If this is not the correct forum for this question, or too specific, feel free to close it. I'm more or less from MSE, and this seemed like the place to ask.

There's this well-known math text called "Characteristic Classes" by Milnor and Stasheff which is regarded as a staple among topology students in the field. It's a classic, and is highly regarded by everyone. The problem is that it's old, comparatively speaking. I don't seem to be able to hunt down a 'new' copy in the sense that it's updated printing, font, etc, and it's kind of an eyesore to read. For me personally, this writing actually causes headaches, to the point where I can't read for much longer once it starts. The copyright date inside the text is from 1974.

I don't know anything about copyright law at all - I'm just a graduate student. Since I have time this summer, I was thinking about trying to transcribe some sections of the book into LaTeX, with modern diagrams and fonts that might be a little easier on the eyes. It would probably involve some headaches along the way, but I can't be the only one who experiences this, and it might help someone else down the line. Plus this would let me insert in footnotes or endnote comments about things that jumped out at me while reading. It's a rich text so there are lots of places where this happens.

Obviously if I do this for myself and don't share it with anyone, there's no problem. But if I were to do this, it would really be for the benefit of other grad students or mathematicians. If I were to post it on a blog or something, it might violate some kind of copyright law, and obviously I want to avoid that kind of problem. Is this sort of thing legal, and more generally, how do I go about finding out what is or is not legal in the United States?

  • 8
    Thanks for proposing to add to the body of legible and searchable text :) seems very scholarly
    – Clumsy cat
    Commented May 25, 2017 at 15:24
  • Might I suggest WorkBooks as a potential place to get involved? I haven't looked to see if they're making progress, but they certainly share a desire for copyright free and searchable text. Perhaps you can instead improve upon what they are doing? Commented May 26, 2017 at 12:50
  • 3
    Publishing someone else's work without permission violates their copyright. Commented May 26, 2017 at 15:15
  • I would submit that being a graduate student is not an excuse for not knowing about copyright law - instead, copyright law is one more aspect of the things you should be learning, and it should be treated with the same due diligence as all other soft skills. This is not to bash you - this is you doing (part of) your due diligence correctly, right here - but "I didn't know it was illegal" is never an excuse.
    – E.P.
    Commented May 28, 2017 at 13:28
  • related: academia.stackexchange.com/questions/15783/… Commented May 28, 2017 at 14:49

4 Answers 4


I would encourage you to contact the authors. Both John Milnor and James Stasheff have their contact info on the web. As Stasheff is emeritus, Milnor might be a better bet.

  • 16
    @AlfredYerger Well, it could still be complicated. Perhaps the authors don't actually own the copyright in the book; it may lie with the publisher, and in that case the publisher would have to be the one to grant permission for the kind of thing you want to do. (I doubt they will.)
    – David Z
    Commented May 25, 2017 at 20:27
  • 2
    I think this is bad advice because of @DavePhD's answer which points out that the copyright is owned by the publisher. So, even if the authors were OK with the proposed derivative work, they couldn't give permission. Or, more likely, they could give permission, and later the OP would discover that the publisher has a problem with it and that it is their opinion which matters... (Yes, it sucks, and this is one of the reasons why I think it's a bad idea to give away one's rights to a publisher.)
    – a3nm
    Commented May 26, 2017 at 15:12
  • 1
    Milnor and Stasheff's text was published 43 years ago by Princeton University Press and a lot has happened since then in the world of typesetting. My guess is that the authors would be the right people to persuade the publisher that it was time for a new edition. Commented May 26, 2017 at 15:24
  • 1
    The author would be in touch with the responsible people at the publisher. They could more authoritatively convey the notion that there is ongoing interest in the text, and desire among those interested for a more modern edition. I would look at this as an opportunity to make a deal with the publisher to prepare a newer edition. Depending on how much the OP thinks he can add content-wise, beyond improved design, it may even be worth co-authorship credit. For instance, if the material can be updated to reflect newer results in the field, cover proof techniques that were unknown at the time, etc Commented May 26, 2017 at 17:15
  • 1
    This isn't a isolated book, it is part of a series. Annals of Mathematics Studies, volume 76. press.princeton.edu/math/series/amnum.html Contacting an author is insufficient.
    – DavePhD
    Commented May 26, 2017 at 17:34

The copyright holder is Princeton University Press.

Princeton University Press says:

Permission to Reprint a Complete Book

A full reprint of a Princeton University Press title requires a license from the Press or other copyright holder. For English language licenses, please contact [email protected].

  • 3
    Feel free to edit in: It is unlikely that Princeton University Press will grant permission for you to digitize the book for free, seeing as they have made an eBook available for sale.
    – shoover
    Commented May 27, 2017 at 4:49
  • Copyright can revert to the authors in case the book goes out of print, depending on the copyright transfer agreement. However, given that the eBook is available from the publisher, this is highly unlikely to apply.
    – E.P.
    Commented May 28, 2017 at 13:22

A very rough rule of thumb for US copyright law is that everything published in 1923 or later is protected by copyright. That means you can't make a copy without the permission of the copyright holder. Making a copy is precisely what you want to do. There are certain "fair use" exceptions, such as for quoting short passages, but they don't appear to apply here.

You'd have to get the permission of the copyright holder; that might be the authors, or their publisher, or some other persons to whom the copyright might have been transferred. They could refuse permission, or demand an arbitrary amount of money; that would be their right.

A nice general overview of US copyright law is at http://www.bitlaw.com/copyright/. It's a good thing for anyone in the US to become familiar with, especially an academic who is likely to publish copyrighted works. See also https://en.wikipedia.org/wiki/Public_domain_in_the_United_States for a discussion of "public domain" which refers to works which are not copyrighted or whose copyright has expired.

  • 9
    Material which was published before the 1940s would have had its copyright expire 28 years after publication unless it was renewed. Since some material published in the 1920s and 1930s had its copyright renewed and some didn't, some such material is in the public domain, while some will be copyrighted indefinitely.
    – supercat
    Commented May 25, 2017 at 18:18
  • @supercat Material published before 1964 had to have their copyright renewed, but the URAA restored copyright to most non-US works. Works published after 1922 will leave copyright 95 years from publication; it's not impossible that that will change, though I would be stunned if this Congress changed the law before 2019, when 1923 works go into the public domain, but "copyright indefinitely" is clearly unconstitutional, and there's no reason to assume that it will be changed.
    – prosfilaes
    Commented May 28, 2017 at 8:42
  • @prosfilaes: Before the extension of copyrights from 75 years to 95 years, it may have been reasonable to expect that works' copyrights would expire 75 years after publication. Is there any more reason to believe that works will expire at age 95 than there would have been to expect expiration at age 75? Personally, I think copyrights should be much shorter by default, but renewable for longer if someone pays an escalating renewal fee. As long as an author is continuing to make use of a work, I have no particular problem letting the author keep exclusive rights to it. My objection...
    – supercat
    Commented May 28, 2017 at 21:46
  • ...is to the vastly more common situation where the creator of a work essentially abandons it after 10-20 years, but other people who might benefit from being able to use it, cannot.
    – supercat
    Commented May 28, 2017 at 21:49
  • @supercat The last copyright extension, that brought us more inline with Europe, was not unexpected. Works did expire under the publication + 75 years standard, and it seems reasonable to assume that if there's another copyright extension, it won't be rushed through in time to stop early 1920s works from falling to the public domain. Given the post-legislative response to the last copyright extension act, it's likely that it won't go through silently. If you object to longer copyright terms, don't treat them as inevitable, plan to fight like hell if Congress proposes them.
    – prosfilaes
    Commented May 29, 2017 at 23:18

No reason you couldn't learn a bit about copyright law yourself. You can actually get a decent understanding of the basics by reading the law itself, title 17 of the United States Code. I'm not a lawyer, but I've read a good chunk of this stuff. I'd point you to these sections to learn the basics:

  • Sections 102 and 103 describe what copyright applies to: any original work "fixed in a tangible medium" (e.g. written down, recorded, stored on a hard drive), as well as compilations and derivative works (i.e. modified versions of someone else's work), but only to the added value you created by compiling or modifying the original. Copyright does not apply to the ideas contained within a work. Also, section 105 says that official works of the US government are not covered under copyright (though there are subtleties in what is considered an official work of the government).
  • Section 106 describes what activities are controlled by copyright law. There are six:

    • reproduction
    • creation of derivative works
    • distribution to the public
    • public performances
    • public display
    • public broadcasting, in the case of sound recordings

    When non-lawyers talk about "copyright", they are typically talking about the right to undertake these activities. To say that someone "owns (or holds) copyright in a work" means that they may reproduce, modify, distribute, perform, display, or broadcast that work, that they may authorize others to do so, and furthermore that other people are not allowed to do these things without authorization from the copyright holder - although there are some exceptions.

  • Probably the most relevant exception for academics is fair use, described by section 107. Fair use means that, under certain conditions, people can do the things that would normally be prohibited under section 106 even without authorization from the copyright owner. The catch is that just what those conditions are has to be decided by a court on a case-to-case basis. The law instructs judges to consider four factors when deciding:

    • "purpose and character" of the use: commercial? educational? for research, criticism, parody? etc.
    • nature of the work, e.g. how complex it is
    • how much of the work is used
    • the effect of the reuse on the market for the original

    All four factors matter, so you couldn't just say "I'm doing this for educational purposes, clearly it's fair use".

  • Section 201 says that the authors of a work own the copyright when the work is created, or if it's a "work for hire", whoever did the "hiring" owns the copyright. They can transfer it to others, or transfer only some of the six protected rights, or do so only under certain conditions, or so on.
  • Sections 302, 303, and 304 describe how long a work is covered under copyright law, i.e. how long people are prohibited from doing the six things without permission. It's an absurdly long time: for works published after 1978, the life of the author plus 70 years, or for works published before 1978, 28 years from publication with the option for renewal for another 67 years. (In your case, I think it's likely the copyright was renewed, so the book is probably still protected.)

There's a lot more to the law but I think most of it is less relevant in your case.

Obviously if I do this for myself and don't share it with anyone, there's no problem.

There's "no problem" in the sense that nobody would ever find out. (Or if they did, let's be honest, 99% of people wouldn't care.) But technically, I think even that might not be legal. After all, I'm pretty sure what you're doing is preparing a derivative work, which is one of the protected rights, even if you never share it with anyone. It might fall under fair use but that would have to go to a court to decide - again, if anyone ever cared enough to go after you for it, which they (almost certainly) wouldn't.

But if I were to do this, it would really be for the benefit of other grad students or mathematicians.

This could easily be irrelevant. It would probably be considered in a decision on whether your project constitutes fair use of the original book, but other than that, copyright law doesn't care who the work benefits.

If I were to post it on a blog or something, it might violate some kind of copyright law

I suppose technically it's the preparation of the derivative work that violates the law, and I'm not sure if distributing the work counts as a whole separate violation on top of that, but the important factor is that putting this on a blog makes it public, which means your copyright violation is in plain view of anyone who would want to hassle you about it. This hassling would probably take the form of a DMCA notice - something like a cease-and-desist letter for suspected copyright infringement - sent to your hosting provider, who would either pass it on to you along with a demand to remove the content (if they're "cool") or would just remove it themselves without warning (if they're not). Now, I'm not going to go in detail about the DMCA here, as this post is getting long enough already, but you should know that these notices are fairly cheap and easy to send, they do carry legal weight but there is a process for challenging them, and if the issue isn't resolved by that point, then you might well wind up in court.

  • 2
    In practice, if you post it on a blog, the most likely outcome (other than "nothing happens because nobody cares") is a DMCA notice sent to your web host, not a lawsuit filed against you personally. Such a lawsuit is always possible, of course, but the incentives are stacked in favor of a notice because lawsuits are expensive and slow.
    – Kevin
    Commented May 26, 2017 at 2:01
  • 1
    @Kevin True, after writing all this I forgot to mention the DMCA. I'll edit accordingly.
    – David Z
    Commented May 26, 2017 at 3:46
  • 4
    One should be very careful about attempting to understand the law by reading legislation. People without legal training are apt to misunderstand legal texts, since common English words very often have very specific meanings. Commented May 26, 2017 at 12:24
  • 2
    @DavidRicherby Recent decisions with similar facts should be studied, like Cambridge University Press v. Mark P. Becker. copyright.gov/fair-use/summaries/… Especially the four fair use factors.
    – DavePhD
    Commented May 26, 2017 at 13:45

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .