Without copyright laws, the GPL would be unnecessary. It was originally devised as a way to use copyright against itself. The big danger with this sort of engineering is the same as with releasing animals into an ecosystem to control some pest - eventually it can get out of control. In this case, there's now a bunch of people who use the GPL specifically because they want to restrict others, not just to have a safe haven of free and open things out there. And of course we've got all sorts of nasty license-conflict issues shooting around taking up time with legalese arguments when that time would better be spent programming.
So yeah, if getting rid of copyright means getting rid of the GPL then I say 'great, it succeeded in its purpose', not 'oh wait, we need to still protect this somehow'.
It's not a fundamental conflict, it just means that the storyteller needs to adjust their art for the medium. Instead of telling one story, the storyteller needs to figure out how to tell 500 stories which still allow the necessary components to come together. And there are well-known tricks for doing that sort of thing. Break the story up into modules which are mostly independant but have threads connecting them in the events that occured before the player arrives. Create some changes of dialogue to acknowledge the player's actions, or even separate branches though you have to be conservative about doing that lest the possibilities balloon. Don't think 'what do I want the villian to do to get the story moving' but rather 'how would the villian react to A? to B? to C?' given that the villian will have some set goal that he/she needs to accomplish, so it will eventually come back to the same things...
And so on. Now perhaps your complaint is 'but then I can't do anything I want to at all'. Thing is, you never could in any game. Each game is a finite universe. Clever designers have figured out how to make it look large, but they can't simulate everything you'd want to do any more than storytellers can create branches for everything you'd want to do. The problems that the gameplay designers face and the problems that the storytellers face are rather similar beneath it all in that way.
Depends on what you're measuring. If you measure the energy of that double pendulum you'll find its more reproducible than the exact x,y,z positions of the bobs. And of course if you measure averages over many runs you're set; the averages of one set of 1000 runs and the averages of a separate set of 1000 runs will be decently similar, and you can predict how different they should be by looking at the statistics you got from those different runs. Chaos doesn't mean 'give up', it means 'measure the things which tie to the qualitative aspects of the system because those things don't change even when the detailed state changes'
2D systems can be chaotic. I think what you're thinking of is systems with 2 degrees of freedom, which is different.
There are some visible differences between 2D and 3D turbulence though. 2D turbulence is marked by point vortices that form various correlated structures, but a lot of the turbulence is just carried by the interactions between point vortices of different sizes.
In 3D, those vortices become threads, and they can get tangled up and so on. Vorticity is no longer conserved and so you have things like as material gets pulled into an upwelling, its rotation is magnified (conservation of angular momentum). It's a process that occurs in tornados, for example.
It has to do with how you go from one to the other. Taking derivatives or integrals will increment or decrement the exponent by 1, and the different scalings are all derived from eachother.
The simplest way to see Kolmogorov scaling is to posit that the effect of viscosity in turbulence occurs only at the smallest lengthscale, and that aside from that energy must be shuttled via conservative mechanisms up and down between lengthscales. That is, energy is only dissipated at the smallest scale, so large scale motions can only lose energy by transferring it to smaller and smaller scales. It helps if we consider the incompressible case as this is simplest (constant density)
You then look at all the meaningful quantities in the system. We'll divide out the density of the fluid so units of mass won't appear. You have the net rate of energy dissipation (this folds in the viscosity) 'epsilon', which is in units of L^2/T^3, you have the lengthscale l which is in units of L, and you have the energy at that lengthscale dE/dl(l) which has units of L/T^2.
So if thats it, then you can write a relation dE/dl(l) is proportional to (epsilon)^(2/3)*l^(-1/3) simply to get the units to work. This gives you dE/dl(l)=A*l^(-1/3). Usually the Kolmogorov relation is given in terms of k, which is proportional to 1/l. So dl = -(1/k^2)dk, so we're left with dE/dk = A*k^(1/3-2) which is k^(-5/3) which is the usual relationship.
The interesting thing to note is, this can be different in 2D because there is a second conserved quantity, vorticity, which can only be dissipated at the viscous scale or the scale at which external forces act on the system. This means that a double-cascade can be observed, where you get a transition between a k^(-5/3) scaling to a k^(-3) scaling (or vice versa depending on how the system is driven).
So I suppose this means we can tell whether Van Gogh got his inspiration from looking at 3D turbulence - patterns in clouds and rising smoke and mixing liquids, or from 2D turbulence - whorls on soap bubbles and other thin film flows.
Whether he minds or not has nothing to do with whether he should be able to prevent you from doing so.
I think that as long as you identify that a modified version you produce is not the original version of the work, you should legally be able to do so and distribute it (though you can't with the current state of the law...), whether or not the original creator approves of the changes you've made. Since you've identified it as a modified version, issues with defaming the author go away, and whats left is the question: do you think a creator of some information should have the right to control that information once other people have obtained it? As for me, I answer 'no'.
I think a stronger argument is that one is the loss of current material possessions whereas the other may involve the loss of potential future possessions. While the owner of the copyright may be granted a legal right to restrict the copying of their material, they aren't granted a legal right to make X amount of sales, no fewer. Similarly, if I set up a donut store next to a local donut shop and undercut them on price because I can buy them bulk somewhere for less than they can, I may be depriving them of their future profits and depending on the full story there may be anticompetitive/collusion/whatever legal consequences, but what I've done is not stealing their donuts.
The problem with calling for alternate licenses is that the GPL's nature is to spread to other code, even if the thing it spreads to is a program that would be more appropriately licensed some other way. Whats being pointed out here is not 'I should be able to do what I want!' but more 'the way the GPL is designed shows that the designers wanted it to eventually cover everything, but look, its also inappropriate for some things and thus there is a contradiction'. There are some attempts to address this, such as the LGPL, but in general I expect that we're going to see a lot of unintended consequences crop up as GPL becomes increasingly common.
It would be sort of funny in a sad way if a ton of code ended up being reimplemented in a cleanroom setting like is done with reverse engineering so that there'd be a safe-for-appA copy and a safe-for-appB copy. I could see a lot of wasted effort coming out of this sort of thing. Really though when it comes down to it I doubt that any single license will satisfy everyone, so it's probably inevitable.
Er, so if most of the heating is say within the first few cm then we have at the most concentrated case something on the order of a cm^3 of material getting 3600 J/hour. At minimum, the density will be that of water which is about 1g/cm^3. I'm not sure what the heat capacity of flesh is, but its mostly water and 1g of water takes about 4000J to increase its temperature 1K. So a reasonable upper bound on heating effects is still within the 1K/h limit.
It seems to me you should be able to start in a higher dimensional space and just restrict it to have certain symmetries. Otherwise how does one make any predictions of physical observables in a three dimensional theory?
Forgive me for not being up on the details of this, but why do division algebras stop at 8? Does 16 not work?
What you're talking about works if I'm intending to spend money on a product anyhow. It's like chocolate that goes to support rainforest preservation or whatever. If what I really want to do is preserve the rainforest and thats my only reason for buying the chocolate, I should just donate to the final destination. But if I was going to buy chocolate anyhow then it makes sense.
So it really depends on what kinds of ads and whats the audience. I've never bought anything from Think Geek. I don't intend to. The stuff they sell doesn't overlap with the stuff I'm willing to buy. The reason online advertising is still going strong is that the lines are grey between gullible people who buy something because of the ad, people for whom branding is very effective, and people who simply don't even have the ad register on their brain despite it flashing up in front of their face because they've been exposed to so much advertising already.
It's sort of taking advantage of the companies in a way due to an implicit threat. If I have two companies that sell an equivalent product and only that, and both pay me to run their ads, and I run both ads, then the net redistribution of buyers is zero but both have had to pay me a little money to prevent the other from gaining an advantage via the eyes that see my sight. In the end perhaps what it should be is just that companies pay a site to not run their competitors' ads. So we get an ad-free internet and the companies don't get their buyers drained by competing ads. Oh well, I can dream right?
But I think its important to realize that this is feeding off of a fundamentally unstable source of income. Because if people stop paying attention, and advertisers realize this, then they will cut off the flow. So while in the short term a site can make some money which can then be stored away for future costs, there will be a point in which sites _will_ have to change the way they support themselves. You can try to convince people not to block ads to put that day off, try to convince them that its stealing the site's bandwidth to not also view the ads or some ridiculous thing like that, but in the end that will create a lot of baggage which becomes useless on the day that the advertising money source dries out or at least reduces its generosity.
I really do think that a site can give free content to the majority of people while depending on the donations of a few. Compared to other media, hosting and bandwidth costs are almost nil. They still exist, but its the sort of thing that very often the personal contributions of the founder can support a site for a thousand or more repeat visitors depending on the type of content (though it will financially strain the founder to support this individually). For text-based sites the ratio is probably something like 10000:1 visitors to donators necessary, maybe even higher. For music or video or large binaries then its another story. But then you have things like bittorrent to help spread the costs.
I guess my point is, the loss of advertising money shouldn't be seen as an armageddon of the free internet. Rather, it just has to be realized that _someone_ is paying either way and what it may come down to is changing how that money gets to the sites and making it clear to people exactly what the costs involved with a site are and how those costs are supported. I think part of what people fear about donations to a site is that it can just be a black hole and they never know if their money really prolonged the life of the site. Or perhaps the people who run the site are taking that money to pay their salaries. Or perhaps they'll say 'we received all this money, lets buy a server to increase the transfer rate for our visitors by 20k/sec' when all you really cared about was keeping the site alive at the current state. Eventually perhaps we'll see free websites run a lot like non-profit businesses, charities, and the like. Whether thats good or bad, well, if its whats necessary to keep them alive...
However those who actually pay for the site aren't people who view the ads but ignore them anyhow. The people who pay for the site are the small fraction who actually go and buy whatever it is being shown. Without that fraction, advertisers see no return on their investment in ads, and would stop funding sites anyhow.
So a logical extension is that we all must buy junk we see advertised to support our favorite sites? Well, in that case, why don't we just give the site the money directly. It may feel weird to financially support a free website but its certainly a better way to do it than to pay some other company so that they keep funding the site via advertisement.
Of course, its not that everyone has to pay in either model. In the advertisement model, you're basically feeding off of those who either have enough money they can toss it at random things at a whim (i.e. just based on seeing some banner ad deciding to spend $50 or so) or who are very gullible or easily manipulated (the same people who purchase things they get spam for).
So to make the analogy go over, what most sites need is a way to get donations from the idle rich and the gullible. They need hooks, gags, things which make no difference to the average site user but which someone with money to burn can donate to get special priveleges. The same sort of stuff that leads people to buy cellphone ringtones and screensavers and the like. Like letting their posts use some special inline images or whatever.
It's a freedom issue if you cannot legally choose to take that risk. That has nothing to do with whether or not the risk is large or small. It would be similar to a law which outlawed bungee jumping because of the risk of injury or death. Yes, not jumping is certainly a much lower risk endeavor than jumping, but its our right to choose to take those risks or not.
There's a problem which if we increase public healthcare at any point is going to rear its ugly head. It's already started in the form of the seatbelt stuff. Namely, you can no longer say that because other people will have to pay for the risks you take. So that gives politicians a huge license to legislate safety regulations. A mix of public and private could avoid that - you could sign some waiver saying that when you participate in whatever the 'unnecessarily risky endeavor' is these days you forfeit your right to public care for any injuries sustained during that activity. Of course, there's problems with that idea too...
If there are accounts for free usage, then this entire thing is a non-issue. Why? Because they can simply sum over all the traffic thats going over a single user account and limit that. Combining resources from multiple physical devices wouldn't work straight out since you'd have to have a separate login for each of them. Of course, some people will register multiple accounts under false names/addresses in which case it won't be too hard to catch if the usage isn't coming from the address that was registered.
On the other hand, most places right now that provide free wireless like coffee houses, universities, etc don't bother with logins. You can just go and use the network.
Now here's an interesting thought. You and your neighbors are all burst users. That is, you use a ton of bandwidth for say an hour or two each day. So you get together and rig up a local network such that you each can tap the sum bandwidth of everyone's accounts. Thats probably not even theft of service since the net used will be the net allocation for all those people, just staggered in such a way that its used to the full extent rather than permitted to go idle.
This raises a point of the difference between something with an explicit terms of service and something with implied terms of service. Something like napkins on the counter in a fast food place: you have some idea that taking many of them is inappropriate but its never written out anywhere. So I think they'd have a hard time going after you for taking 20 napkins (if you didn't go into a non-public area or take something thats not placed out there for public use). Or the opposite being that unless there's a signed form from the restaurant manager saying that you can take napkins, even if you take a single one for use with your meal its theft.
Similarly, I'd imagine the wireless service would have some problem going after people if all they do is make wireless available without any sort of registration or explicit signed agreement from the one receiving the service. It makes me wonder what the legal status of me placing a bin of items out in front of my house is. Have I somehow waived my property rights on those items and have given blanket permission for people to take them, or do I retain property rights and the right to prosecute someone who goes and takes one of the items. Seems like a good scam in the making there...
But of course if there's a TOS/registration/etc it's a much clearer case.
You'll end up crashing a lot of social events that way. Is it one guy in that house using 20 people's allocation of bandwidth or twenty people getting together for a party/study session/cohabitation/whatever. It'll be fun seeing all the lawsuits that the victims of indiscriminate fishing expeditions like driving around the neighborhood with a signal meter and barging into any house with a spike will levy against the company or local police.
Nah, what you've gotta do is keep a catalogue of the state of each country in the world on that graph and plan your moves so that you get out before you get to dependency or bondage. Of course, if you're apathetic you probably won't go through the trouble to do that.
The way I see it, arguments based on morality end up amounting to after-the-fact justifications. It's a way to convince yourself you shouldn't feel guilty for doing whatever you were going to do anyhow for more visceral reasons.
I mean, take the Hitler example. Does it really matter if it was a moral right of people to stop him? Such has happened, and most people are happier as a result of that. Those who acted to stop it made life better for themselves and those who would have been killed had Hitler been allowed to continue to expand unchecked.
Or a less Godwin-esque case, posit the situation that you observe a mugging in progress and have the ability to stop it. Does it matter whether you stop it because you feel its morally wrong for the mugger to steal, or if its just because you feel compassion for the victim and desire to help them?
Acting out of that compassion is dangerous when you misunderstand the desires of the victim - thats the point of relevance in cultural differences and why non-interference can make sense. If it turns out the mugger and the victim were just people practicing a play, or if the victim had some odd fetish for being mugged or something like that, then you'd achieve the opposite of the effect you wanted due to your compassion. You can try to judge if thats likely or not - in the case of the mugging, you observe how the victim is responding, or since you probably have a decent grasp of how humans in general respond to being mugged you take a chance and assume their reaction.
When you have wildly different cultures than your own, that sort of assumption can fail. Or when you have wildly different belief systems or anything like that where your ability to judge how another person would respond to your interference is compromised.
Interesting argument. I wonder, does this apply to software?
Right now, if Microsoft donates a million copies of Windows to various school systems/businesses/homeless people/wherever, do they get to deduct it? Even though it essentially only costs them a lost opportunity to sell that particular copy of the information?
As I see it the main reason other people are even folded into the idea of an RPG is that at the time tabletop RPGs were developed - and that is where the term and genre of game originated - the only way to have any sort of satisfying interaction between a player's character and their world was if those interactions were 'calculated' by humans. That is, the other players and the GM.
As game AI gets better and programmers actually take that sort of thing into account, you can very much approach that sort of dynamic in a CRPG. It won't be AS rich or open-ended, but its clearly something above and beyond the very first attempts at CRPGs. When the world in-simulo provides an increasingly convincing audience, I think that should even satisfy the hard-line view that an audience is needed for role-play. The computer thats running the game is the audience.
Otherwise we could easily get into Descartian arguments about 'well how do I even know the other players are observing me? I can't prove they're real!' sorts of things, which I think is for the most part unproductive.
Or to put it another way - if the player feels like they're immersed in a role, then its roleplaying. The audience may be necessary to invoke that feeling, but whether its a real audience or a very well simulated one doesn't matter so long as that feeling is produced.
Psst. If they're afraid they can always refuse to make an ebook. Of course, if they go out of business because no one wants to buy their book since there ARE electronic versions of that information available out there one way or another then good riddance to redundancy.
Reference books won't simply disappear unless there exists something out there which replaces their function and information adequately. If freely contributed materials from professionals performs that function, then so be it. If publishers go out on a limb and make electronic books available for a price, either the piracy of those materials will kill that publishing company in which case another will step up that won't do electronic materials, or they'll be able to make enough income to stay afloat despite piracy.
All of this only becomes a problem if some governing body decides to make laws saying 'publishers can't, unless its free'. Which is of the same vein of problem as a governing body making a law that says 'once the content is collected and published by one company/person/etc then no one else can publish that information'.
Information is cheap to copy and distribute but expensive to organize in a cogent fashion. Perhaps we'll lose some of that cogence when it becomes less profitable to be an organizer-of-information, but the information itself will stay around. And if that organization is absolutely critical to actually using the information, then there will still be a market for it in whatever form it takes. Of course, if that market is filled by people willing to do it for free thats no less valid than if its filled by people who will only do it for money. It just remains to be seen where it will fall.
Well, it'd be an example of a piece of experimental evidence in a field where basically you can't test anything because of the lack of experimental evidence. If that mass ratio varies, then this also means we can measure how it varies, and each theory can make a prediction and... I suspect they bring up string theory because thats the one everyone knows, not because this particular piece of evidence gives support to string theory only. But on the other hand, I can't really say off the top of my head which of the successor theories predicts that the masses of composite particles can have a time dependence (which would amount to either the strength of the strong force or electromagnetic force having a time dependence, or the quark and/or electron self-masses have a time dependence). If all of them do, and can't actually predict how it would vary, then yeah this isn't so useful.
But if its true its still pretty neat and its like water in the desert for the field of post-QM/GR theories.
There are plenty of text adventures still being made, just not commercially. The engines have gotten pretty sophisticated these days (in the sense that there's a lot less of the 'okay, exactly what grammar does it want me to use to tell it to do this thing I have in mind' as well as having a lot of the small details be taken care of by reasonable default object behaviors).
Do an experiment. You won't get an exact value. You'll get scatter in your data. Numerics are no different. If you're careless and treat your numerical result as an exact analytical result then yes you'll get burned. If you treat your numerical result the same way you'd treat the output of a monte-carlo analysis of something then you know you aren't exact but by doing it over and over with small variations, by analyzing the nature of your errors, etc, you have some idea of just how far off you are.
This is part of the reason that the best numerics work validates itself against analytical solutions first. The second thing that can really help numerics work be useful is when the purpose isn't just to determine a number like the critical temperature of a certain phase transition (notoriously nasty to do with numerics since the number you get depends on the system size and the time of simulation scaling a certain way together - if you simulate too long on a small system, you get too small a Tc, and if you don't simulate long enough, its too big...) But rather they construct something that amounts to a binary check which can be extracted from the quantitative data but which also has a huge visible signal. Like a system-wide symmetry breaking. Something like 'at this point, I see no oscillations, but when I add in the term I'm testing for I suddenly get oscillations'. Or 'at this point, the system is zero everywhere, but when I add in this term the system divides itself up into regions of -1 and 1'. Its not a sufficient condition for a theory but it is a necessary one that it reproduce the qualitative properties that are actually observed. Numerics can help you get at that in cases which would be impossible on paper.
Anyhow, looking at my own history of calculations of the paper variety, I'm just as likely to drop a negative sign when doing paper math as I am to insert a bug in my code. Of course, there's that old addage... computers let you make the same mistakes you'd normally make, but a lot faster.
Why is it necessary that companies get involved with linux? This is sort of like 'if you ever want to see any closed-source commercial software you'd better go out and buy some to support the guys who make it!'. Well, I don't particularly care to see that - if I'm in the mood for that, I can just go and use a windows environment for a bit. The additional hardware support/etc that would be gained by getting companies interested in selling on linux isn't necessarily worth the dilution it brings to opensource software in my opinion. Not to mention how some programs have gone the path of MS Office and have become bloated in order to make them appeal to a wider audience. Snobbery? Perhaps, but if Linux loses technical and ideological advantages to gain market share, then what's the point?
Slashdot Mirror
Without copyright laws, the GPL would be unnecessary. It was originally devised as a way to use copyright against itself. The big danger with this sort of engineering is the same as with releasing animals into an ecosystem to control some pest - eventually it can get out of control. In this case, there's now a bunch of people who use the GPL specifically because they want to restrict others, not just to have a safe haven of free and open things out there. And of course we've got all sorts of nasty license-conflict issues shooting around taking up time with legalese arguments when that time would better be spent programming.
So yeah, if getting rid of copyright means getting rid of the GPL then I say 'great, it succeeded in its purpose', not 'oh wait, we need to still protect this somehow'.
It's not a fundamental conflict, it just means that the storyteller needs to adjust their art for the medium. Instead of telling one story, the storyteller needs to figure out how to tell 500 stories which still allow the necessary components to come together. And there are well-known tricks for doing that sort of thing. Break the story up into modules which are mostly independant but have threads connecting them in the events that occured before the player arrives. Create some changes of dialogue to acknowledge the player's actions, or even separate branches though you have to be conservative about doing that lest the possibilities balloon. Don't think 'what do I want the villian to do to get the story moving' but rather 'how would the villian react to A? to B? to C?' given that the villian will have some set goal that he/she needs to accomplish, so it will eventually come back to the same things...
And so on. Now perhaps your complaint is 'but then I can't do anything I want to at all'. Thing is, you never could in any game. Each game is a finite universe. Clever designers have figured out how to make it look large, but they can't simulate everything you'd want to do any more than storytellers can create branches for everything you'd want to do. The problems that the gameplay designers face and the problems that the storytellers face are rather similar beneath it all in that way.
Depends on what you're measuring. If you measure the energy of that double pendulum you'll find its more reproducible than the exact x,y,z positions of the bobs. And of course if you measure averages over many runs you're set; the averages of one set of 1000 runs and the averages of a separate set of 1000 runs will be decently similar, and you can predict how different they should be by looking at the statistics you got from those different runs. Chaos doesn't mean 'give up', it means 'measure the things which tie to the qualitative aspects of the system because those things don't change even when the detailed state changes'
2D systems can be chaotic. I think what you're thinking of is systems with 2 degrees of freedom, which is different.
There are some visible differences between 2D and 3D turbulence though. 2D turbulence is marked by point vortices that form various correlated structures, but a lot of the turbulence is just carried by the interactions between point vortices of different sizes.
In 3D, those vortices become threads, and they can get tangled up and so on. Vorticity is no longer conserved and so you have things like as material gets pulled into an upwelling, its rotation is magnified (conservation of angular momentum). It's a process that occurs in tornados, for example.
It has to do with how you go from one to the other. Taking derivatives or integrals will increment or decrement the exponent by 1, and the different scalings are all derived from eachother.
The simplest way to see Kolmogorov scaling is to posit that the effect of viscosity in turbulence occurs only at the smallest lengthscale, and that aside from that energy must be shuttled via conservative mechanisms up and down between lengthscales. That is, energy is only dissipated at the smallest scale, so large scale motions can only lose energy by transferring it to smaller and smaller scales. It helps if we consider the incompressible case as this is simplest (constant density)
You then look at all the meaningful quantities in the system. We'll divide out the density of the fluid so units of mass won't appear. You have
the net rate of energy dissipation (this folds in the viscosity) 'epsilon', which is in units of L^2/T^3, you have the lengthscale l which is in units of L, and you have the energy at that lengthscale dE/dl(l) which has units of L/T^2.
So if thats it, then you can write a relation dE/dl(l) is proportional to (epsilon)^(2/3)*l^(-1/3) simply to get the units to work. This gives you dE/dl(l)=A*l^(-1/3). Usually the Kolmogorov relation is given in terms of k, which is proportional to 1/l. So dl = -(1/k^2)dk, so we're left with dE/dk = A*k^(1/3-2) which is k^(-5/3) which is the usual relationship.
The interesting thing to note is, this can be different in 2D because there is a second conserved quantity, vorticity, which can only be dissipated at the viscous scale or the scale at which external forces act on the system. This means that a double-cascade can be observed, where you get a transition between a k^(-5/3) scaling to a k^(-3) scaling (or vice versa depending on how the system is driven).
So I suppose this means we can tell whether Van Gogh got his inspiration from looking at 3D turbulence - patterns in clouds and rising smoke and mixing liquids, or from 2D turbulence - whorls on soap bubbles and other thin film flows.
Whether he minds or not has nothing to do with whether he should be able to prevent you from doing so.
I think that as long as you identify that a modified version you produce is not the original version of the work, you should legally be able to do so and distribute it (though you can't with the current state of the law...), whether or not the original creator approves of the changes you've made. Since you've identified it as a modified version, issues with defaming the author go away, and whats left is the question: do you think a creator of some information should have the right to control that information once other people have obtained it? As for me, I answer 'no'.
I think a stronger argument is that one is the loss of current material possessions whereas the other may involve the loss of potential future possessions. While the owner of the copyright may be granted a legal right to restrict the copying of their material, they aren't granted a legal right to make X amount of sales, no fewer. Similarly, if I set up a donut store next to a local donut shop and undercut them on price because I can buy them bulk somewhere for less than they can, I may be depriving them of their future profits and depending on the full story there may be anticompetitive/collusion/whatever legal consequences, but what I've done is not stealing their donuts.
The problem with calling for alternate licenses is that the GPL's nature is to spread to other code, even if the thing it spreads to is a program that would be more appropriately licensed some other way. Whats being pointed out here is not 'I should be able to do what I want!' but more 'the way the GPL is designed shows that the designers wanted it to eventually cover everything, but look, its also inappropriate for some things and thus there is a contradiction'. There are some attempts to address this, such as the LGPL, but in general I expect that we're going to see a lot of unintended consequences crop up as GPL becomes increasingly common.
It would be sort of funny in a sad way if a ton of code ended up being reimplemented in a cleanroom setting like is done with reverse engineering so that there'd be a safe-for-appA copy and a safe-for-appB copy. I could see a lot of wasted effort coming out of this sort of thing. Really though when it comes down to it I doubt that any single license will satisfy everyone, so it's probably inevitable.
Er, so if most of the heating is say within the first few cm then we have at the most concentrated case something on the order of a cm^3 of material getting 3600 J/hour. At minimum, the density will be that of water which is about 1g/cm^3. I'm not sure what the heat capacity of flesh is, but its mostly water and 1g of water takes about 4000J to increase its temperature 1K. So a reasonable upper bound on heating effects is still within the 1K/h limit.
It seems to me you should be able to start in a higher dimensional space and just restrict it to have certain symmetries. Otherwise how does one make any predictions of physical observables in a three dimensional theory?
Forgive me for not being up on the details of this, but why do division algebras stop at 8? Does 16 not work?
What you're talking about works if I'm intending to spend money on a product anyhow. It's like chocolate that goes to support rainforest preservation or whatever. If what I really want to do is preserve the rainforest and thats my only reason for buying the chocolate, I should just donate to the final destination. But if I was going to buy chocolate anyhow then it makes sense.
So it really depends on what kinds of ads and whats the audience. I've never bought anything from Think Geek. I don't intend to. The stuff they sell doesn't overlap with the stuff I'm willing to buy. The reason online advertising is still going strong is that the lines are grey between gullible people who buy something because of the ad, people for whom branding is very effective, and people who simply don't even have the ad register on their brain despite it flashing up in front of their face because they've been exposed to so much advertising already.
It's sort of taking advantage of the companies in a way due to an implicit threat. If I have two companies that sell an equivalent product and only that, and both pay me to run their ads, and I run both ads, then the net redistribution of buyers is zero but both have had to pay me a little money to prevent the other from gaining an advantage via the eyes that see my sight. In the end perhaps what it should be is just that companies pay a site to not run their competitors' ads. So we get an ad-free internet and the companies don't get their buyers drained by competing ads. Oh well, I can dream right?
But I think its important to realize that this is feeding off of a fundamentally unstable source of income. Because if people stop paying attention, and advertisers realize this, then they will cut off the flow. So while in the short term a site can make some money which can then be stored away for future costs, there will be a point in which sites _will_ have to change the way they support themselves. You can try to convince people not to block ads to put that day off, try to convince them that its stealing the site's bandwidth to not also view the ads or some ridiculous thing like that, but in the end that will create a lot of baggage which becomes useless on the day that the advertising money source dries out or at least reduces its generosity.
I really do think that a site can give free content to the majority of people while depending on the donations of a few. Compared to other media, hosting and bandwidth costs are almost nil. They still exist, but its the sort of thing that very often the personal contributions of the founder can support a site for a thousand or more repeat visitors depending on the type of content (though it will financially strain the founder to support this individually). For text-based sites the ratio is probably something like 10000:1 visitors to donators necessary, maybe even higher. For music or video or large binaries then its another story. But then you have things like bittorrent to help spread the costs.
I guess my point is, the loss of advertising money shouldn't be seen as an armageddon of the free internet. Rather, it just has to be realized that _someone_ is paying either way and what it may come down to is changing how that money gets to the sites and making it clear to people exactly what the costs involved with a site are and how those costs are supported. I think part of what people fear about donations to a site is that it can just be a black hole and they never know if their money really prolonged the life of the site. Or perhaps the people who run the site are taking that money to pay their salaries. Or perhaps they'll say 'we received all this money, lets buy a server to increase the transfer rate for our visitors by 20k/sec' when all you really cared about was keeping the site alive at the current state. Eventually perhaps we'll see free websites run a lot like non-profit businesses, charities, and the like. Whether thats good or bad, well, if its whats necessary to keep them alive...
However those who actually pay for the site aren't people who view the ads but ignore them anyhow. The people who pay for the site are the small fraction who actually go and buy whatever it is being shown. Without that fraction, advertisers see no return on their investment in ads, and would stop funding sites anyhow.
So a logical extension is that we all must buy junk we see advertised to support our favorite sites? Well, in that case, why don't we just give the site the money directly. It may feel weird to financially support a free website but its certainly a better way to do it than to pay some other company so that they keep funding the site via advertisement.
Of course, its not that everyone has to pay in either model. In the advertisement model, you're basically feeding off of those who either have enough money they can toss it at random things at a whim (i.e. just based on seeing some banner ad deciding to spend $50 or so) or who are very gullible or easily manipulated (the same people who purchase things they get spam for).
So to make the analogy go over, what most sites need is a way to get donations from the idle rich and the gullible. They need hooks, gags, things which make no difference to the average site user but which someone with money to burn can donate to get special priveleges. The same sort of stuff that leads people to buy cellphone ringtones and screensavers and the like. Like letting their posts use some special inline images or whatever.
It's a freedom issue if you cannot legally choose to take that risk. That has nothing to do with whether or not the risk is large or small. It would be similar to a law which outlawed bungee jumping because of the risk of injury or death. Yes, not jumping is certainly a much lower risk endeavor than jumping, but its our right to choose to take those risks or not.
There's a problem which if we increase public healthcare at any point is going to rear its ugly head. It's already started in the form of the seatbelt stuff. Namely, you can no longer say that because other people will have to pay for the risks you take. So that gives politicians a huge license to legislate safety regulations. A mix of public and private could avoid that - you could sign some waiver saying that when you participate in whatever the 'unnecessarily risky endeavor' is these days you forfeit your right to public care for any injuries sustained during that activity. Of course, there's problems with that idea too...
If there are accounts for free usage, then this entire thing is a non-issue. Why? Because they can simply sum over all the traffic thats going over a single user account and limit that. Combining resources from multiple physical devices wouldn't work straight out since you'd have to have a separate login for each of them. Of course, some people will register multiple accounts under false names/addresses in which case it won't be too hard to catch if the usage isn't coming from the address that was registered.
On the other hand, most places right now that provide free wireless like coffee houses, universities, etc don't bother with logins. You can just go and use the network.
Now here's an interesting thought. You and your neighbors are all burst users. That is, you use a ton of bandwidth for say an hour or two each day. So you get together and rig up a local network such that you each can tap the sum bandwidth of everyone's accounts. Thats probably not even theft of service since the net used will be the net allocation for all those people, just staggered in such a way that its used to the full extent rather than permitted to go idle.
This raises a point of the difference between something with an explicit terms of service and something with implied terms of service. Something like napkins on the counter in a fast food place: you have some idea that taking many of them is inappropriate but its never written out anywhere. So I think they'd have a hard time going after you for taking 20 napkins (if you didn't go into a non-public area or take something thats not placed out there for public use). Or the opposite being that unless there's a signed form from the restaurant manager saying that you can take napkins, even if you take a single one for use with your meal its theft.
Similarly, I'd imagine the wireless service would have some problem going after people if all they do is make wireless available without any sort of registration or explicit signed agreement from the one receiving the service. It makes me wonder what the legal status of me placing a bin of items out in front of my house is. Have I somehow waived my property rights on those items and have given blanket permission for people to take them, or do I retain property rights and the right to prosecute someone who goes and takes one of the items.
Seems like a good scam in the making there...
But of course if there's a TOS/registration/etc it's a much clearer case.
You'll end up crashing a lot of social events that way. Is it one guy in that house using 20 people's allocation of bandwidth or twenty people getting together for a party/study session/cohabitation/whatever. It'll be fun seeing all the lawsuits that the victims of indiscriminate fishing expeditions like driving around the neighborhood with a signal meter and barging into any house with a spike will levy against the company or local police.
Nah, what you've gotta do is keep a catalogue of the state of each country in the world on that graph and plan your moves so that you get out before you get to dependency or bondage. Of course, if you're apathetic you probably won't go through the trouble to do that.
The way I see it, arguments based on morality end up amounting to after-the-fact justifications. It's a way to convince yourself you shouldn't feel guilty for doing whatever you were going to do anyhow for more visceral reasons.
I mean, take the Hitler example. Does it really matter if it was a moral right of people to stop him? Such has happened, and most people are happier as a result of that. Those who acted to stop it made life better for themselves and those who would have been killed had Hitler been allowed to continue to expand unchecked.
Or a less Godwin-esque case, posit the situation that you observe a mugging in progress and have the ability to stop it. Does it matter whether you stop it because you feel its morally wrong for the mugger to steal, or if its just because you feel compassion for the victim and desire to help them?
Acting out of that compassion is dangerous when you misunderstand the desires of the victim - thats the point of relevance in cultural differences and why non-interference can make sense. If it turns out the mugger and the victim were just people practicing a play, or if the victim had some odd fetish for being mugged or something like that, then you'd achieve the opposite of the effect you wanted due to your compassion. You can try to judge if thats likely or not - in the case of the mugging, you observe how the victim is responding, or since you probably have a decent grasp of how humans in general respond to being mugged you take a chance and assume their reaction.
When you have wildly different cultures than your own, that sort of assumption can fail. Or when you have wildly different belief systems or anything like that where your ability to judge how another person would respond to your interference is compromised.
Interesting argument. I wonder, does this apply to software?
Right now, if Microsoft donates a million copies of Windows to various school systems/businesses/homeless people/wherever, do they get to deduct it? Even though it essentially only costs them a lost opportunity to sell that particular copy of the information?
As I see it the main reason other people are even folded into the idea of an RPG is that at the time tabletop RPGs were developed - and that is where the term and genre of game originated - the only way to have any sort of satisfying interaction between a player's character and their world was if those interactions were 'calculated' by humans. That is, the other players and the GM.
As game AI gets better and programmers actually take that sort of thing into account, you can very much approach that sort of dynamic in a CRPG. It won't be AS rich or open-ended, but its clearly something above and beyond the very first attempts at CRPGs. When the world in-simulo provides an increasingly convincing audience, I think that should even satisfy the hard-line view that an audience is needed for role-play. The computer thats running the game is the audience.
Otherwise we could easily get into Descartian arguments about 'well how do I even know the other players are observing me? I can't prove they're real!' sorts of things, which I think is for the most part unproductive.
Or to put it another way - if the player feels like they're immersed in a role, then its roleplaying. The audience may be necessary to invoke that feeling, but whether its a real audience or a very well simulated one doesn't matter so long as that feeling is produced.
Psst. If they're afraid they can always refuse to make an ebook. Of course, if they go out of business because no one wants to buy their book since there ARE electronic versions of that information available out there one way or another then good riddance to redundancy.
Reference books won't simply disappear unless there exists something out there which replaces their function and information adequately. If freely contributed materials from professionals performs that function, then so be it. If publishers go out on a limb and make electronic books available for a price, either the piracy of those materials will kill that publishing company in which case another will step up that won't do electronic materials, or they'll be able to make enough income to stay afloat despite piracy.
All of this only becomes a problem if some governing body decides to make laws saying 'publishers can't, unless its free'. Which is of the same vein of problem as a governing body making a law that says 'once the content is collected and published by one company/person/etc then no one else can publish that information'.
Information is cheap to copy and distribute but expensive to organize in a cogent fashion. Perhaps we'll lose some of that cogence when it becomes less profitable to be an organizer-of-information, but the information itself will stay around. And if that organization is absolutely critical to actually using the information, then there will still be a market for it in whatever form it takes. Of course, if that market is filled by people willing to do it for free thats no less valid than if its filled by people who will only do it for money. It just remains to be seen where it will fall.
Well, it'd be an example of a piece of experimental evidence in a field where basically you can't test anything because of the lack of experimental evidence. If that mass ratio varies, then this also means we can measure how it varies, and each theory can make a prediction and... I suspect they bring up string theory because thats the one everyone knows, not because this particular piece of evidence gives support to string theory only. But on the other hand, I can't really say off the top of my head which of the successor theories predicts that the masses of composite particles can have a time dependence (which would amount to either the strength of the strong force or electromagnetic force having a time dependence, or the quark and/or electron self-masses have a time dependence). If all of them do, and can't actually predict how it would vary, then yeah this isn't so useful.
But if its true its still pretty neat and its like water in the desert for the field of post-QM/GR theories.
There are plenty of text adventures still being made, just not commercially. The engines have gotten pretty sophisticated these days (in the sense that there's a lot less of the 'okay, exactly what grammar does it want me to use to tell it to do this thing I have in mind' as well as having a lot of the small details be taken care of by reasonable default object behaviors).
Check out http://www.ifarchive.org/ for a bunch of games and
http://www.tads.org/ for a particularly good IF engine.
Do an experiment. You won't get an exact value. You'll get scatter in your data. Numerics are no different. If you're careless and treat your numerical result as an exact analytical result then yes you'll get burned. If you treat your numerical result the same way you'd treat the output of a monte-carlo analysis of something then you know you aren't exact but by doing it over and over with small variations, by analyzing the nature of your errors, etc, you have some idea of just how far off you are.
This is part of the reason that the best numerics work validates itself against analytical solutions first. The second thing that can really help numerics work be useful is when the purpose isn't just to determine a number like the critical temperature of a certain phase transition (notoriously nasty to do with numerics since the number you get depends on the system size and the time of simulation scaling a certain way together - if you simulate too long on a small system, you get too small a Tc, and if you don't simulate long enough, its too big...) But rather they construct something that amounts to a binary check which can be extracted from the quantitative data but which also has a huge visible signal. Like a system-wide symmetry breaking. Something like 'at this point, I see no oscillations, but when I add in the term I'm testing for I suddenly get oscillations'. Or 'at this point, the system is zero everywhere, but when I add in this term the system divides itself up into regions of -1 and 1'. Its not a sufficient condition for a theory but it is a necessary one that it reproduce the qualitative properties that are actually observed. Numerics can help you get at that in cases which would be impossible on paper.
Anyhow, looking at my own history of calculations of the paper variety, I'm just as likely to drop a negative sign when doing paper math as I am to insert a bug in my code. Of course, there's that old addage... computers let you make the same mistakes you'd normally make, but a lot faster.
Why is it necessary that companies get involved with linux? This is sort of like 'if you ever want to see any closed-source commercial software you'd better go out and buy some to support the guys who make it!'. Well, I don't particularly care to see that - if I'm in the mood for that, I can just go and use a windows environment for a bit. The additional hardware support/etc that would be gained by getting companies interested in selling on linux isn't necessarily worth the dilution it brings to opensource software in my opinion. Not to mention how some programs have gone the path of MS Office and have become bloated in order to make them appeal to a wider audience. Snobbery? Perhaps, but if Linux loses technical and ideological advantages to gain market share, then what's the point?