I complained about that, but it was too late to do anything about it. (You don't usually even see those frontmatter pages during the book-writing/editing/proofing process.)
Why complain? It's a little funny: the latter paragraph ends "... without the prior written permission of the copyright owner and the publisher", which, apparently, the former paragraph grants. No?
I'm not sure this is the case here, but it might be the difference between the actual text itself and the representation, arrangement, and style of the text as exhibited in the book. This happens in music, where a given Mozart concerto, for example, is in the public domain, but that doesn't mean you can photocopy a given sheet music of the concerto and hand it out; the notes themselves are in the public domain, but the particular typesetting of the specific document is not.
This was an issue recently with the Google Book Search deal. Even though the books were out of copyright or orphaned, Google owned the copyright to the scans, and would only give a non-transferable license back to the universities.
Google does have a copyright on the collection as a whole, even as they do not gain any extra rights on the underlying public domain material. It's sort of weird, but it does make sense.
I see the first paragraph as being "written permission" to "copy distribute and/or modify" the document under some quite specific conditions, and the second as reserving all other rights.
I suspect this leaves Apress at the "Require all further authors to contractually promise us, in writing, that they are assigning Apress the exclusive right to publish/distribute the content and have not written any legal landmines which contradict this clause into the sections we routinely don't read".
As the person who commissioned the book at Apress, I can give a little insight into how this came about. I had known Mark for a little while and had seen his (then unfinished) work with DiP. I knew about the license conditions, and spent a good few hours talking with Mark via IRC to convince him to sign up to Apress, let us advance him and essentially pay him to finish it.
I vaguely remember some terms that we agreed which made it favorable for Apress for the first few years, to give apress breathing room to recoup the investment we had made in Mark.
I took this to Gary and the rest of the editorial board, proposed why we should do it and how it could happen, and they agreed. I left not long after, but I understand Mark's book went on to several reprints and it seems new editions - all continuing to make for an excellent book.
I'm proud of bringing the book to Apress, and at the time we felt it was a really good match for a smaller pub to be able to do something a little non traditional and fun. I'm glad to see that it's made lots of people a good deal of money - and I'm sure that Apress don't want to lose the stream from one of their star performers. :)
It is hard though to keep making such unique deals as pubs grow- and, yes Joel - unless you are you, dictating terms isn't really something that most tech authors can do, as the secret is that only a very small percentage of tech books truly sell well. But all it takes is one committed editor who's willing to put their neck on the line for a good product-- and i think dip was one of those. :)
That is a likely response. But that will push authors who disagree with that to other publishers.
I've noticed a number of technical books developed in an open manner (though usually not under the GPL) that then sold well. There are also a lot of good people around software these days who strongly believe in the ideals of free and/or open source software. Therefore I suspect that insisting on strict copyright control will lose more sales than it gains. But people will make mistakes as they learn this.
That is the standard Apress contract now. Mark made his initial contract with Apress quite a long time ago when it was a far smaller company where special deals could be made.
Save for time-travel, they can't really do that when they decide to pick up and "publish" a book that's been freely available in electronic form for eight years.
Or it may drive them to offer much lower royalties on books offered that way -- and perhaps only pay them out after the profitability has been established.
So do Apress pay royalties to Mark currently, on a per unit sold basis? If these new people can publish under the GFDL then presumably Apress can (are?) as well.
>such a contract must specifically remove any rights the GPLD gives Apress
I assume the contract commits Apress to pay royalties to Mark based on their sales of the book's print edition. If they contractually agreed to pay royalties, they would be in breach of contract regardless of the licence under which Mark released his copyrighted content.
Notwithstanding the licence, Apress must have decided that they could make enough money on sales to justify the money they're paying to Mark.
Look at it this way: he might have released the book under a free document licence that another party could exploit, but he arguably wouldn't have written the book if not for his contract with Apress. So if they didn't agree to pay royalties, there would be no book to sell.
Simply because I like when I purchase classics (not covered by copyright any longer) that I have the ability to choose the book based on criteria I cannot usually factor in.
I can opt for quality paper, for a nice and well-spaced font, for good-binding, for a nicer jacket, for a specific translation.
I like that there is the potential that I could purchase this book and be able to choose such things, that I could say "I'll pay this guy over here a little more as he's invested in the quality of the medium.".
I was very happy with the contents of Coders at Work for example, but disappointed with the paper selection and choice of font. I would have happily paid more to have a nicer package, so I love authors who would permit multiple companies to do this.
Just to be clear, translations are themselves protected by copyright. So, for example, if you want to read the Edith Grossman translation of Don Quixote, first published in 2003, you do not have the freedom to select from a wide variety of editions, as described in the parent comment.
This is what free software / free licenses needs, people who parade and exemplify it rather than compare it to non-free licenses or even hate on and fight against non-free software, by force.
I don't entirely understand his explanation though. All I understand is that Mark approves of what happened. I do not understand why it is a good thing and why he wants the world to work like this. He didn't explain that.
The whole logic of what and how you are allowed to charge for in the free software/content world seems completely arbitrary to me, even though I have read so much about it.
I think the point is that once Mark released his work under the GPL, then he no longer has any control on how his work is used or distributed, even though it might be in disagreement with his wishes ... and he accepts that, because that's what "Free" is all about.
... that in contrast to forcing people to abide by your rules, without explicitly saying that in the license (e.g. by activism, harassment and public insults ... for reference see for example GNU/Linux [1] or the endless discussions about apologists/traitors[2] ).
I'm not a lawyer, but my layman's understanding is something like this:
* If you wrote/created/designed it, you get to pick the license. You can give it to some people under the GPL, you can give it to other people under the BSD license, you can negotiate all kinds of crazy licenses for arbitrary recompense you dream up - it's up to you.
* If you received a copy of some GPL'd code, freely or in exchange for money, you have the right to give copies to other people - freely, or in exchange for money. If you paid $1000 for some GPL'd code, you're allowed to burn it to a stack of CD-Rs and hand them out on the street corner. If you got it for free off the Internet, you can copy it to your own website and charge people $29.95 per download (whether these actions are moral is another question, but they're legal).
* If you give somebody some binary executables whose source was under the GPL, whether you gave them in exchange for money or for free, you must give them the source too, or give them the right to ask for the source later. If they ask for a copy of the source later on, you're only allowed to charge the cost of blank media - regardless of what you charged for the binary executables in the first place.
If you wrote/created/designed it, you get to pick the license
I do get that part. But it doesn't help me understand the merits of any particular license.
Thanks for your attempt to explain the GPL to me. I already knew what the GPL says, I just don't see how it achieves its goals of increasing my freedoms.
For instance, I'm not allowed to use the source code of some of the software google has linked to GPL code. Even when I pay for using it (like when I run my code on google's servers), I lose all freedoms related to some of the code covered by the GPL.
If they let me download their code, they would have to let me use the source and sell it as well. But they can simply not let me download it and force me to run the code on their servers exclusively.
So, in this scenario using the GPL means that I can not view or modify the source code of the software I build on. I have to pay for using it, and I have to abide by further arbitrary rules included in their terms of service, taking away even more fundamental freedoms.
The GPL increases your freedoms by forcing people to give back, for free, the work they've done on top of GPL'd work they in turn received for free. With software as a service, this breaks down completely, because the derived work is never actually distributed.
The GPLv3 explicitly tries to deal with software as a service, but it's definitely more problematic since it's not as solidly grounded in copyright law as v2 is, it's harder to detect breaches of it, harder to prove breaches, and harder for the people in the justice system to understand it.
> The GPL increases your freedoms by forcing people to give back, for free, the work they've done on top of GPL'd work they in turn received for free
No, the GPL forces people to distribute the source code of their own modifications, but that's only on redistribution. That's why with software as a service this doesn't work.
This has to be emphasized, because that's its central strenght.
The majority of all software is made and run in-house, without ever being distributed. And open-source would not be what it is today without contributors that have SaaS as a business model.
> The GPLv3 explicitly tries to deal with software as a service
No it doesn't. The GPLv3 is still only a copyright license which only covers redistribution (as GPLv2). It is more problematic because it tries to define what distribution means (to prevent tivoization), and copyright laws have their own definitions, but it still is within normal bounds ... so it remains to be seen if it works, but FSF has got some pretty smart lawyers over there, and they did their homework.
The license that tries to deal with SaaS is AGPL.
But that's not a copyright license anymore, and while it is considered to be "open-source", some people think that it shouldn't be, since it places restrictions on the actual usage of the software (like an EULA).
And you'll have a hard time convincing companies that have contributed to open-source to switch to AGPL. Hell will freeze over or a new gap in AGPL will be discovered before that happens. One way AGPL can be "monetized" is with dual-licensing. But that's just dishonest and definitely not free.
If they let me download their code, they would have to let me use the source and sell it as well. But they can simply not let me download it and force me to run the code on their servers exclusively.
This is where the AGPL comes into play. It's a modified form of the standard GPL3 license, but with extra provisions specifically designed to include the of accessing the software from a remote server as distribution. By licensing web applications or server-based software with the AGPL, you then "force" the hosting entity to provide source code for any users of the service.
You're pretty much spot on. You don't just get the right to redistribute, though, you also get the right to modify. If you do modify, and distribute your modifications, you're obligated to release them under the GPL as well (this is what FUDers always cry 'viral' over).
The reason is that this is desireable is because anyone who's willing to spread what Mark has written helps Mark. Think of it as free advertising.
This could be much, much longer. I actually spoke on a panel at my local PodCamp about all of this a few weeks ago... It's a very complicated topic. A lot of the ideas are very counter-intuitive to a lot of people.
You've read a lot about Free software, but have you read the primary source? The Free Software Definition (http://www.gnu.org/philosophy/free-sw.html) would be a good place to start, if you haven't. Does that still not make sense? If not, I could try, but this is basically what I'd tell you, without some further context.
The definition of what is "Free Software" is arbitrary, because anyone could use a completely different definition and wouldn't be less right or wrong.
The proposition that "Free Software" is moral is also arbitrary, for there's no rationale. There's just some agreement because of the bandwagon effect and multiple meanings of words like "freedom", "sharing" and so on.
The simple truth is: It's sometimes useful and so is "non-free" software.
I sort of read this as more BSDish philosophy. One of the fundamental things he cites is the freedom for others to do something you might not like, which, IMO, is more of a BSD notion.
I have several little writing projects that I distribute, but I chose the Creative Commons 'no commercial use, no derivatives' license because that was what I felt good with.
People who create stuff have the right to choose the license.
I just had an Apress book published, and the contract gives APress the distribution rights - fine with me since they effectively paid me to write the book: their advances allowed me to have a really good time and spend most of my time writing for 4 months. (I enjoy consulting, but writing is even more fun.)
Im confused too; in my basic Contract Law class, there's something called Value Received. If I don't pay for something, there is by definition NO CONTRACT. So any words (gpl etc) printed on the thing are just decoration.
You're describing the concept of "consideration" in contract law -- i.e., that each party to an agreement needs to exchange something of value in order for a valid contract to be formed. However, one doesn't need to pay money for something in order for there to be an exchange of value.
Here, the copyright holder (Mark) has clearly given you something of value -- a copy of the text of his book. In exchange, you give the him something of value, as well - your a promise to abide by the terms of the GNU FDL (as referenced on the book's copyright page). Your promise has value to Mark -- if it didn't, he (presumably) would have simply released his work into the public domain, rather than under a GNU license.
The enforceability of open source licenses turns on this issue. The implication of your statement is that an open source license is unenforceable whenever a work is distributed for free (and obviously, this kind of distribution happens all the time). While there were some in the legal community who took this view in the past, the courts have rejected it. Check out the 2008 Federal Circuit decision in Jacobsen v. Katzer (http://www.cafc.uscourts.gov/opinions/08-1001.pdf), which turned on the enforceability of the open source Artistic License. Some of the more salient quotes:
- "Traditionally, copyright owners sold their copyrighted material in exchange for money. The lack of money changing hands in open source licensing should not be presumed to mean that there is no economic consideration, however. There are substantial benefits, including economic benefits, to the creation and distribution of copyrighted works under public licenses that range far beyond traditional license royalties."
- "The choice to exact consideration in the form of compliance with the open source requirements of disclosure and explanation of changes, rather than as a dollar-denominated fee, is entitled to no less legal recognition."
Which is why it's handy to remember that, while copyright licenses can be received as part of a contract, copyright licenses are not in themselves contracts.
Basically, if you want to do something that copyright forbids then you can't, unless you get permission. The GPL (and other share-alikes) gives you permission, as long as you obey its rules. Break the rules and you no longer have permission, just as if you were paying a monthly fee for permission and then stopped paying.
These two paragraphs are pretty odd sitting next to each other: http://timfreeman.org/copyright_dip.png