# Somebody scraped our answers and sold them as a book

I don't know if this has come up in discussion, but somebody scraped various tags on Phys.SE and copy-pasted the results together into several books, including one on QFT, one on GR, and one on optics. The books in print cost about \$6 while the ones out of print can cost up to \$1000. The prices are so exorbitant I thought it was one of those Amazon money laundering operations, but it seems to be serious.

• The book reviews on a few of them suggest there isn't even any indexing. Sounds like it's basically just a big list of questions and answers in some arbitrary order. I wonder if anyone actually buys those. – JMac Apr 7 '18 at 15:13
• Oh hey, it contains one of my questions. Neat! – AccidentalFourierTransform Apr 7 '18 at 15:23
• You can see sample pages from these on Google books. – Keith McClary Apr 7 '18 at 16:50
• Related: Printed Wikipedia volumes for about $80\,\frac{\text{USD}}{\text{volume}}$. – Nat Apr 7 '18 at 23:43
• A similar undertaking was proposed previously on this meta. – tpg2114 Apr 8 '18 at 12:24
• Wow, I finally got published! – Dvij Mankad Apr 9 '18 at 0:07
• " I wonder if anyone actually buys those." - not more than once I imagine @JMac – Alfred Centauri Apr 9 '18 at 0:32
• @JMac - The only goal of this "author" George A Duckett is likely to make some money. Otherwise he would probably not do that. Most of the publications are rather expensive paperbacks. – freecharly Apr 9 '18 at 11:39
• How do you know what's in the GR book? I can't see any content on the amazon page. The QFT book does have the "look inside" feature enabled. – Ben Crowell Apr 9 '18 at 23:49
• I'm not sure if I'd prefer these "books" not existing over them existing (with proper attention to the licensing), but since they do I must say I would have liked them to be a little neater... It doesn't look like they were proofread even once, it's all just very crude copy-pasting. I mean, there are things like ö etc. in them that are being shown as Ã¶. – Wouter Apr 12 '18 at 14:20
• @freecharly Note that the exorbitant prices being charged on Amazon (anything over ~6 USD) are exclusively (as far as I can tell) re-sale prices for used copies of the paperbacks. This likely means that they were set by (algorithms within?) Amazon, and not the author, and those sales do not benefit Duckett. As for JMac's question, the existence on sale of used copies conclusively settles the existence question (though not a quantitative estimate on how many copies were sold). – Emilio Pisanty Apr 13 '18 at 19:13
• Duckett seems to be making a living from producing "books" from all sorts of sources. I see historical texts from the British Library. Probably useful and interesting (legal? I don't know.) But if you have \$1,710 (at the time of this writing) doing nothing, you can buy a book of a scraped Mathematica forum. Surely, this is the deal of a lifetime. – garyp Apr 16 '18 at 13:55
• @garyp Again, those exorbitant prices would not go to Duckett, as they're resale prices (likely set by some algorithm inside Amazon that's trying to get as much money as possible without enough data to really inform the pricing) which benefit the reseller. For the Mathematica book you linked to, that appears to be vegasbooks, who somehow acquired six hard copies that they're trying to sell. – Emilio Pisanty Apr 17 '18 at 19:09
• I would guess most of Duckett’s profits have come from these bots buying copies! That’s really delightful; making a living off algorithms. – knzhou Apr 17 '18 at 19:11
• @knzhou I don't think it's bots buying the copies, either. My reading is that actual humans bought the paperbacks for whatever they were priced at some two-three years ago and then offloaded them to their local used-book shop at some point, which has an Amazon web storefront. I don't think bots come in any earlier than setting the prices on those used copies. And much as I dislike the collections (both because of their abysmal quality and their license infringements), it is unfair to lay the prices of the re-sale copies at Duckett's feet. – Emilio Pisanty Apr 17 '18 at 19:18

Actually, on further reflection, I don't think this is OK at all.

I'm all for re-use of the Creative Commons content that I post on this site, including commercial use if people can find ways to monetize it that are compatible with its licensing conditions. However, I'm not OK with reuse of that content in ways which infringe the attribution requirements (i.e. providing electronic copies without hyperlinks to the source post and author profile, and making clear on the text which website the content came from) or the licensing requirements (details below).

Generally speaking, it is very easy to get immediately incensed when you see your content used in ways you didn't anticipate, particularly if those ways include money changing hands. In those situations, say, when somebody says "somebody is printing Wikipedia and selling it for a profit", I often find it useful to append "... and it's being used to tremendous effect in schools in [insert developing country] where internet access is [insert suitable restriction]", and see if that changes how I feel about it. However, that is not the case with this book series.

Rob Jeffries is quite right in his initial statement

you can go to google books and read these Q+A without being aware that the content has been stripped from elsewhere, with no obvious indication of the CC-BY-SA license and with no proper links to the original source of the material.

and although the details vary depending on the precise presentation at issue, most of the ways in which the material is available are problematic enough to be infringing in way way or another.

And frankly: Rob is right, it's time someone took the time to write some sharp lines to that effect. So, I've written a pretty curt email (much longer than I intended, partly to leave no room for wriggling and partly because I didn't have time to make it shorter) to Duckett which I intend to send shortly I've just sent. I will update this answer as and when I get any news from that.

Before I do, though, I'm posting it here so I can get some community feedback in case my interpretation of the license is off in some way I'm not seeing right now.

Dear Mr. Duckett,

I am writing to you about your book series "Questions and Answers", collected from the Stack Exchange network of sites, as available through the Amazon and Google Books sites. This email requires action on your part.

As regards attribution, the requirements for reusing Stack Exchange content are clearly spelled out in the Stack Exchange blog (https://stackoverflow.blog/2009/06/25/attribution-required/) and they essentially follow the basic common-sense requirement that any reproduction of the work in electronic means must contain a hyperlink to the location of the original work as well as a hyperlink to the author's profile.

As regards the license, you are required to license your derivative work only under the terms of the CC BY-SA 3.0 license; moreover, as the license makes it clear in section 4.a,

You must include a copy of, or the Uniform Resource Identifier (URI) for, this License with every copy of the Work You Distribute or Publicly Perform.

Similarly, as noted in that same section of the license,

You may not impose any effective technological measures on the Work that restrict the ability of a recipient of the Work from You to exercise the rights granted to that recipient under the terms of the License.

Finally, as regards the license, I draw your attention to section 7, which clarifies that redistribution of the work in breach of the license's terms will cause its termination, at which point you lose the right to use the work as originally licensed.

This email concerns content whose copyright rests with me, which was made available on the website Physics Stack Exchange (https://physics.stackexchange.com/) as well as other sites of the Stack Exchange network, and which you have redistributed as part of several collections of your 'Questions and Answers' series, which are available

• as Kindle ebooks on Amazon.com, currently in print and for sale whose profits (ostensibly) benefit you directly;
• as out-of-print paperback hardcopies from Amazon.com;
• as samples of the paperback hardcopies, displayed as electronic copies and attributed to you, on Google Books;
• as eBook samples, displayed as electronic copies and attributed to you, on Google Books;

You can find a selection of the books in your series which contain content under my copyright using the Google Books search feature (https://www.google.com/search?tbm=bks&q=inauthor%3A%22George+A+Duckett%22+emilio-pisanty). These include your books on Physics, Thermodynamics, Quantum Mechanics, Energy in Physics, Electrostatics, Particle Physics, Fluid Dynamics, Physics Exercises, Optics, and Mathematica.

Your redistribution of the work is in infringement of the CC BY-SA 3.0 in (at least) the following ways:

• The Google Books samples, as provided by default, do not contain any hyperlinks to either the source of each work nor the profile of the author.
• The Google Books samples, as provided by default, do not contain any indication of the origin of the work as globally contained in the collection: they have an 'About this book' page which does not reference the website from which the book's content originated.
• The Google Books samples, as provided by default, often contain no indication of the licensing conditions of the work, as the final page (containing the scant copyright and licensing information) is not provided.
• Where the Google Books samples, as provided by default, do contain licensing information, they still do not contain suitable indication of the CC BY-SA license: where the final licensing page is present, the information provided, "All questions and content within this book are licensed under cc-wiki with attribution required" is insufficient. To clarify this point further:
• The name "cc-wiki" is essentially meaningless and it was clearly deprecated by both Creative Commons and Stack Exchange as early as June 2009 (https://stackoverflow.blog/2009/06/04/stack-overflow-creative-commons-data-dump/). The correct name of the license is CC BY-SA 3.0, including the license version (which does matter).
• The collection does not contain either a copy of the license text or the URI of the license, as required by §4.a of the license.
• The Google Books samples, in their eBook version, contain no indication of the licensing conditions of the work, as the final page (containing the scant copyright and licensing information) is not provided.
• The Kindle ebook versions as provided in samples by Amazon.com do not contain any indication of the CC BY-SA license, nor do they contain the license text as required.
• The Kindle ebook versions contain technological measures (generally known as Digital Rights Management) that prevent their users from redistributing the work, a restriction that is explicitly forbidden by the license.
• The paperback copies (as depicted in the Amazon.com "Look Inside" feature) do not feature any indication of the website from which the work originates, as their 'About this book' page does not reference the Stack Exchange website from which the book's content originated.

As indicated above, the infringement of the license terms causes the license to expire.

More generally, the core problem is that readers can see the content, both on Google Books as well as on the paperback copies, without having any idea as to where the content came from.

I should also note that several others have found your approach problematic, as discussed on the Physics Meta site (Somebody scraped our answers and sold them as a book), (as well as a previous discussion at Are these eBooks that copy from SE illegal?); indeed they may contact you independently. If you wish to address these issues in good faith directly with the Physics Stack Exchange community, I would ask that you do so at that thread.

In the final page of each book you include a statement ("If you have or know of copyrighted content included in this book and want it to be removed please let me know at georgeduckett@hotmail.com") that indicates that you intend to do right by the license terms. Given the fact that the books are already published and that several of your mistakes cannot be undone without altering the record, your options are quite limited, but if you want to retain the license to the work, you do need to fix the problems.

To be clear, I am not requiring you to remove the content under my copyright from your collections: instead, I am requiring you to stick to the CC BY-SA license's terms as stipulated above. This means that you must:

• Ensure that all electronic copies of the work, as provided by both Google Books and Amazon.com, contain appropriate attribution, including hyperlinks to the source post and its author's profile page.
• Ensure that all electronic copies of the work, as provided by both Google Books and Amazon.com, contain appropriate licensing information, including the correct name of the license and its full text or the URI of a complete copy of the license.

This will likely require you to ensure that the hyperlink-less copies are removed from Google Books and that the eBook sample is made available without cutting any attributions and without removing the license statement (i.e. likely to make the eBook sample available in full), and ensure that the full and correct license information is available from the landing page.

Similarly, this will require you to ensure that the Kindle eBooks (currently still on sale) are amended to include the correct license and a URI of a complete copy, and to ensure that the DRM is removed from all copies of the work, as well as ensuring that the full licensing information is available 'above the fold' on the Amazon.com landing pages.

If you are unwilling or unable to fix these problems, or if you do not respond to this communication, my next steps will be, at my convenience, to initiate Digital Millenium Copyright Act takedown proceedings against the infringing copies of your books on both Google Books and Amazon.com.

I look forward to hearing from you regarding your planned actions on resolving the infringing aspects of your published collections.

Sincerely,

Emilio Pisanty

Oh, and finally: I'm releasing this post under CC0, in case you want to, say, privately distribute slightly-modified copies of this text by email.

I'm the creator of these books. Firstly I should say it was not my intention to break any copyright or cause any ill feeling among contributors. When initially evaluating submitting the books to Amazon/Google/iTunes I checked the terms of the license and checked with Stack Overflow and believe I did everything needed to comply.

The eBooks and physical books should no longer be available on Amazon. I followed some example links however in each case I checked it said it was unavailable as I'd expect. If you see any that you can buy (that aren't second hand) please let me know and I'll follow up with them. I do believe I complied with all relevant terms for these books while they were available. Print books had all relevant urls for any link (to user/post/license) in the footer of the page the link appeared on for example. FWIW I have not set the prices of the used books that are available, nore the prices any resellers choose to charge.

Regarding the eBooks available via google play, there is no DRM applied. This is a setting per book and every one has the DRM option disabled. I was unaware of the wiki license name change (it's included in that 2009 post, but it's unclear when the notice re. the update was added). I'll update all links where I reference it, and also include the bare link as well as the license name/hyperlink. This is not an instant change but I will update this post when completed. I've updated all books to indicate that I'm the Publisher, rather than the Author.

Finally, if any user would like their content removed from all current and future publications I will do so. Likewise, if the physics.se community collectively decide they do not wish to be included in any publications I will remove them from sale and not use this site as a source for any future publications.

-George Duckett

Update:

Samples. I've found an option within google to only allow the epub format as a sample, rather than an auto-generated PDF which mangles hyperlinks. This is in effect now; when clicking "free sample" I get a message that my browser is incompatible (presumably because it can't open the epub file). I've yet to be able to open the sample. I will keep trying various methods until I can verify that hyperlinks are preserved or receive confirmation that the samples will not work anywhere. I will move the copyright section to be immediately below the about section so a casual reader should be made aware more easily. Once that is done for all books I will probably reduce the sample from 100%. Any changes to books' contents will happen over the course of a few weeks as I re-generate and re-publish them on google's platform.

Author. I've changed my name to be "publisher" rather than "author" on Google's stores. As the books are no-longer on Amazon I can't change the data unfortunately. I also cannot find a way to change the Author field on iTunes.

Update 18/04/2018

Samples. I've submitted updated samples to google for all books sourced from physics.stackexchange. The rest of the books will follow. Google says to allow for at least 24 hours to see the changes. I will be keeping an eye on this as I want to make sure the updates worked correctly. Samples are still available via Google Books with hyperlinks not rendering correctly in their preview. I've got a support ticket passed on to the "engineering team" at google (presubably a team specifically for google books) regarding this. They couldn't give me a timeline for when they'd get back to me.

Update 24/04/2018 I have updated all books. Google will take time to update them though, probably no more than a day or two. I have however asked for all books to be deactivated so they won't be listed on Google. This will take time as their content team has to do it given the volume of books.

• I've just checked and the PDF download also doesn't include links. I've removed that option for now. I should be able to provide a PDF myself, hopefully preserving hyperlinks however it will take some time to generate/create. I will of course update. – George Duckett Apr 16 '18 at 21:50
• Is there a timeline to these updates? Kudos for changing “author” to “publisher” but Amazon still shows you as author as of time of my writing. – ZeroTheHero Apr 16 '18 at 23:36
• @ZeroTheHero: Editing the books will take some time as I'll have to regenerate them each individually. On the order of a few weeks if I'm being realistic. Re. Amazon still displaying me as the "author" unfortunately since all books have been removed from their platforms I have no control over that. – George Duckett Apr 17 '18 at 11:23
• Thank you for your response - it is good to see that you're taking this seriously. As of right now several books are unchanged (so it would e.g. be helpful if you indicate which books you think are no longer infringing) but it is fair that this can take a few weeks. (The version 4.0 of the CC BY-SA license establishes a 30-day deadline; it's not in effect for 3.0, but it's definitely a marker past which it'd be unreasonable for the licensing errors to remain.) I look forward to seeing the fixes implemented. – Emilio Pisanty Apr 17 '18 at 19:02
• Good luck everyone. Hope that the issue is resolved quickly ;-) – tatan Apr 18 '18 at 5:52
• @EmilioPisanty: See my latest update. tl;dr:physics books updated pending google processing them fully, following up with google as to why the preview in google books doesn't show hyperlinks (and in general doesn't render 100% correctly). FWIW I think the samples are a bit of a grey area (although of course I want to fix the problem) but I still believe all my published books are in line with the license. These changes are to remove any doubt (and are the right thing to do, having found out about the problems). – George Duckett Apr 18 '18 at 14:24
• For the record, I don't have an issue with you using my content in any financially beneficial way, but if you have used my content in making any of your books, the least you could do is send me a free copy of that book – Jim Apr 24 '18 at 13:15
• Wouldn't the publisher still be CreateSpace? I feel like usually when a book has content from multiple authors, the person who makes the compilation is the "editor". – martin Jun 29 '18 at 8:28

• This doesn’t prevent people from writing reviews of the book, indicating that the contribution of the author is merely in selecting questions and answers that are available freely from this site. – ZeroTheHero Apr 7 '18 at 14:05
• This answer is incomplete. SE content isn't just on a CC BY license - there's also a ShareAlike requirement, which restricts the type of license that book can be released under. They're allowed to charge for print versions but if there's an ebook then it needs to be openly available or they're breaking the license terms. – Emilio Pisanty Apr 8 '18 at 12:39
• @EmiloPisanty so you're saying that animuson's answer to the meta question may be wrong (there are Kindle eBooks as far as I can tell)? – ACuriousMind Apr 8 '18 at 13:35
• As an add-on to my previous comment: watch for the conditions on writing Amazon reviews: only links to other Amazon products are allowed, else the review will not be published, and you don't have a chance to correct the review. – ZeroTheHero Apr 8 '18 at 14:07
• @ACM I do think that the ShareAlike status of the Kindle versions of those books is insufficiently clear, and animuson's answer isn't wrong so much as purposefully inconclusive. On the other hand, of the 72 SE-scraped books listed, only two are available as Kindle ebooks, and the "author" (better termed as editor) is indeed a long-time user of Stack Overflow, so I do partly share animuson's assessment that license violations are unlikely to be present, and that they're relatively unimportant if they're there. – Emilio Pisanty Apr 8 '18 at 15:56
• @EmilioPisanty: I think you're right that it violates the SA part of the license, but I don't think your analysis is quite right. I think the violation is that nowhere in the book does it carry a copyright notice or a statement of the license. I don't think the license treats electronic books differently than paper ones, nor does it forbid charging for electronic books. – Ben Crowell Apr 9 '18 at 23:51
• @BenCrowell I agree that it's a complex issue and you're probably right - but then again, the Kindle versions are likely have some form of DRM, which is unlikely to be SA-compatible right off the bat, I should think. One would have to get a full copy and examine it (or maybe just have a full look at a free sample? but even then it's a drag) to tell for sure. Given that there's 72 books on the series and only two are still have Kindle versions available - do we really care? – Emilio Pisanty Apr 10 '18 at 0:17
• The format of the advertisement suggests that George Duckett is the author of the questions and answers, rather than just the editor or the compiler of these. – ZeroTheHero Apr 10 '18 at 21:45
• @ACM I've changed my mind. I do think the work is infringing and I do think it warrants action - see my answer. – Emilio Pisanty Apr 13 '18 at 20:00

I can see that the CC-BY-SA license allows people to produce this kind of thing, but I don't think that excuses the fact that you can go to google books and read these Q+A without being aware that the content has been stripped from elsewhere, with no obvious indication of the CC-BY-SA license and with no proper links to the original source of the material.

• The "Contact" link (on the bottom of every Stack Exchange page, including this very page) has the specific reason "Stack Exchange content is being reproduced without attribution" (under "What can we help you with?"). More important, action is actually taken by Stack Exchange for these. – Peter Mortensen Apr 11 '18 at 12:33
• @PeterMortensen SE is already aware of the book series, which covers all of the network ─ see the link in ACM's answer. – Emilio Pisanty Apr 11 '18 at 15:14
• @Rob Apparently Duckett uses the term "cc-wiki" (cf. the last page here) ─ and apparently he considers that to be a sufficient description of the license. – Emilio Pisanty Apr 11 '18 at 15:25
• @EmilioPisanty When I go to this address books.google.co.uk/… I see no licensing information and no proper attribution (perhaps because I am not allowed to see the copyright page without buying it). I can read lots of everyone's questions and answers and be none-the-wiser about where they come from (though it is obvious that they are not George Duckett's work). – Rob Jeffries Apr 11 '18 at 15:55
• @EmilioPisanty A minimal attribution would be to include a hyperlink to each and every Q+A. Perhaps he does that and the problem is that this isn't reproduced when you look at on google books? – Rob Jeffries Apr 11 '18 at 15:56
• @RobJeffries I suspect that the final page containing the (scant) copyright and licensing information (as on the MathOverflow "book") is standard for the series and has been cut out from the Google Books preview of that specific issue. The Google Books copy may not be under the control of Duckett, and it may be a Google scan of a print copy, or a similar mechanism that stripped links (and possibly some pages) from the content. Either way, that would've been prevented by correct licensing on Duckett's part, so it's not an excuse. – Emilio Pisanty Apr 11 '18 at 16:08
• There's also an argument to be made that the omission of pages on the part of Google Books counts as the imposition of DRM, which is specifically forbidden by the CC licenses. – Emilio Pisanty Apr 11 '18 at 16:09
• As to being 'none the wiser' about where the content comes from, the 'about this book' link has a clear description, so maybe it's a tad too harsh of a description. Also, if you go here and click the drop-down on 'View sample', there's a (sample of an) ebook copy with correct hyperlinks to both posts and users. – Emilio Pisanty Apr 11 '18 at 16:12
• @EmilioPisanty OK, I see - in the small link on the left. I was referring to the book itself. There is an "About this book" section there which says nothing about where the stuff comes from. – Rob Jeffries Apr 11 '18 at 16:17
• @RobJeffries Sure, that's definitely true. For me, the bigger problem is that the CC BY-SA license has an explicit requirement that "You must include a copy of, or the Uniform Resource Identifier (URI) for, this License with every copy of the Work You Distribute or Publicly Perform." and this is is nowhere in sight in either copy shown by Google Books. – Emilio Pisanty Apr 11 '18 at 16:19
• @EmilioPisanty Yes I see. The problem is that you can look at most of the book on google without seeing any of these hyperlinks. All you see is that rob-jeffries wrote this answer. – Rob Jeffries Apr 11 '18 at 16:21
• Well I just put in a review pointing out that all this content can be obtained for free, so you would be a fool to buy it. – Rob Jeffries Apr 11 '18 at 16:22
• @RobJeffries Please see my answer. You might want to consider a similar course. – Emilio Pisanty Apr 13 '18 at 19:58
• I think Google Books only publishes extensive sample pages with permission of the copyright owner. If so, he must have represented himself to Google as such. – Keith McClary Apr 16 '18 at 4:51
• @Keith The copyright for the collection itself still rests with Duckett, unless (in the US) a court were to rule that the effort in doing the collection was so trivial that the collection effort itself does not actually generate copyright. There's some grounds to that (given that the selection appears to just be from the site voting data with no actual active curation) but it's still a pretty large stretch for the collection copyright to be voided. (That's the copyright only, mind you - the CC license of the content still restricts the conditions under which the collection can be distributed). – Emilio Pisanty Apr 17 '18 at 19:02