Thanks for posting that, I didn't even realize ZFS was in BSD. On a scale from "rock solid for critical production use" to "for data you don't care about anyway," how stable is it in the latest FreeBSD release? Not that OpenSolaris has been without glitches, but I'm always nervous when contemplating big changes to our storage.
I often print things out so that I can read them on the run. A decent e-reader would eliminate that need entirely, but tragically no one makes one yet. I have high hopes that this will be fixed sometime this year or next. That's not the only obstacle to a paperless office, but that's by far the #1 paper consumer for me.
Misuse of statistics is well-represented in scientific articles. Other things that are well-represented are poor knowledge and reasoning in the area of the subject discipline, inept writing, misleading or unhelpful graphics, poor scholarship, etc. Sturgeon's Law applies across the board.
Having read a fair number of sky-is-falling articles about statistics in science, and having worked with my share of researchers (MDs and PhDs in a variety of fields) who think everything is rosy, I'm pretty sure that the truth is somewhere in between. A minor saving grace is the fact that getting the statistics wrong is not the same as getting the answer wrong. Although it's certainly quite common to find published articles that make claims with no support whatsoever, in my experience it's much more common to find articles where the inappropriate statistics just mean the support isn't nearly as strong as claimed. Spurious results tend, though not as reliably as we'd like, to get weeded out by the literature. I rarely read an article that isn't specifically about methodology in which the methods/statistics are really solid, but I also rarely read an article in which unsound statistics undermines the entire contribution.
Re: the licensing issues, I was really just asking if there's any chance Sun might consider re-licensing the ZFS code. Last I checked, this was a non-starter, but a lot has changed in the past year or so. That said, I forgot about the "rampant layering violations," and I can see how this would be a permanent problem even if the licensing issues could be worked out. I'm still struggling, though, because I hate maintaining an OpenSolaris box solely for ZFS. At the same time, I love ZFS, and it's going to be a while before I can feel equally comfortable with btrfs.
I'm surprised you've found Linux filesystems such comparatively poor performers, but that's good information to have. I live in a somewhat lower performance neighborhood, and would not have noticed this.
So far as I can tell, zfs is the only piece of opensolaris that's exciting enough to make anyone want to install if if they'd otherwise want to install a linux distribution. With that in mind, could someone post an authoritative update on the supposedly intractable licensing issues that prevent ZFS from being incorporated into the linux kernel? Is it still hopeless?
I'm sure this device has a lot going for it, but for my reading enjoyment, I'm still waiting for devices with better screens from qualcomm/mirasol or pixel qi. In nicer packages, I hope. I don't need two flawed screens hinged together, I need a single screen that's more functional than e-ink.
It sounds like you're assuming the charger will exactly max out a dedicated residential circuit. I don't honestly know how much current these things draw, but I'll bet it's a lot less than 20 amps. If your assumptions are correct, then the quickcharger would be extremely impractical in residential settings.
There is no such thing as the "correct" viewing distance. I like to watch at such a distance that the screen takes up about 20 degrees of visual angle. That makes me comfortable, even though I know a lot of people enjoy 30 degrees, and some of them consider me "insane." I don't watch movies to demonstrate my sanity, I watch them for enjoyment. It doesn't matter to me even a tiny bit that I might be missing out on some of the resolution of the image. If some new video standard had a zillion times the resolution of blu-ray, I wouldn't sit with my face pressed up against the tv.
Recent versions of gcc support OpenMP, and there's now experimental support for a multithreading library that I gather is going to be in the next c++ standard. These don't solve everyone's problems, but certainly it's getting easier, not harder, to take better advantage of multi-processor multi-core systems. I recently test retrofit some of my own code with OpenMP, and it was ridiculously easy. Five years ago it would have been a much more irritating process. I realize not everyone develops in c/c++, nor does everyone use a compiler that supports OpenMP. But I doubt it's actually getting harder, probably just the rate at which it's getting easier is not the same for everyone.
This is a little misleading. fMRI can't tell the difference between blood flow associated with excitatory and inhibitory neural activity, and quite likely many or all BOLD responses involve some combination of both. That does not, however, mean that fMRI can't tell the difference between any activity and the willful inhibition of that activity. That may be true in some cases, but it's not true in general.
It's also not true that fMRI can't in principle tell the difference between your reaction to your wife and some attractive celebrity. Even if it were true that the BOLD signal associated with active inhibition of some process were equivalent to the BOLD signal associated with that same process, this would still be a perfectly reasonable comparison, even if we can't tell the difference between involuntary responses and responses meant to mislead the system. If this weren't the true, then none of the most basic uses of fMRI would work at all. E.g., you can tell the difference between finger tapping and not finger tapping. I don't know about willful inhibition of finger tapping specifically.
I agree that the use of fMRI as a lie detector is fruitless, but it's more an issue of practicality and devising a useful contrast than some in-princple reason why we can't tell the difference between a mental state and its opposite.
I do work in fMRI and I've read many hundreds of articles on the subject, so I feel pretty comfortable with the above. Others working in fMRI might disagree, but I don't think anyone in the field would say that these issues are settled.
I'm with the OP on this. As the father of a 5 and a 3 year old, I know from experience that I can't tell the 3 year old to pick up the blocks, but I can point to a block, tell him to pick up *that* block, and then point to the bucket and tell him to put the block in the bucket. Then I can repeat the process x times where x equals the number of blocks on the floor.
So as the OP said, if these eggheads would just have kids, they would know the outcome of their "research" through experience and intuition.
Maybe the eggheads have learned to base their research on more than just a single parent's experiences with two kids. I know from experience that 3 year olds can learn to pick up blocks without going one-by-one, and do many other more complex things, given the right mood or the right motivation. That's different from your experience, I guess. But I don't know if my experience is typical.
Concerning behavioral research (my career has nothing particular to do with this, I'm a molecular biologist concerned with early brain development), there are repeatedly studies presented here on/. with crappy conclusions (and good commentaries from the/. crowd).
When you say "crappy conclusions," do you mean conclusions that are poorly supported by the evidence, or conclusions that are well supported by the evidence but not that interesting, or what?
I've rarely seen much in the way of good commentaries on science from the crowd here. Sometimes a few people who actually know the area in question will post some insightful comments. But the rest generally respond to the popular report that's been posted, which almost invariably misses the point of the research. I'd be curious to hear what most Slashdotters think the take-home message was meant to be from the study under discussion here.
As a behavioral researcher, I certainly sympathize with your contempt for the field. But at the same time, I think there are good reasons to ask scientific questions that involve behavior (even more so neurology). The fact that some people do it poorly doesn't change that, nor does the fact that some people prefer an extreme reductionist approach.
No, they're not. They "merely" apply science to specific well-known problems.
Anecdote and casual observation accumulated over time equate to empirical evidence.
Absolutely not! If we went by your standard of evidence, we would consider there to be a mountain of evidence that the Sun goes around the Earth. Nowadays it's easy to see that it's the other way round, but if we went by your standard of evidence it's doubtful that our collective scientific knowledge would actually have gotten far enough to discover that.
You're no scientist and have no idea what scientists actually do.
No, they're not. They "merely" apply science to specific well-known problems.
The correct answer here is that some engineers are scientists and some aren't. Among those engineers who are scientists, there are basic scientists, whose aim it is to understand the principles of engineering, and applied scientists, whose aim it is to understand how our knowlege of engineering interacts with real world problems.
That's the cartoon version, of course, but it should clear up some unnecessary confusion.
I gave the original article a good solid skim. There are some truly interesting things in it, few having much to do with mind reading, even at a superficial level. The authors have presented some interesting evidence concerning how space is represented in the medial temporal lobe. The mind reading, such as it is, is not a parlor trick to make headlines. It's a demonstration that the activity in a given region is sufficient to predict a specific mental state, namely knowledge of spatial location.
Of course, just about every BOLD fMRI study ever carried out could be called an exercise in mind reading. Does this study get us any closer to the science fiction version? Well, it's a nice application of multivariate analysis methods, i.e., methods that allow us to investigate relationships between some cognitive state of interest and the pattern of neural activity over a larger area of the brain (not just a single "voxel"). These methods are not exactly new, but they do have a bit of a mind-reading flavor to them, in that they can allow us to extract information about mental states from their distributed representations in the brain. So maybe I have to put aside my knee-jerk reaction and concede that there is a little something in there that you could call mind reading if you wanted to.
I've read a lot of comments this way and that about the differences between high-quality MP3s and CD-quality audio are, but I've never seen any hard evidence one way or the other. No study is going to settle the issue completely, but it would be nice to have some hard data to work with instead of anecdotes, intutions, and listening tests designed by (a) people who don't know what they're doing; or (b) people with an interest in seeing the results come out a particular way. Can someone post some links to reasonably well run studies?
I find it highly likely the easy availability of timed release would cause some authors who would have released immediately under a public license or into the public domain to use the timed release instead. Consider the simplest case, where one could choose a time delay from the CC license chooser. I bet many people would select it just because they could, just as well over half of people select the NonCommercial option, even though in many cases doing so is counter to what one would hope sharing to accomplish.
Certainly there are both costs and benefits, and we don't really know the balance. I understand your concern about the time release costing the commons a few years of access to something useful, and I'm sure you understand my concern that works may end up with many more years of copyright restrictions than their creators really wanted. I have to say, I suspect the real reason many people choose the noncommercial option is that it really expresses their feelings about what rights they would like to grant (even if those feelings aren't well thought out). Noncommercial clauses have been written into various software licenses for a long time, certainly since long before it was possible to generate one from a web interface. Many people don't like the idea of freeloaders. Removing that option would obviously reduce the amount of stuff licensed under those terms. But I'm not very comfortable with the idea of bolstering the commons by trying to herd people into choosing the licenses I like best.
I'll also add that my original reason for asking about this wasn't because I just thought it would be a great idea to help increase the common good, it was just because in the past I've wanted to use a license like that, but the Founders' Copyright was too much trouble.
1: That's exactly the question I was raising initially. If it were that easy, presumably there would be no reason for the Founders' Copyright. But the team of lawyers at the creative commons felt the Founders' Copyright was necessary, for some reason.
3: Consider music. Music you write now might, in 20 years, still make a nice soundtrack for a school project movie you want to make available publicly, or a viral video. It might make, after a bit of processing, good background music for a free computer game. It might make a nice ring tone, or maybe someone would want to sample or otherwise adapt it for a new composition of some kind. There are a million potential uses for music, most of which are restricted by copyright, and many of them would be quite relevant in 20 years. The music would have much more value to the commons than to the copyright holder. By contrast, code ages much more rapidly, especially the kind of code liable to be released under this kind of license. The kind of code liable to be released into the public domain would generally be valueless to anyone unless released much sooner.
4: I thought the analysis was obvious enough, but I'll be explicit. Consider books. If it's not possible to create a timed-release license, virtually no books will be released into the public domain immediately. Most books will have no special license, which is to say that they will enjoy the almost perpetual copyright protection they do currently. In 20 years, all such books written today will still be under copyright protection, and will not be available for various uses. While many of the authors of such books will still be alive in 20 years, some will be dead, some won't want to bother trying to get the word out about their 20 year old books, etc. If such a license is available and legal, then 20 years down the road it will magically come into effect, and it will already be attached to every existing print or electronic copy.
It seems surpassingly unlikely that the timed release option would be used by someone who would otherwise release their work into the public domain sooner, and that this would somehow discourage them from releasing their work anyway. Four years from now, if you want to release your self-published opera score into the public domain, you're not going to be somehow deterred just because you originally published it with a timed release license on page one instead of a plain copyright notice. But with a timed release license, you'll definitely release it in 20 (or whatever) years. Without it, you most likely won't. It's really hard to imagine a scenario in which a timed release license will decrease the value of the commons, but it's quite easy to see how it might increase value.
These are interesting points, thanks for your input. I have some quick responses. Re: 1, how would you put a piece of work into the public domain n years into the future if not via a license? Re: 2, I think added complexity is only inherently bad if it's for no reason. For example, rocket fuel is more complex than water, but rockets don't go if you put water in them. In this case, the simpler license is certainly better in terms of simplicity, but it's much worse in that it fails to grant rights that I would like to grant, namely the rights to use my work for anything whatsoever beginning in, say, 20 years. Re: 3, software is very different from other kinds of creative works, like books and music. Certainly I agree that this kind of license would make little if any sense for software. Re: 4, does this analysis consider only software, or are books and music considered as well?
Certainly the value of the commons would be greatly increased if works written today would fall into the public domain while still relevant. That horizon is much longer for books and music than for software, but not infinite.
Sure, but that would lose two of the main attractions of the CC licenses: (1) that they've been vetted by actual lawyers who supposedly know what they're doing; and (2) that they're readily available on the CC web site for anyone who wants to use them. If I wrote my own license, it would not encourage other people to use the timed-release license, and I'd leave out some magic legal word that would probably invalidate the whole thing.
I've always wondered why the creative commons doesn't offer a timed-release license, so to speak -- a license that kicks in at a certain future date. For example, instead of "you are hereby granted the right to do x," we might imagine, "you are hereby granted the right to do x on january 1, 2020 and any date thereafter." At one point they had something called the Founders copyright (do they still?), but it required transferring copyright to the creative commons, or some kind of nonsense like that. It seems like it would be quite easy to write this kind of license. Is there some technical legal reason why it can't be done?
I have to argue with this a little, even though I'm sympathetic. While it can be an administrative headache, we're only in the first year of this policy, and most of the stickier issues will get better. Costs of $10-20k/year sound like a lot, but how much is that as a percentage of your budget? That may be less science for you, but it's more science for the many thousands of researchers whose institutions can't afford the massive costs of online access to journals. Also, PLoS claims to be able to waive fees for authors who don't have the money.
In an ideal world, journals like the PLoS journals would proliferate, your library's costs would drop dramatically, and your university would negotiate a slightly lower F&A rate, putting more money in the pool for other awards. I realize there are some pretty weak links in that chain, but it's worth considering that just because the fees to access pdfs of journal articles aren't coming out of your direct costs, that doesn't mean you're not, at some level, paying the price. If your library recoups some costs and spends the money on solid gold bookcases, that's a different issue. It's also worth considering that when an article can potentially benefit many thousands of researchers, it's probably a better model to inconvenience the authors than to inconvenience or exclude a substantial minority of the readers. All the more so when it's part of a process that will eventually cut out some middlemen that many of us see as abusive.
Slashdot Mirror
Thanks for posting that, I didn't even realize ZFS was in BSD. On a scale from "rock solid for critical production use" to "for data you don't care about anyway," how stable is it in the latest FreeBSD release? Not that OpenSolaris has been without glitches, but I'm always nervous when contemplating big changes to our storage.
zfs
I often print things out so that I can read them on the run. A decent e-reader would eliminate that need entirely, but tragically no one makes one yet. I have high hopes that this will be fixed sometime this year or next. That's not the only obstacle to a paperless office, but that's by far the #1 paper consumer for me.
Misuse of statistics is well-represented in scientific articles. Other things that are well-represented are poor knowledge and reasoning in the area of the subject discipline, inept writing, misleading or unhelpful graphics, poor scholarship, etc. Sturgeon's Law applies across the board.
Having read a fair number of sky-is-falling articles about statistics in science, and having worked with my share of researchers (MDs and PhDs in a variety of fields) who think everything is rosy, I'm pretty sure that the truth is somewhere in between. A minor saving grace is the fact that getting the statistics wrong is not the same as getting the answer wrong. Although it's certainly quite common to find published articles that make claims with no support whatsoever, in my experience it's much more common to find articles where the inappropriate statistics just mean the support isn't nearly as strong as claimed. Spurious results tend, though not as reliably as we'd like, to get weeded out by the literature. I rarely read an article that isn't specifically about methodology in which the methods/statistics are really solid, but I also rarely read an article in which unsound statistics undermines the entire contribution.
Re: the licensing issues, I was really just asking if there's any chance Sun might consider re-licensing the ZFS code. Last I checked, this was a non-starter, but a lot has changed in the past year or so. That said, I forgot about the "rampant layering violations," and I can see how this would be a permanent problem even if the licensing issues could be worked out. I'm still struggling, though, because I hate maintaining an OpenSolaris box solely for ZFS. At the same time, I love ZFS, and it's going to be a while before I can feel equally comfortable with btrfs.
I'm surprised you've found Linux filesystems such comparatively poor performers, but that's good information to have. I live in a somewhat lower performance neighborhood, and would not have noticed this.
So far as I can tell, zfs is the only piece of opensolaris that's exciting enough to make anyone want to install if if they'd otherwise want to install a linux distribution. With that in mind, could someone post an authoritative update on the supposedly intractable licensing issues that prevent ZFS from being incorporated into the linux kernel? Is it still hopeless?
For what it's worth, you can comment on the proposed legislation here:
http://www.regulations.gov/search/Regs/home.html#documentDetail?R=0900006480a75fb2
Of course, do your research first.
I'm sure this device has a lot going for it, but for my reading enjoyment, I'm still waiting for devices with better screens from qualcomm/mirasol or pixel qi. In nicer packages, I hope. I don't need two flawed screens hinged together, I need a single screen that's more functional than e-ink.
Last I checked, a few months ago, there didn't seem to be any simple options for running skype on a 64-bit GNU/Linux machine.
It sounds like you're assuming the charger will exactly max out a dedicated residential circuit. I don't honestly know how much current these things draw, but I'll bet it's a lot less than 20 amps. If your assumptions are correct, then the quickcharger would be extremely impractical in residential settings.
There is no such thing as the "correct" viewing distance. I like to watch at such a distance that the screen takes up about 20 degrees of visual angle. That makes me comfortable, even though I know a lot of people enjoy 30 degrees, and some of them consider me "insane." I don't watch movies to demonstrate my sanity, I watch them for enjoyment. It doesn't matter to me even a tiny bit that I might be missing out on some of the resolution of the image. If some new video standard had a zillion times the resolution of blu-ray, I wouldn't sit with my face pressed up against the tv.
Recent versions of gcc support OpenMP, and there's now experimental support for a multithreading library that I gather is going to be in the next c++ standard. These don't solve everyone's problems, but certainly it's getting easier, not harder, to take better advantage of multi-processor multi-core systems. I recently test retrofit some of my own code with OpenMP, and it was ridiculously easy. Five years ago it would have been a much more irritating process. I realize not everyone develops in c/c++, nor does everyone use a compiler that supports OpenMP. But I doubt it's actually getting harder, probably just the rate at which it's getting easier is not the same for everyone.
This is a little misleading. fMRI can't tell the difference between blood flow associated with excitatory and inhibitory neural activity, and quite likely many or all BOLD responses involve some combination of both. That does not, however, mean that fMRI can't tell the difference between any activity and the willful inhibition of that activity. That may be true in some cases, but it's not true in general.
It's also not true that fMRI can't in principle tell the difference between your reaction to your wife and some attractive celebrity. Even if it were true that the BOLD signal associated with active inhibition of some process were equivalent to the BOLD signal associated with that same process, this would still be a perfectly reasonable comparison, even if we can't tell the difference between involuntary responses and responses meant to mislead the system. If this weren't the true, then none of the most basic uses of fMRI would work at all. E.g., you can tell the difference between finger tapping and not finger tapping. I don't know about willful inhibition of finger tapping specifically.
I agree that the use of fMRI as a lie detector is fruitless, but it's more an issue of practicality and devising a useful contrast than some in-princple reason why we can't tell the difference between a mental state and its opposite.
I do work in fMRI and I've read many hundreds of articles on the subject, so I feel pretty comfortable with the above. Others working in fMRI might disagree, but I don't think anyone in the field would say that these issues are settled.
I'm with the OP on this. As the father of a 5 and a 3 year old, I know from experience that I can't tell the 3 year old to pick up the blocks, but I can point to a block, tell him to pick up *that* block, and then point to the bucket and tell him to put the block in the bucket. Then I can repeat the process x times where x equals the number of blocks on the floor.
So as the OP said, if these eggheads would just have kids, they would know the outcome of their "research" through experience and intuition.
Maybe the eggheads have learned to base their research on more than just a single parent's experiences with two kids. I know from experience that 3 year olds can learn to pick up blocks without going one-by-one, and do many other more complex things, given the right mood or the right motivation. That's different from your experience, I guess. But I don't know if my experience is typical.
Concerning behavioral research (my career has nothing particular to do with this, I'm a molecular biologist concerned with early brain development), there are repeatedly studies presented here on /. with crappy conclusions (and good commentaries from the /. crowd).
When you say "crappy conclusions," do you mean conclusions that are poorly supported by the evidence, or conclusions that are well supported by the evidence but not that interesting, or what?
I've rarely seen much in the way of good commentaries on science from the crowd here. Sometimes a few people who actually know the area in question will post some insightful comments. But the rest generally respond to the popular report that's been posted, which almost invariably misses the point of the research. I'd be curious to hear what most Slashdotters think the take-home message was meant to be from the study under discussion here.
As a behavioral researcher, I certainly sympathize with your contempt for the field. But at the same time, I think there are good reasons to ask scientific questions that involve behavior (even more so neurology). The fact that some people do it poorly doesn't change that, nor does the fact that some people prefer an extreme reductionist approach.
No, they're not. They "merely" apply science to specific well-known problems.
Absolutely not! If we went by your standard of evidence, we would consider there to be a mountain of evidence that the Sun goes around the Earth. Nowadays it's easy to see that it's the other way round, but if we went by your standard of evidence it's doubtful that our collective scientific knowledge would actually have gotten far enough to discover that.
You're no scientist and have no idea what scientists actually do.
No, they're not. They "merely" apply science to specific well-known problems.
The correct answer here is that some engineers are scientists and some aren't. Among those engineers who are scientists, there are basic scientists, whose aim it is to understand the principles of engineering, and applied scientists, whose aim it is to understand how our knowlege of engineering interacts with real world problems.
That's the cartoon version, of course, but it should clear up some unnecessary confusion.
I gave the original article a good solid skim. There are some truly interesting things in it, few having much to do with mind reading, even at a superficial level. The authors have presented some interesting evidence concerning how space is represented in the medial temporal lobe. The mind reading, such as it is, is not a parlor trick to make headlines. It's a demonstration that the activity in a given region is sufficient to predict a specific mental state, namely knowledge of spatial location.
Of course, just about every BOLD fMRI study ever carried out could be called an exercise in mind reading. Does this study get us any closer to the science fiction version? Well, it's a nice application of multivariate analysis methods, i.e., methods that allow us to investigate relationships between some cognitive state of interest and the pattern of neural activity over a larger area of the brain (not just a single "voxel"). These methods are not exactly new, but they do have a bit of a mind-reading flavor to them, in that they can allow us to extract information about mental states from their distributed representations in the brain. So maybe I have to put aside my knee-jerk reaction and concede that there is a little something in there that you could call mind reading if you wanted to.
I've read a lot of comments this way and that about the differences between high-quality MP3s and CD-quality audio are, but I've never seen any hard evidence one way or the other. No study is going to settle the issue completely, but it would be nice to have some hard data to work with instead of anecdotes, intutions, and listening tests designed by (a) people who don't know what they're doing; or (b) people with an interest in seeing the results come out a particular way. Can someone post some links to reasonably well run studies?
I find it highly likely the easy availability of timed release would cause some authors who would have released immediately under a public license or into the public domain to use the timed release instead. Consider the simplest case, where one could choose a time delay from the CC license chooser. I bet many people would select it just because they could, just as well over half of people select the NonCommercial option, even though in many cases doing so is counter to what one would hope sharing to accomplish.
Certainly there are both costs and benefits, and we don't really know the balance. I understand your concern about the time release costing the commons a few years of access to something useful, and I'm sure you understand my concern that works may end up with many more years of copyright restrictions than their creators really wanted. I have to say, I suspect the real reason many people choose the noncommercial option is that it really expresses their feelings about what rights they would like to grant (even if those feelings aren't well thought out). Noncommercial clauses have been written into various software licenses for a long time, certainly since long before it was possible to generate one from a web interface. Many people don't like the idea of freeloaders. Removing that option would obviously reduce the amount of stuff licensed under those terms. But I'm not very comfortable with the idea of bolstering the commons by trying to herd people into choosing the licenses I like best.
I'll also add that my original reason for asking about this wasn't because I just thought it would be a great idea to help increase the common good, it was just because in the past I've wanted to use a license like that, but the Founders' Copyright was too much trouble.
1: That's exactly the question I was raising initially. If it were that easy, presumably there would be no reason for the Founders' Copyright. But the team of lawyers at the creative commons felt the Founders' Copyright was necessary, for some reason.
3: Consider music. Music you write now might, in 20 years, still make a nice soundtrack for a school project movie you want to make available publicly, or a viral video. It might make, after a bit of processing, good background music for a free computer game. It might make a nice ring tone, or maybe someone would want to sample or otherwise adapt it for a new composition of some kind. There are a million potential uses for music, most of which are restricted by copyright, and many of them would be quite relevant in 20 years. The music would have much more value to the commons than to the copyright holder. By contrast, code ages much more rapidly, especially the kind of code liable to be released under this kind of license. The kind of code liable to be released into the public domain would generally be valueless to anyone unless released much sooner.
4: I thought the analysis was obvious enough, but I'll be explicit. Consider books. If it's not possible to create a timed-release license, virtually no books will be released into the public domain immediately. Most books will have no special license, which is to say that they will enjoy the almost perpetual copyright protection they do currently. In 20 years, all such books written today will still be under copyright protection, and will not be available for various uses. While many of the authors of such books will still be alive in 20 years, some will be dead, some won't want to bother trying to get the word out about their 20 year old books, etc. If such a license is available and legal, then 20 years down the road it will magically come into effect, and it will already be attached to every existing print or electronic copy.
It seems surpassingly unlikely that the timed release option would be used by someone who would otherwise release their work into the public domain sooner, and that this would somehow discourage them from releasing their work anyway. Four years from now, if you want to release your self-published opera score into the public domain, you're not going to be somehow deterred just because you originally published it with a timed release license on page one instead of a plain copyright notice. But with a timed release license, you'll definitely release it in 20 (or whatever) years. Without it, you most likely won't. It's really hard to imagine a scenario in which a timed release license will decrease the value of the commons, but it's quite easy to see how it might increase value.
These are interesting points, thanks for your input. I have some quick responses. Re: 1, how would you put a piece of work into the public domain n years into the future if not via a license? Re: 2, I think added complexity is only inherently bad if it's for no reason. For example, rocket fuel is more complex than water, but rockets don't go if you put water in them. In this case, the simpler license is certainly better in terms of simplicity, but it's much worse in that it fails to grant rights that I would like to grant, namely the rights to use my work for anything whatsoever beginning in, say, 20 years. Re: 3, software is very different from other kinds of creative works, like books and music. Certainly I agree that this kind of license would make little if any sense for software. Re: 4, does this analysis consider only software, or are books and music considered as well?
Certainly the value of the commons would be greatly increased if works written today would fall into the public domain while still relevant. That horizon is much longer for books and music than for software, but not infinite.
Right, that's the one I meant, thanks for the link. You can see it's a tremendous hassle compared to slapping a notice on your work.
Sure, but that would lose two of the main attractions of the CC licenses: (1) that they've been vetted by actual lawyers who supposedly know what they're doing; and (2) that they're readily available on the CC web site for anyone who wants to use them. If I wrote my own license, it would not encourage other people to use the timed-release license, and I'd leave out some magic legal word that would probably invalidate the whole thing.
I've always wondered why the creative commons doesn't offer a timed-release license, so to speak -- a license that kicks in at a certain future date. For example, instead of "you are hereby granted the right to do x," we might imagine, "you are hereby granted the right to do x on january 1, 2020 and any date thereafter." At one point they had something called the Founders copyright (do they still?), but it required transferring copyright to the creative commons, or some kind of nonsense like that. It seems like it would be quite easy to write this kind of license. Is there some technical legal reason why it can't be done?
I have to argue with this a little, even though I'm sympathetic. While it can be an administrative headache, we're only in the first year of this policy, and most of the stickier issues will get better. Costs of $10-20k/year sound like a lot, but how much is that as a percentage of your budget? That may be less science for you, but it's more science for the many thousands of researchers whose institutions can't afford the massive costs of online access to journals. Also, PLoS claims to be able to waive fees for authors who don't have the money.
In an ideal world, journals like the PLoS journals would proliferate, your library's costs would drop dramatically, and your university would negotiate a slightly lower F&A rate, putting more money in the pool for other awards. I realize there are some pretty weak links in that chain, but it's worth considering that just because the fees to access pdfs of journal articles aren't coming out of your direct costs, that doesn't mean you're not, at some level, paying the price. If your library recoups some costs and spends the money on solid gold bookcases, that's a different issue. It's also worth considering that when an article can potentially benefit many thousands of researchers, it's probably a better model to inconvenience the authors than to inconvenience or exclude a substantial minority of the readers. All the more so when it's part of a process that will eventually cut out some middlemen that many of us see as abusive.