Note: this is only one page in a larger set of pages on this wiki that talk about copyright. See also:
With a large project like PlanetMath that actively accepts contributions from users around the globe, copyright concerns are bound to arise. Especially this one: are the contributors actually contributing works that they have a right to contribute?
The same issue applies to GNU and to Wikipedia and other large projects (both free and non-free) and historically this issue has been resolved in different ways, depending on the size and nature of the project in question. Project Gutenberg must do little beyond being certain of the publication date for books that contributors would like to add, but that is a particularly simple case.
One of the key copyright concerns for most CBPP projects is that some contributors may not understand the relevant parts of copyright law. Copyright law can be confusing, and even smart people whose backgrounds lie elsewhere will not necessarily be familiar with some important facts about copyright, like the difference between Ideas and Expression.
We might blame citizen education! But rather than pointing fingers, on this page we're going to try to actually do something about the problem, by writing some about what you need to know about copyright to be a good free CBPP contributor. We will also discuss the steps that a project like PlanetMath can take to limit its liability, while continuing to be a productive and socially responsible organization.
Throughout the discussion presented here, our rhetorical focus will shift back and forth between PM and some general "theoretical" CBPP system. The concerns we speak of are generally applicable, but they are made more immediate and concrete in the case of PM. You may find these rhetorical shifts of perspective somewhat disorienting. If the condition persists, you should speak with your physician.
The number one copyright concern is that a user of a CBPP system may "contribute" work that they don't own the copyright to. This might happen if the user was ignorant, incautious, or even malevolent.
The possible negative results of copyright infringements in CBPP are various. First, an ignorant user could get him or herself into trouble - there are always risks associated with ignorance! Second, the maintainers of the CBPP system (and de facto publishers of the infringing work) could get in trouble - there are always risks associated with being "open". Third, in the case of free content, persons making derivative works under the terms of the license could get in trouble or end up wasting time by working with something that they later learn is not really legal to work with - there are always risks associated with copying.
I wish to emphasize that this complex of risks is never going to go away. The best thing we can do is work to minimize the impact of problem situations that may arise.
Our concerns about the copyright status of contributed content will remain nebulous until we have some documentation of the process used to create the content. As the most basic example, unless users cite the references they used when working on a given article, looking for works that may have been used inappropriately would be somewhat like looking for a needle in a haystack. If an article cites the references that were used, these references could, at least in theory, be checked for examples of outright copying and the like. The more detail that a user provides about their process and methodology, the easier it will be for an independent content auditor to check the copyright status of the work.
Unless we have a clear understanding of what makes a document legally OK, even good process documentation will not be that helpful. We can make guesses and conjectures, and say "well, it looks like it is OK" when it does – however, without guidelines for an auditor, he or she will find it harder to make good judgements. Since responsible authors can be expected to apply the same sort of procedures to their own work (both during and after composition), clear guidelines for "copyright cleanliness" would also help authors avoid infringements in the first place.
Finally, unless we have a clear understanding of the legal responsibilities of the parties involved, we will find it harder to respond to or prevent problems. Assuming an possible infringement is identified, how should the CBPP system respond? In the unlikely event of a lawsuit, who would be implicated?
We would hope that users have so far only contributed content that they could contribute rightfully. Certain instances of copying that have been assumed to be legal should be checked, and to the extent it is possible, the process that led to the creation of all of the (purportedly) original articles should be documented and checked. Until these processes are complete, our understanding of the copyright status of the encyclopedia probably really is largely a matter of faith. We can do better than this, and we should.
Luckily for us, even if infringements were considerably more widespread than we would guess, it seems unlikely at that a lawsuit would be filed by anyone at this stage in the game. As PM grows, our beginner's luck in this regard may fade. Indeed, the more comprehensive the site becomes and the greater the range of services it provides, the more eyes will be on our work, and the greater chance there will be of legal difficulties if we don't keep our legal apparatus growing in scale with the rest of the site.
Thus, the concerns should be thought of as long-term but of growing importance. In the long term, copyright problems could completely destabilize the site. By contrast, good solutions to our concerns will help make the site a really great place.
So, there you have it, the tell-tale mixture of danger and opportunity. We should proceed with caution.
Add info here, but basically:
Add info here, but basically:
The (draft) card-based temporary rating system could contain some useful ideas to help with the mechanics of checking new contributions, and also for dealing with problem articles found in the current collection. Since addressing copyright concerns is a big priority for PM, edits to that page that enhance its usefulness in this area are particularly welcome. Feel free to discuss any related ideas here or on that page!
Phone companies are not responsible for illegal activities conducted by telephone. However, if they evesdropped on conversations and reported some conversations to the police, they might end up being held responsible for all conversations (including ones that they didn't listen in on). By adopting a completely hands-off policy, and letting anyone at all use their services, the phone companies secure a position in which they are not liable for activities carried out using their services.
It would be good if PM could secure a position like this vis a vis the articles that are published on PM. I don't know enough about the situation to be sure that we could successfully make the arrangements… but if we could, it would be pretty swell. Anyway, it is something to add to the discussion.
Discussion about common carriers
The major reason why the issue of copyright infringement is discussed here at all is that most sources of mathematical information are decidedly not free — this means not simply that they are protected by copyright, but that publishers and authors have chosen to reserve all or most of the rights which they can reserve under the law. To understand the difference between copyrighted vs. non-free, consider the example of the GNU license. A work released under the GNU license is not in the public domain. It is copyrighted every bit as much as a book published by a commercial firm. The difference lies in the fact that that, while the commercial publisher does not allow anyone else the right to reproduce the book, the author who releases a work under GNU license has granted everyone permission to reproduce the book and certain other rights provided certain conditions (such as reproducing the GNU license when copying the work) are met. In fact, copyright law is what allows the free license to work — if one simply relinquished all rights to the work and put it in the public domain, then there would be no grounds for requiring that the conditions of use be met!
One day, we hope the situation will be different and one will be able to obtain all one's mathematical information from free sources or, at the very least, sources with relatively liberal license agreements, but now the situation is quite diffrerent. Publishers require that authors surrender their rights to books and articles to the publisher and have jealously guarded their copyrights, even suing authors whose books they were printing.
Given this situation, one is forced to rely on information which appears in books which have "all rights reserved" stamped on their title pages. What can one do in such a situation? Waiting for the copyright to expire is not very practical because, by the time a work passes into the public domain, it will be rather dated — whilst one can freely reproduce works of nineteenth century mathematicians, most of twentieth-century mathematics is off-limits. One can always ask for permission, but it is unlikely that a commercial publisher will allow a free encyclopedia permission to do something which may possibly impact on sales of their books. Once a book is out of print, one can ask an author to consider releasing the book into the public domain or rereleasing the book under a free license. However, a math book can stay in print for decades, so this, while certainly worth doing, is not necessarily going to keep one from having to refer to non-free material.
In such a situation, one needs to tread carefully lest one run afoul of the law. In particular, this means that those of us involved in free math needs to become familiar with the doctrine of fair use. Roughly speaking, it allows one to legally do things which would otherwise be considered infringement of copyright. That is the good news — the bad news is that, as even the copyright office itself states, this is a somewhat murky area of the law. There are no clear-cut rules stating how much material may be copied without permission. Rather, the law (section 107 of the copyright law) lays out four broad guidelines which the court is to use in determining whether or not an act constitutes fair use or infringement.
Therefore, it should be imperative that those of us involved in free math take some time to familiarize ourselves with these guidelines and their interpretation as established by the courts. Likewise, we need to be careful to avoid lulling ourselves into false security with popular misconceptions about copyright. One can find a list of some such mistakes at the following website: [1]
Among the misconceptions, the following way be worth pointing out as being of relevance to free math projects:
To finish on a more positive note, it might be worth remebering what the law does not and can not protect. Section 102(b) of the copyright law lays it down rather clearly:
Since the subject matter of mathematics consists of "ideas, procedures, processes, systems, methods of operation, concepts, principles, and discoveries" and mathematical entities are suppsed to be invariant under isomorphism and independent of embedding, it is clear that mathematics is free for all to use according to the law. Likewise, "laws of nature, natural phenomena, and abstract ideas" cannot be patented.[3] To be sure, like all other freedoms, this freedom is subject to erosion (in particular, issues having to do with software copyrights come to mind) and must be defended.
In particular, this means that there is a sure-fire way of being sure that one does not infringe on existing copyrights — only write on material which one has already thought about at length to the point that one could rederive it from scratch and take care to write from one's understanding. However, this may be asking too much, especially from younger contributors, so some sort of guidelines for extracting mathematical content from its textual presentation and cautions to keep away from potential legal difficulties may be in order.
Finally, there is something of a paradox in this — while mathematics is unquestionably free in principle, in practise it is inextricably bound with its verbal expression, which is subject to the strictest legal protection. It is this disparity which the free math movement seeks to remedy by providing sources of mathematical knowledge which are as free as the ideas they embody.
Note that going beyond the particular set of ideas, facts, etc., that you find in a source is not in and of itself enough to solidify your case under fair use. The Ideas and Expression page discusses a case where it seems that the core ideas were used again, and significantly expanded upon, but were also used together with some of the expression from the original work. The case hasn't been decided, but the plaintiff certainly seems to have a point. And, ironically, it may be the very way in which the defendants "went beyond" that gets them into trouble. The claim is that they expanded upon an unpublished manuscript, retaining many of its core features while changing many others, and adding plenty of new things. Still, the core is apparently there – and this is a core of expression, not a core of facts.
In the math world, similarly, someone basing an article on someone else's expression might get themselves in trouble. True, basing one or two paragraphs on one or two paragraphs from a book with hundreds of pages is not likely to be the same as incorporating the core of the book's expression. (I assume that Fair Use is somewhat forgiving of minor trespasses like this? - maybe I'm wrong). But if this same activity was carried out systematically throughout the pages of the book in question, then trouble would be almost certain.
And even though fair use makes individual facts free, in general, even a bare collection of facts has some "expression" to it, simply by dint of those particular facts being collected together. Thus even the driest work (a logarithm table, for example) is afforded some protection under copyright law.
Just as an author of a screenplay who bases his or her work on the work of a previous author must be careful not to have "too much" of the expression in common, an author of a new mathematical guidebook that includes logarithm tables must be careful not to have "too much" of someone else's logarithm table.
But "too much" here means too many rows – if you can compute even one more column for all the rows you might be OK, because then it is clear that you're computing things for yourself. (I actually don't know if this statement is mathematically correct, since maybe there are tricks you can use to compute more decimal places for your logarithms, but anyway hopefully you see where the comment is coming from in principle!) Even so, if the actual choice of arguments to the log function (i.e. the pattern behind the rows) is at all "artful" or creative, you might also wish to choose some other pattern.
Basically, copyright protection for the original work says that to the extent that it is reasonable to do so, anything new in the same area shouldn't be a "derivative work" unless it is written by the author of the original work. There is really a continuum of derivativeness, from being an outright copy to being just vaguely inspired by the other work.
Finding some ideas in a source text does not immediately make your work a "derivative". Going beyond the ideas you find in a source text does not immediately make your work not a derivative. The important thing is that your work should be your own – by which I mean, it should contain your expression of the ideas you are writing about, not someone else's. This makes good legal sense, and for a project like PM it also makes good expository/pedagogical sense.
I don't necessarily want to advocate doing this… but… if the project facilitators do create a notice that "makes it clear that illegal behavior is the responsibility of contributors", then I wonder whether the project could prosecute persons who contribute infringing text? It makes sense to me: first of all, it would be a breach of contract, since by accepting the terms of this Contract and submitting text to this site, you guarantee that to the best of your knowledge you have the legal right to post this text, and that you have familiarized yourself with the Copyright Guidelines for CBPP and that you will comply with the requirements listed there. Second of all, there could be additional damages that have accrued to persons making derivative works. These persons could, presumably, press charges on their own against the person who published the work under an illegitimate license. But PM could also add to its contributor's contract something like the Contributor agrees to be held solely responsible for damages caused to PM or other legal entities brought about through negligence in following the Copyright Guidelines. Thus, even if a third party was to sue PM, PM could point to this contract and redirect the lawsuit (or, anyway, turn around and sue the person who contributed the non-legal content). I don't know how effective any of these suits would be, especially if they involve crossing international borders.
I don't suppose that PM would want to prosecute anyone who contributed infringing text right off the bat (unless they were clearly doing it just out of malevolence and a desire to try to get PM in trouble). If no derivative works were made and the original owner of the copyright was oblivious, then its really no skin off PM's back. But we would still want to reserve the right to delete the offending content before it did become a problem – at least, contributors should know that red cards could be used by other site members to delete suspicious contributions and possibly to get the contributor blocked from using the site in the future.
Organizational entities are trying to assert copy rights over public domain materials simply because they have digitized, or otherwise captured or disseminated them. We have heard this called "thin" copyright, and (1) it is not supported by the law, and (2) it has been opposed by courts (see Corel v. Bridgeman).
It is important to pay attention to this issue, because the contemporary sources one might encounter of public domain material may be asserting restrictions which would halt CBPP-style re-use, at least without burdensome release procedures. In addition, there is no statutory guarantee that a content user won't be sued in this situation, so "thin" copyright still poses a real risk for content builders! In other words, if Bridgeman had been facing an organization like PlanetMath instead of Corel, it could have wiped us out financially, even if we won in court (more likely, we probably would have to give up use of the content upon extra-legal threat).
So far we know that Cornell's DL (which contains historical mathematics works) is asserting "thin" copyright over PD materials, and it is likely Google Print will attempt to do the same. Stay tuned.
The authors of AsteroidMeta (and this page in particular) are not attorneys or legal scholars. The material presented here is simply general information presented in order to raise public awareness about issues of copyright and foster discussion of these issues and their impact on commons based peer-production of mathematical content. While every attempt has been made to check that the information presented here is accurate, the opinions and suggestions presented are in no way intended as legal advice; for authoritative and reliable advice, the reader should consult a qualified expert in the field of copyright law.
But I don't see evidence on the 'net that Adobe was actually challenged over this thing, and I think it is quite likely that the guy I was talking to simply didn't know what the hell he was talking about.
Lessig does know what he is talking about, and since he didn't seem to see a problem with Adobe put a license onto the thing in the first place, maybe they did have a copyright.
Maybe time to turn to a different case.
Here's one thing I've found:
And here's more info:
I mean, what could be more confusing?
Oh, here you go! Copyright protection of computer program software.
This thin copyright business seems to me to be a real tar baby of confusion. Maybe time to turn to a local expert or something…
--jcorneli