grounds for having a law thrown out. (or if not the entire law, your particular case) This is because when the law is selectively encorced it is usually because of racial/ethnic/politicol profiling, someone with an axe to grind, or they couldn't pin anything else on you.
For example : Did you know that it is illegal to drive your car across state lines if you have a loan on it, without first informing the holder of the loan that you intend to do so?
Of course, everyone does this, and nobody gets busted. Until they (yes, the infamous they) want to get you to testify, or give information, or just toss you in jail because they can't prove you were the "bad guy"
Of course, this has no relevence in the google case, which is not a government entity, and can make whatever terms it wants with whomever it wants.
If the improvements are obvious, they are not patentable. If they are pateneted, they will be invalidated.
Its pretty rare to get an improvement patent, since you have to be improving the process is a radical way. Thats usually a different process all together.
Slashdot Mirror
It is most certainly a copyright violation, and not covered by fair use, since it is the entire article, with no commentary.
:)
That being said, thanks for posting it
Yeah, Its not a bell curve, you are right. However, the ppoint was not the particular shape a histogram would make from rolling die. The point was that the histogram would look the same if you measured every roll, or a random sample.
The thing to remember about statistics :
They are very good at caclulating how an entire population will act.
The are very poor at determining how an individual will act.
Even things like "What books you will like, based on what other books you like" kinda stuff you see at amazon is pretty weak.
If 1/10 of of the people who like book A like book B, then that tells you not much about me, even if you know I like book A.
But if 1/10 of people who like book C like B, and D like B and E like B (etc) AND I like C,D, and E, then you have a 4/10 chance that I will like B. And that starts to be usefull.
Well, here is an example. I wrote a security camera program for a company once. They had a camera hooked up to a computer, and wanted it to save a GIF every time the picture changed (They had it pointed at a safe, they wanted to record every time someone accessed the safe)
:)
We were taking pictures at 640x480 which is 288000 pixels. If I sampled as few as 50 pixels, and compared them to what the pixel was in the last frame, I got the same results as if I sampled 10,000 pixels, and the same as if I sampled every pixel.
It was very important that I pick my sample randomly, because obviously if I picked the first 50 pixels every time, and the motion was in the bottom of the frame, I would never see it.
Simmilarly, if 1/10 people in the US watches friends on Thursday nights, it doesn't matter if I pick 100 people, 1000 people, 10,000 people, or 100,000,000 people. 1/10 is 1/10.
Because the population is so large, and can be highly "pocketed" (people who live near eachother tend to watch the same stuff, because their friends tell them about it, or the local paper writes about it, or because someone local stars in it, or whatever) they need a VERY random, and very large (by statiscal measuring) sample.
Also, there is a wide variety of programming available, so if you only ask 100 people, you might get 1 person for each show so you ask 100000 people and then you notice that "Bob's Accordion World" only gets 1 viewer, and friends gets 10000.
In any case it works.
Here is a test you can do at home.
Take 2 dice. Roll them a few hundred times , and keep track of what you got each roll.
At the end, count how many ones you got, how many twos, how many threes etc.
You should end up with a nice bell curve, peaking around 6.
Now pick a random 50 rolls, and do the counts again. The bell curve should look almost identical, even though you only sampled a small percentage.
In this case, 50 is a large % of a few hundred rolls. But the graph would look the same if you rolled the dice 10,000,000 times. I just didn't think you would want to roll that many times
Uh. the DMCA specifically refers to circumventing access controls that prevent access to a copyrighted work. Since he doesn't get to acecss any "work" on your computer, it wouldn't be covered.
:)
You could however set up an autoresponder to any ping or incoming email, that sends out a copyrighted work, and say the only way you could get in was by having a license, then he would be circumventing, and then you could maybe nail him
This is simmilar to the people putting haiku headers into email so you know it isnt spam. If a smapper copies the header, they are infringing the copyright
If the sample is random, very small numbers give accurate results.
well, I totally disagree with this current lawsuit, but double jeopardy doesn't come into account.
Double Jeopordy refers to criminal action, all these anti trust suits are civil action (based on prior criminal acts tho)
Also, even if it was tried under criminal code, the suit is for a seperate market, which would presume seperate actions, which would be triable. (Just because you get off on a murder rap doesn't mean you can murder someone else for free).
Thirdly, Double Jeopordy is primarily a US thing (though much of europe has it as well. However, UK recently introduced legistlation weakening or canceling double jeopordy (Though I think it hasn't been signed yet, and may have been killed in process)
see : http://216.239.39.100/search?q=cache:JeAJHIKSZIQC
The EU competition regulator is concerned that Microsoft might take over the market for mobile phone software, according to a senior adviser to the European Commission. "Mobile phones are a very important market in Europe. The Commission is worried that Microsoft might take advantage of the strength of Windows in order to dominate in mobile phones," he said Uh, so they are suing because MS might do something? I forget what the legal term for this is, but I think this guy has a big stick up his ass. Besides, Stinger has 0 market penetration right now. Everyone is on just a big kick MSs ass because they are an easy target kick.
Actually, its not legal if the only intent is to damage the competitor, if its just a good marketing strategy, its not illegal.
Additionally, the fact that there is a subscription (XBox live) and accessories involved makes the case more complex. Companies can, will, and has argued successfully that dumping should be determined on the costs of everything the average buyer buys.
The dynamic you are talking about is NOT due to MS being a monopoly. It is because MS has more than one product line.
If Pepsi has a bad year, they still get to rely on taco bell, pizza hut, and doritos to make it through the year. Thats not a crime, its called diversification.
Sony does the same thing, they have TVs.
Note, you get to pay the taxes, even if you don't use cash. Using some made up cash would fall under barter rules, since at some point you either had to give them real cash, or something that someone else would pay real cash for.
Legal Tender is certainly a service provided for BY taxes, but is not in itself the source of tax revenue
Lion King was ripped from a Manga called Jungle Taitei. (Jungle Prince?)
Lead character : kimba. (To be fair, kimba/simba/timba and some others are african words for "cub" so its not like they ripped off a made up word)
However, some of the scenes were taken too, notable the cliff outcroping scene that is on all the lion king posters.
This was a huge controversy when lion king came out, Disney totally denied, said they had never even heard of the guys. Japanese guys said "We don't care, we take it as an hommage, and we ripped off bambi for our first picture anyway". Later, disney admited that several of the main animators were big fans, and had copies of the manga at their desks while making lion king.
This was also parodied in a great simpsons episode, with Lisa's mentor, (forget the first word) gums.
Gums was played by J.E. Jones, and after he died, they did this great thing where his ghost said a few words, then Darth vader said a few words, then mufasa said "Take care kimba.. I mean simba", and then JEJ said "This is CNN".
It was hilarious, and the number of people that would get the joke is really small.
Very true, but I really bet they bought the rights, or at least obtained them.
Also, those are pretty old movies, many may have fallen out of copyright already
commentary certainly allows quoting of the original. Just like a movie reviewer might show a still, or part of a scene.
But its not "quoting" if you put the whole thing. Can I quote your new book, in its entirety, and just add a few footnotes? Can I sell the result? Nope.
Note, I'm not saying this information shouldn't be out there. I'm just saying that a commentary or satire defense would quickly get squashed.
If he actually just quoted the document, or talked about it but didn't quote, then he would be much more in the clear.
Hrm, I think his defense on satire and commentary or non existant. Could I go to a movie, record the whole thing on my camcorder, and talk through it, saying what I thought, and then distribute the end product? Nope.
This also isnt satire, as he is making legitimate comments, not making jokes.
Now, if he re-wrote the document, but made everyhitng outrageous, that could be satire, but not the way it is now.
He is of course perfectly within his righs to publish commentary, but publishing the full text of the commented document I think is out of bounds (from a legal perspective)
which specifically was not written for preformance, but for readbility and as an example for proper coding practices
Are you saying that performance shouldn't be taken into consideration when deciding what "proper coding practices" are?
I disagree. If I write an elegant solution that takes up 500 lines, and you write a clunky solution that takes 1000 lines, who was more productive?
Now if we come up with the same solution, but I just type faster, so i have 1000 lines done, and you have 500 lines done, who is more productive?
Performance of a developer should be measured in (features implemented - bugs found)/time * some_constant_for_how_maintainable_the_code_is
anything else, and you are lying to yourself.
There weren't tricks to get the descrepancy large. Sun wrote the original J2EE version. MS wrote the .net version. If Sun chose to write their code in an inefficient manner, (and you care about lines of code) then perhaps that tells you something about OTHER apps that sun writes?
This is going to sound long a snotty troll, but i don't mean it that way.
Perhaps the reason you are looking at replacing the system by the time upgrades come up, is that you didn't buy something powerfull enough to begin with. It could be that the sizing tools are taking that into account.
Bah, typo. I meant to say water was finite.
The pipe is [should be] metered because you are using up a % of it at any given time. There ARE things as bandwidth hogs (I am one of them). And If I am tying up 50% of the bandwidth for a day, I should pay for 50% of the bandwidth for the day.
Water is not finite, It will all get recycled eventually. However the amount of water that can be pumped at any given time, and the number of gallons per minute is finite. Simmilarly, the number of megabytes per second is finite.
Until you get to the backbone (and even then, as the recent DoS attack shows) bandwidth is most certainly finite.
Actually, a dynamically linked library DOES impose this restriction. The GPL was specifically modified to handle this. Peopel were taking GPL code, and just wrapping it into a library, so they could release their code under a different license. FSF got upset, and GPL was modified.
In the article, the authors refer to TCPIP. I think this is an excellent example. If the TCP-IP protocol were released under GPL, then anyone writing an application that used TCPIP in any fashion would have to be GPL as well. That means : No commercial web browsers. No commercial email programs. No commercial ANYTHING that does updates via the web. No operating systems (Whens the last time an OS got released without TCPIP support?)
Thats the point. The authors are not saying everyone shouldnt have access to the govt. funded code. They should. they are saying that govt funded code can be used in a non free app. This is correct.
As long as the original work created by the funding is free for everyone to use, the world is in a good spot.
The government invented velcro. Velcro is free for everyone to use. Should anything that uses velcro have to be free too?
(Question: Can you realease a spec under GPL, or just an implementation of the spec. If you can release a spec under the GPL, would the GPL just apply to extentions/modifcations of the spec, or anything that implemented the spec (or happened to comply with the spec)
Acutally, selective enforcement is grounds for having a law thrown out. (or if not the entire law, your particular case) This is because when the law is selectively encorced it is usually because of racial/ethnic/politicol profiling, someone with an axe to grind, or they couldn't pin anything else on you.
For example : Did you know that it is illegal to drive your car across state lines if you have a loan on it, without first informing the holder of the loan that you intend to do so?
Of course, everyone does this, and nobody gets busted. Until they (yes, the infamous they) want to get you to testify, or give information, or just toss you in jail because they can't prove you were the "bad guy"
Of course, this has no relevence in the google case, which is not a government entity, and can make whatever terms it wants with whomever it wants.
If the improvements are obvious, they are not patentable. If they are pateneted, they will be invalidated.
Its pretty rare to get an improvement patent, since you have to be improving the process is a radical way. Thats usually a different process all together.