Authors´ Guild vs. reality: Kindles and read-aloud

In yesterday's New York Times, Author's Guild president Roy Blount Jr. rails against the Kindle's text-to-speech feature, opining that it infringes copyright because it provides a "derivative work" by creating audio editions of your textfiles. Blount says that eventually, text-to-speech will be so darned perfect, the audiobook market will be destroyed by it, so he aims to do something about it right now.

He doesn't actually say what that thing is. Presumably, he'd like Amazon to simply remove the feature. But if you take Blount at his word, then we can only assume the feature will spread to other platforms -- does he think that iPhones shouldn't be able to read your email to you as you jog? And if it can read your email, what's to stop it from reading your ebooks? Blount implies that there's a simple solution to his technological problem, but if you truly believe that it should be illegal to ask software to produce audio of copyrighted works, then Blount's lining himself up to fight the entire future of technology that can convert text to speech.

Continuing to take Blount at his word, let's assume that he's right on the copyright question, namely, that:

1. Converting text to speech infringes copyright

2. Providing the software that is capable of committing copyright infringement makes you liable for copyright infringement, too

1. is going to be sticky -- the Author's Guild is setting itself up to fight the World Blind Union, phone makers, free software authors, ebook makers, and a whole host of people engaged in teaching computers to talk.

But 2. is really hairy. If Blount believes that making a device capable of infringing copyright is the same as infringing copyright (something refuted by the Supreme Court in Betamax in 1984, the decision that legalized VCRs), then email, web-browsers, computers, photocopiers, cameras, and typewriters are all illegal, too.

Time and again, the Author's Guild has shown itself to be the epitome of a venal special interest group, the kind of grasping, foolish posturers that make the public cynically assume that the profession it represents is a racket, not a trade. This is, after all, the same gang of weirdos who opposed the used book trade going online.

I think there's plenty not to like about the Kindle -- the DRM, the proprietary file format, both imposed on authors and publishers even if they don't want it -- and about Amazon's real audiobook section, Audible (the DRM -- again, imposed on authors and publishers even if they'd prefer not to use it). But if there's one thing Amazon has demonstrated, it's that it plans on selling several bazillion metric tons of audiobooks. They control something like 90 percent of the market. To accuse them of setting out to destroy it just doesn't pass the giggle-test.

One of the most powerful weapons in the publishing industry's arsenal is that it isn't the record or film industry. By and large, publishing is undertaken by bookish people who love books and bookselling and readers and writers. By and large, writers get a decent deal from their publishers -- especially relative to recording artists; most writers don't have to sign over their copyrights, don't have the promotion of their books deducted from their royalties, etc. By and large, publishers don't sue tool-makers or accuse readers of being crooks.

Unlike the record and film industries, who seem bent on doing everything in their power to build the moral case for ripping them off -- to convince the public that they are a passel of greedy, clueless technophobes who deserve to have their industries killed, if only to protect the 21st century from them -- there are very few people who feel this way about publishing and authorship.

Unless, that is, groups like the Authors' Guild continue to make us all out to be cut from the same cloth as media execs like Universal Music's Larry Kenswil, who once bellowed "FAIR USE IS THE LAST REFUGE OF THE SCOUNDREL" at me from a stage at the RSA in London.

Dear Mr Blount: you don't represent me. You don't represent the future of authorship. You and your group are jeopardizing the future of authorship and of society with your petty little grabs and ridiculous posturing. Cut it out before someone gets hurt.

True, you can already get software that will read aloud whatever is on your computer. But Kindle 2 is being sold specifically as a new, improved, multimedia version of books – every title is an e-book and an audio book rolled into one. And whereas e-books have yet to win mainstream enthusiasm, audio books are a billion-dollar market, and growing. Audio rights are not generally packaged with e-book rights. They are more valuable than e-book rights. Income from audio books helps not inconsiderably to keep authors, and publishers, afloat.
The Kindle Swindle?

152

  1. Maybe they’ll go on strike, like the scriptwriters.

    (… and give independent authors some room to break into the market).

  2. And yet, Cory, and yet. Is a book spoken aloud by a computer an audiobook? Hard to make the case that it doesn’t represent an assumption of rights not granted. If you’re listening to it, it’s clearly not just the written word you’ve purchased. This is material for negotiation. Amazon should simply purchase audio rights, albeit at a discount given the elimination of overhead for publishers in producing such audio versions. There’s a lot of sound and fury here about principles that distract from the fact that its really just a question of pricing.

  3. Wired’s article on the Kindle vs. open standards (which also quotes you) also quotes an Amazon representative insisting that the Kindle is “DRM-agnostic.”

    Amazon says authors and publishers can choose to submit their content without DRM, using the Amazon Digital Text Platform, so books can be shared with more than six Kindle users. The Kindle also reads a small collection of non-DRM, Amazon-owned Mobipocket e-books.

    “Amazon is agnostic when it comes to DRM with e-books,” Amazon spokesman Cinthia Portugal told Wired.com. “We give content owners the choice — we just want to make the content available to readers.”

    Most publishers these days, however, are pushing for the DRM protection out of fear of what happened with the music industry with piracy and file sharing.

  4. Both Windows and OSX ship with text to speech software. It’s not that great but you can get better. Even then I wouldn’t want to listen to a whole book read by it.

    I don’t have a Kindle but if I did could I go to a park and let it read a book out load to some people gathered there?

    What if someone were to call in to “Wait Wait… Don’t Tell Me” and then had their Kindle read one of Mr. Blount’s books out load? He’s very funny but I wonder if he’d get the joke?

  5. Without wading too deep into the hyperbole here, Blount does make an interesting point about the advancement of technology. I see a lot of people dismissing the whole issue on the grounds that text-to-speech can never compete with a real live voice reading a book. I think he’s right that this will change, though, just as animation has begun to take the place of real live actors.

    I’m not taking his side, mind you, just wondering how things are going to look when you can sample a favorite actor’s voice and have them read your textfile books to you, dramatic inflection and all?

    The specific performance of a piece of music, and the music itself, have different rights and protections applied. If you make a machine that automatically performs a piece of music, who owns that performance?

    And why not have a seperate license for automated performance? Independent writers can deal with it the same way the license the rest of the work, and greedy publishers can choose to charge for it and lose more business.

  6. Agreeing with Noen: I’m not sure anyone who’s reading for pleasure or the love of a book or author would want to hear the machine speech version.

    To anyone with a Kindle: just how good is the text-to-speech?

    It seems like the technology would have a long, long, long way to go before it could replace one of the very few (a few hundred?) people worth listening too when they read a book out loud. Maybe an audio equivalent of the Uncanny Valley involved here.

    ^ Which is all secondary to the main issue: is it a derivative work, or a fair use of the same work?

  7. Right! We also need to stop any attempts to develop a cure for cancer since it will undoubtedly destroy the market for radiation and chemotherapy!

  8. um, i think it will be a LONG time before a computer can create the equivalent PROFESSIONAL PERFORMANCE of an audiobook read by a talented narrator from an ebook source document.

    for incidental listen-reading, sure, read-aloud is handy, but having Kindle’s equivalent of Mac’s Alex voice stuttering through an ebook novel is not my idea of entertainment!

  9. The point in which a piece software can replace a human reading a book aloud is way, way off. Yes, there is software that ‘sounds’ human and it’s getting better all the time. But how will software understand the emotional context of a scene, when to whisper, when to shout? When to speed up and when to slow down. How will it determine just the right inflection to indicate sarcasm or bitterness?

    Now, I can envision a way to place queues in the text that directs the software to the appropriate vocal inflections and pacing and such, in essence, programming the super advanced reader to behave like an audio book. But that would take a lot of work and would be very deliberate, which is a far cry from an electronic reader cold-reading text.

    It shouldn’t have to be said, computers can only do what they are programmed to do. They cannot ‘understand’ anything. It’s that understanding only a human reader can convey when it comes to audio books.

    As far as the comment about CG animated characters replacing human actors: it takes a great amount of time, planning, and manpower to create a digital performance, and is still usually driven by a human actor (i.e. Andy Serkis), at the very least, a voice actor (which is very demanding).

  10. anonymous#6:

    Plus he’s consistently the least funny panelist on “Wait, Wait, Don’t Tell Me…”

    ad hominem. I always thought he was pretty funny. But Paula Poundstone is flat-out hilarious.

  11. These are mostly strawman arguments – the heartstring-pulling about blind people is particularly offensive, especially considering that Blount mentions in his article that he has no truck with any sort of adaptive devices. The comparison between email and an e-book is silly, because email is personal communication over which few people ever need to claim copyright, whereas an e-book is a copyrighted work of fiction which a person purchases.

    Copyright law gives authors certain rights to control whether and how their works are distributed, and one of the ways they have worked it out with publishers and distributors is by establishing separate licensing terms for e-books and audiobooks. Sure TTS sounds like a drunken Swede now, but that’s not the point; the point is that whatever the technological process involved in creating the rendition, Amazon is basically selling an audiobook for the price of an e-book, which is lovely for Amazon, because then they’ll sell more of their ebooks for the same price. It’s the authors who will get the shaft.

    Recap: buying an e-book: ok. Using your own TTS software to read your e-book: okay. Reading the e-book aloud with adaptive device: okay. Reading the e-book to your kids, hiring somebody to read the e-book, or any of the other silly scenarios that have been proposed: okay. What’s NOT okay is creating a platform that enables you to buy e-books and sell them to other people as audiobooks without paying the royalties to the authors. Now here’s an example that’s actually relevant: it wouldn’t be okay for me to record Harry Potter and sell it w/o the permission of the publisher. So why should it be okay for Amazon to effectively sell recordings of e-books?

    It’s nice that you are so supportive of free speech and copyleft. And what fiction of yours I’ve read, I’ve enjoyed. It sounds like you’ve had some unpleasant experiences with the Author’s Guild; I’m not sure what. But not all authors have world-famous blogs from which to promote their work, and the eradication or abrogation of copyright law wouldn’t serve our culture at all. And if you read Blount’s editorial without foaming at the mouth over the fact that he dares to question the “free information” doctrine, you might find it’s not so unreasonable after all.

  12. Under English law, for the reading to be copyright infringement it would either (1) have to be a public performance; or (2) the resulting work would have to be a derviative work itself capable of copyright protection. Copyright protection only arises once a work is “fixed” in some medium. Since, as I suspect, the electronic reading process is a transient process, and the resulting audio is not stored somewhere, then there is no fixation, and no copyright. If someone (anyone) does record the audio output, then fixation would occur.

  13. If producing software that can create a credible text-to-speech is infringement, who will make software for blind people?

    And if making a tool that *can* commit an infringement is an infringement in and of itself (“creating a platform that enables you to buy e-books and sell them to other people as audiobooks without paying the royalties to the authors”) then why isn’t a browser an infringement?

    Your argument is incoherent and dwells on the percieved unfairness of Amazon making a device that allows the owners of copyrighted works to get more value from them. Adding value to copyrighted works with technology isn’t — and has never been — unlawful.

  14. Going along with EICOS on this one – one of the first times contrary to Cory’s contention. Here are my thoughts – technological hard determinism – the reductionist idea that since the technology is “out there”(*in email in this case) inevitably wide adoption must take place is jst not reflective of reality or the past. While technology unquestioningly brings decisive change with its new developments, adaptation of that new technology is not pre-defined without our input but rather our tools and their us is influenced by concerns about power and politics. Over time,many developments have been regulated or even prohibited without all going down the tubes – in the case of Atomic bombs it’s a godsend in fact.

    Secondly, and the strongest support of Blount in this, is his claim that the Kindle feature essentially represents a derivative work. And as someone above pointed out – a performance of the work – and an unsanctioned one at that. Up until last year’s strikes, creators/writers of TV/Film were not benefiting from web and even DVD “re-purposing”. They had to fight for that share b/c guess what, those priming the pumps weren’t planning on cutting them into the deal. Why should authors have a right to determine how there work is heard if people want to hear it.

    And finally, there’s quality. My kid and I sometimes listen to “THGTTG” or The Hitchhiker’s Guide to the Galaxy. How disembodied and to some degree disemboweled would our consumption pleasure be if we had a passable, but arbitrary reading voice instead of, in our version, the author himself. As the work was intended to be read – as a “fixed performance”, and so properly declared without digital subterfuge.

    Cheers

  15. Well, I haven’t listened to a Kindle (and don’t plan to), but if they sound anything like every other text-to-speach program I’ve ever heard… I really, really, never want to.

    As much as I appreciate their effort, I find it annoying enough listening to some of the less talented Librivox volunteers.

    T2S is downright painfull, and not even remotely a threat to audiobooks at the moment.

    And really, if T2S stops being crap in the future, and it probably will, because the markets don’t overlap a lot, they’ll basicly be able to sell to both markets, without the added costs of producing the audiobook.

  16. from Neil Gaiman’s post on his agent’s view … We’ve sold audiobook rights and print book rights as separate things. We must stop this.

    Initially, I read that to mean “we must stop selling the rights separately”. Buy a book, you get to read it, you get to have it read to you, your kids are allowed to turn it a play and perform it for you, etc. Sounds entirely reasonable to me.

  17. The real issue is that the publishing industry created a compensation structure that didn’t anticipate this technology, and they are too lazy to do the work of fixing it. Trying to use the legal system to make technology conform to their business model is the same failed approach the recording industry tried against digital audio.

  18. Mr. Blount’s arguments are ridiculous.

    1. Does there exist, or are there any soon-to-be-released text-to-speech systems that actual parse out and comprehend content? Such would be the requirement to properly render a text file as is done by a Narrator.

    2. Voice sampling as a means to “skin” a text-to-speech” device? Again, Mr. Blount makes a claim against future technology that is patently unrealistic in even the mid-future. One could possibly create an electronic voice with similarities to a particular Voice talent, but would the book sound the same? Ridiculous. Take for example, Susan Erickson’s rendition of the J.D. Robb novels. Anyone who has ever listened to a such work knows that a machine will not produce that sort of content until “HAL” is a serious threat to “Dave.”

    The idea that the we are even close to having the processing power and software to mimic a human’s interpretation of the author’s story, to understand and deliver it with inflection, that is, with the inflections of a skilled Narrator, and with the voice characterizations of a skillful Narrator, reflects a complete ignorance of computer technology and belittles the skills of those he purports to represent.

    The J.D. Robb audio books are a prime example of performance by a pro who is in top form. Another example would be Patrick Tull, who has done such an excellent job reading Patrick O’Brian’s Aubry/Maturin series (of which “Master and Commander” is one).

    Mr. Blount, have you ever listened to an audio book? I heartily recommend that you try it before claiming that a Kindle will prevent people from buying an audio book.

    Sir, you are indeed a luddite, a panic monger, and a fool.

  19. Shit.. does this mean I can no longer read books to my children? If I have children? When is using a bathroom or walking on the sidewalk going to make me a pirate?

  20. Blount should be working on behalf of authors to get their voices digitised and sold as add-ons to the kindle will read in the author’s own voice, and pull it back that way.

  21. Can I just ask who the Author’s Guild actually represents? It seems from what I can gather that pretty much no authors I’m interested are backing them at all in their bizarre technophobic crusade. Maybe it’s time to get a sizeable group of sane authors together to make a definitive statement dissociating themselves from the AG, that’d be a kick up their backsides. I’m assuming that without authors behind them they’d be nothing…

  22. I think the Author’s Guild’s reasoning is something like the following, once you set aside all of the talk about copyright.

    The Authors Guild wants to continue the option of separately monetizing text and audio. They want Kindle 2 to provide the option to lock text-to-speech so that if you want text-to-speech for a title, a buyer must pay for it.

    But the copyright discussion is irrelevant to them and to me for entirely different reasons. I think the current situation is such a temporary condition that I expect their objective to be obsolete.

    I think that in the near future, your book content is going to come with everything, for no extra charge: text, text-to-speech, and Jim Dale reading the text. Right now the only reason it doesn’t happen is because Jim Dale’s stream takes up too much space on the media. Well, how long is that awful situation going to last? Not very long, certainly not as long as it will take Amazon to get the lead out and produce a braille-capable, voice controlled Kindle.

    The reason I think the Author Guild argument is obsolete is because book sales generally could always be better. If I could triple the sales of my book by including a stream of me reading it to you, I’m gonna do it! Or I can get Jim Dale to do it. Maybe. The option to do so has not really been available, the book industry doesn’t sell a lot of books for use on laptops and desktop machines. Mass market Kindle-ish hardware presents us the opportunity.

    At any rate, being able to have the buyer read some text, have Jim Dale read part of it aloud, and maybe have T2S read part of it if there is no audio stream, that’s really going to be a regular feature.

    If Bezos focuses on Kindle 3, it’ll get done by this time next year, and the world will truly rejoice.

  23. Cory,

    I think that you are setting up several straw men in order to avoid the legitimate core point which Blount makes.

    Software which reads your email does not compete with an audio book. Software which reads your books on a Kindle does compete.

    Yes, I know it is the same software, and someone could use their OS text to speech software to read their audio books, but that is not what they are mainly used for.

    The Kindle is in part sold on the ability to read to you. That means it is ‘performing’ the work.

    The legal question is at best ambiguous, but as I read it his point is pretty strong.

    And software does get better, and so automatic systems will get better and better and will replace a substantial number of human read audio books.

    And authors who are not A-list bloggers, as well as voice artists, stand to lose out.

    That is fine. Nobody has a guaranteed business model, but under our legal system authors have certain rights. Kindle 2 does things with text which most authors did not agree to when they licensed their work.

    An audio performance of a written work, even a computer created audio performance, is different from the written work. You are arguing that because software exists which can perform a work that there is no difference. But you are absolutely wrong in that claim.

    I have no trouble imagining software which could analyze a written work and create a full blown CGI movie from that work. Should I be able, without an additional agreement with an author, be allowed to use that software?

    I don’t think so.

    I think it is important to recognize that Blount is making a very real, and true, claim that we have always recognized a distinction between a written work and a performance of that work and that this distinction remains true even when the performance is created by software.

    The answer seems painfully simple: just recognize his claim, and then use the relatively straight forward tools which exist to decide what percentage of the increased value should go to the authors of the source material and which percent to the creators of software.

    But the first step is to recognize that at the core, Blount is right and Cory is wrong. Performance, by a person or a program, is different from text.

  24. I don’t agree with Blount but some of the self righteous comments are out of hand and obviously you have not bothered to read what he wrote. For one, he notes that the Kindle speech engine sucks but makes the obvious point that they are more concerned about the future. Anyone who thinks they know for sure what will happen with technology is the fool. Also, corey claims he is taking on the Blind but fails to note that Blount addresses that:

    Blount, “In fact, publishers, authors and American copyright laws have long provided for free audio availability to the blind and the guild is all for technologies that expand that availability. (The federation, though, points out that blind readers can’t independently use the Kindle 2’s visual, on-screen controls.) But that doesn’t mean Amazon should be able, without copyright-holders’ participation, to pass that service on to everyone.”

    You can disagree with him but not to it that he talks about is is pretty disingenuous. Lastly, if he doesn’t represent thethe future of authorship how can he jeopardize it?

  25. First, let me say that I don’t buy that Amazon is creating a derivative work by providing a TTS tool — no more than Xerox creates a derivative work by manufacturing a photocopier.

    *If* a TTS reading is a derivative work, and *if* its unauthorised production is an infringement, then the person committing the infringement is the owner of the Kindle, when s/he presses the button to convert it to speech (just as the infringement that takes place if you record a movie with your digital camera takes place when you press the record button, not when Panasonic puts the camera in the box).

    What’s more, even if you stipulate that a fulltext TTS reading of a work is an infringement, there are plenty of instances in which a partial text reading would not be, or in which a fulltext reading would not be (say, for example, if you took your in-copyright-in-the-USA ebook of George Orwell’s 1984 to Canada, where it is in the public domain and then pressed play).

    The Kindle’s TTS can be used to read ANY book or other text you load on your Kindle — just as TTS on your PC can be used to read any text. There are literally tens of thousands of public domain books that you can buy in the Kindle store, and millions of works you can download from the Internet in which no copyright inheres, or works that are covered by CC licenses and other licenses.

    It is neither technologically feasible nor socially beneficial for technology firms to be in the position of determining, a priori, whether or not a use made in the field does or doesn’t infringe upon copyright law. Not only is it impossible for Amazon write software that can tell when a use would be held to be infringing in a court of law, but the apportionment of liability to Amazon for getting it wrong would stop technology dead in its tracks.

    Your Web browser has exactly *zero* in-built features to reduce copyright infringement. It does not detect copyrighted works. It doesn’t attempt to stop you from printing them orhaving them read aloud. Your email client has exactly zero in-built copyright protection features. Your phone, your camera, your VCR, your photocopier, your scanner — all these general-purpose devices that are capable of infringing upon copyright (that are used every day to commit billions of acts of infringement!) are totally free from an obligation to provide for means to detect and estop infringement.

    This is a good thing. It means that engineers and inventors and innovators don’t have to ensure that everything they make is capable of being used only for good. WIthout this principle — embodied in US common law since the Supreme Court ruled on Betamax in 1984 — no one could have built and capitalized on the Web, email, phones, etc.

    The parallels in the Betamax decision are deep here: Sony advertised its VCRs as being capable of performing a number of functions, including:

    1. Time-shifting, and;

    2. Librarying

    When the Supremes ruled in Betamax, they held that only time-shifting was unambiguously non-infringing, and that “librarying” may indeed infringe upon the copyrights of the studios.

    Nevertheless, they said that so long as the device was “capable of sustaining a substantial non-infringing use” it *did not matter* if it could be used to infringe. It *did not matter* if the device was advertised as suited to an infringing purpose.

    All that mattered is that the device is capable of sustaining a substantial non-infringing use.

    The studios argued then that Sony could prevent infringement by adding “features” such as the detection of an analog watermark that said “No copying allowed”, and the Supremes ruled that Sony had no obligation even to tweak its featureset to reduce the likelihood that its VCRs would be used to infringe.

    Is there anyone who will say that the Kindle’s TTS cannot be used in a substantial, non-infringing way?

  26. It is time to Just Acknowledge the doctrine of first sale and private use.

    It is just Modern technologically informed common sense expectations.

    It should be legal to privately rip tracks of a CD, transforming/converting them to MP3/Vorbis etc, to listen to on your computer/media player.
    You should not have to re-purchase the same music in mp3 formats that you have on CD just because the copyright holder says he loses money.

    Viewing the output of a visual effect generated from that music ( ie Gloom in totem/rhythmbox ), which is just another transformation, should
    be legal.

    Transforming text to speech by use of a automated mechanism/program should be no different than any other form of private media shifting. You cannot redistribute the resulting output, as it is derived, but there should be NO limits to what you can do privately to the work.

  27. I think Blount’s argument contains a kernel of truth but he’s chosen the wrong target, the problem is not the technology but the agreements that authors might have with publishers.

    It may be a long way off but eventually TTS is going to be good enough that a significant number of people will choose to buy ebooks to be read aloud on their device instead of buying an audiobook narrated by a human being.

    This should be of no concern to authors as long as they get the same amount of compensation whichever form their work is sold in. I’m not privy to the rights agreements that authors might have with their publishers but if their is a difference then authors should take the advice of Neil Gaiman’s agent.

    We’ve sold audiobook rights and print book rights as separate things. We must stop this.

    I think that the only people who should be concerned by the future in which ‘audiobooks’ are all computer generated on the fly (if indeed that ever does come to pass) are the actors who work as narrators and the recording studios where audiobooks are recorded.

  28. Cory,

    I continue to disagree with you.

    Performance rights on written materials are currently considered to be different from the text.

    Time shifting and Librarying were _not_ considered to be different at the time of the Betamax decision (or now of course).

    You did, and do, have the right to time shift your video viewing. You do not, right now, have a right to perform a written work.

    Does the Kindle perform that work? It seems pretty clear that it does.

    Does that performance have value? Again, it seems clear that it does have value. Amazon is selling the Kindle in part on that ability.

    Does the person whose work is being performed have any rights to the economic value of that performance? Obviously they do.

    Does that right vanish if the performance is created by software rather than a person? I don’t think so, but that is pretty much the key question.

  29. You’re still not addressing the point. Even if we stipulate that the TTS performance is unlawful infringement in the case of copyrighted books, and that this infringement has some characteristic that distinguishes it from librarying (though what this is I cannot guess — Sony’s Betamax was competing head-on with Discovision’s optical discs that sold commercial copies of feature films for home librarying), you still haven’t answered the real question:

    Why is this Amazon’s problem?

    If Betamax says that you don’t have to build devices to prevent infringement, that you can advertise a device to promote an infringing use, provided that the technology can “sustain a substantial non-infringing use,” then why is it illegal for Amazon to build TTS into a device?

    If TTS infringes, then why is Amazon the infringer, as opposed to the reader who uses the technology to commit infringement?

    The fact that rightsholders presently sell audio rights separately does not mean that the law should step in to preserve that right.

    Imagine that rightsholders did a thriving trade in selling books that were only licensed to be read under incandescent bulbs — is it incumbent on the law to ensure that rightsholders get a regulatory toolkit to enforce that right against the makers and users of compact fluorescents?

  30. Does converting text to an image on the screen count as a derivative work also?

    This whole argument only make sense because of an archaic legal system.

    Text-to-speech is just another *interface*, to access the *same* data.

  31. @28

    Cory’s point was that even if performing copyrighted books through TTS were infringing there are thousands of books that are not copyrighted so the tecnology itself should not be blamed.

    “Does that performance have value? Again, it seems clear that it does have value. Amazon is selling the Kindle in part on that ability”

    so what? value does not automatically translate to money or infringement. the TTS is a nice feature but not the main one. the iPhone phone sells in part on the ability to play Mp3’s and can view Video and eBooks too. in the case of music it can play it aloud for large audiences. so by the same argument Apple is selling an infringing device.

  32. “You do not, right now, have a right to perform a written work.”

    I don’t have the right to read a book to my child? I don’t have the right to recite poetry to a few friends or act out a play? I think I do. Just as I also have the right to play a tape or DVD movie for a few friends. Those are all private performances and as far as I know they’ve always been allowed.

  33. According to Mr. Blount’s argument would that make Gutenberg’s printing press infringing on copyright because it provides a “derivative work”?

  34. One day they are going to outlaw eyes (or looking in general) – because you copy an image in your brain – clearly infringement if you ask me.

  35. Cory,
    first time I read “I, Robot” (your “I, Robot”) I thought that you just were exaggerating to send a warning about the abuse of some kind of laws.

    Recently, and even more after this embarrassing behaviour of Authors Guild, I think that in United States the scenario of “I, Robot” could be a real one.

  36. Someone mentioned that animation is already removing actors from movies. That’s true, sort of. However, note that while animation is removing actors, nothing is removing voice performance yet. The only time I can think of a computer creating a voice performance is MacTalk from Wall-E, and that was done intentionally to sound computerized.

    I think we have a ways to go in that department.

  37. Rich:

    Performance rights on written materials are currently considered to be different from the text.
    You did, and do, have the right to time shift your video viewing. You do not, right now, have a right to perform a written work.

    Noen’s response is exactly right.

    Only public performance is restricted.

    You can dance around your room, individually performing all the chracters in the book for all the law cares. You just can’t sell a resulting recording, or sell tickets to the performance itself.

    All of these arguments have been roundly trounced in the last TTS thread:

    http://gadgets.boingboing.net/2009/02/10/authors-guild-claims.html

  38. Cory,

    You say that I am not addressing ‘the point,’ but I say you are not addressing ‘the’ point.

    You have said nothing in response to the points which I made, instead you create strawmen.

    No one is arguing for an’incandescent bulbs’ license. For you to create that fake argument is to disengage from very real issues.

    You ask ‘why is this Amazon’s problem?’

    Amazon is selling Kindles because there is content to read on them. Much of that content is being sold/licensed pursuant to agreements between amazon and the publishers which ignore the reality that the Kindle allows for a kind of use which is different from a written text.

    The Authors of the works did not (with the apparent exception of you) grant a performance right to their work.

    The Betamax decision allowed VCR’s to be sold. That is not particularly on point to the Kindle.

    Here content is being sold by Amazon, to be used on a device, also sold by Amazon, which allows that work to be ‘performed’ in a way which is not specifically licensed by the authors of that work.

    At this point, and with this technology, I don’t think that performance right is worth all that much. But it is a right which is, I think without question, owned by the authors.

    Or do you argue that a ‘performance’ created by software running in silicon is different from a performance created by software running in meat?

  39. Animation is already removing actors from movies. That’s true, sort of.

    Remotely true (in very limited circumstances) but entirely irrelevant. Animation will never take over from live-action. There are the production costs for one thing, but also they are different mediums. Although something like Benjamin Button used a digital head for the entire first hour of the movie, Brad Pitt (and some stand-ins) still had to act the face and body, for the motion reference.

    http://www.ted.com/talks/ed_ulbrich_shows_how_benjamin_button_got_his_face.html

    We do not have the time, money or technology to make every film animated, or every actor digital.

  40. @arkizzle

    “Only public performance is restricted.”

    I strongly disagree with you.

    An audio book is not a ‘public performance,’ and yet it is restricted.

    I am confused why there is such resistance to the basic idea that a performance created by a person is functionally the same as a performance created by software.

    I don’t know what should be done about this, but it doesn’t help anything to deny the similarity between an audio book recording of a book and a Kindle real time rendering of a book.

  41. Rich, “performed” is not an issue. You can legally perform anything you like, in private, right now.

    “Publically performed” is the problem, which Kindle doesn’t (necessarily) have. Once you let go of that erroneous point, your argument falls down.

  42. No Rich, an audiobook is a derivative work in a fixed medium that is published for profit.

    You are not arguing audiobooks when you talk about performance. They are separate things legally.

  43. I am confused why there is such resistance to the basic idea that a performance created by a person is functionally the same as a performance created by software.

    You are conflating two different things.

    1. the right to publically perform a work
    2. the right to distribute a work on a fixed medium.

    There is no ‘private performance’ right, because it is unrestricted legally. You are making it up by conflating the two restrictions. You are mistaken.

  44. The accessibility argument needs to be dropped on this guy’s head directly so he can be shut down once and for all: Text screen readers are legally required allowances for the blind. This is something any aspiring web designer learns when they take dreamweaver tutorials – there is a legal requirement for websites to have accessibility content that can be accessed by screen readers for the blind (this is only typically applied to government and school websites, and what qualifies as “accessible” is subjective, but the screen reading attributes are not up for debate – blind people need them, period.

    This accessibility MUST be available to those who need it, so unless his proposal is to provide a free audio version of the book with every single copy sold – thus defeating the purpose of his battle, he will lose, this is futile.

    I’m just surprised that hasn’t been turned into the crux of the opposing argument yet, though. It’s the kind of thing that would make this an open and shut case with the supreme court laughing him off the stand.

  45. but the kindle works only with eBooks purchased from Amazon? will it display plain .txt and .doc files loaded from a pc?

  46. Blount it a nut. He’d prob be against libraries if he thought he could get away with it. Alot of people check out books on CD and don’t pay a dime there.

  47. @arkizzle

    “Only public performance is restricted.”

    I strongly disagree with you.

    Than you disagree with the copyright law. The copyright law provides to creators certain monopolies. Among them are the right to make copies, the right to create derivative works, and the right to public performance.

    An audio book is not a ‘public performance,’ and yet it is restricted.
    An audio book is a copy of a derivitave work. PLAYING an audio book is a performance. Even if you have purchased an audio book you do not have the right to perform (play) it publicly, or to copy it.

    What poster #13 said about English law stands for U.S. law as well. TTS constitutes a performance of the underlying text, no different than reading the text aloud. Now recording or publicly performing of either would probably be an infringement.

  48. IMHO, the argument boils down to format shifting: is it allowed, or not? Publishers hate format shifting because they are set up to profit from separate formats. Customers like format shifting (in as much as they know what it is) because it allows them to be flexible with the thing they just bought.

    My opinion: publishers need to get with the times. They must sell the rights to the information, not the rights to the format of that information.

  49. I wonder if Blount’s chagrin comes from the amount of $$$ he makes from his voice as well as from his writing. This seems so odd to me, but then again, it is from the man who wrote: “Me, I like my oysters fried / Cuz then I know my oyster’s died.” I hope I didn’t violate anything my quoting him here. He just lost a fan in me.

  50. Rich Gibson said, “Or do you argue that a ‘performance’ created by software running in silicon is different from a performance created by software running in meat?”

    I think there is a distict differenct, for now.

    I’m currently listening to Stephenson’s Anathem, which is being performed by some exceptional reading talent (a great book to read or listen to). Will a device like Kindle compete with humans reading their parts with inflection, accent, emotion? No. Stephenson didn’t even allow for that kind of reading talent in the Diamond Age. And as Cory said, there is already ‘ware that does this for the blind, and no one is screaming bloody murder over copyright infringement (or maybe they are and I don’t know it.) I think we need to talk about copyright issue with regard to problems that could emerge with our expectations of increasing technological sophistication. I agree that our semantic content with regard to these issues needs to be sorted out. Right now silicon is different than meat, but that may not be the case for long

  51. Publishers of audio-books really can’t think much of their product if they think text-to-speech is a valid competitor. I’m trying to imagine a computer programme matching Patrick Tull’s reading of the Aubrey/Maturin novels, Rob Inglis’ Lord of the Rings or Stephen Fry’s Harry Potter… It’s a bloody joke.

    I love good audiobooks because they are a performance of a work, as infused with added meaning and interpretation as any stage production – they give me something of the giddy expectation I so clearly remember when I was 3 or 4, waiting for my Dad to come home to read me my stories. No robotic monotone is going to replace that experience anytime soon, and the argument that it will is doing the audiobook industry a grave disservice.

  52. Cory,

    If Sony had bundled every Betamax with blank tapes and a copy of TV Guide with all the movies tagged, would Universal have prevailed?

    Part of the problem here seems to be that the Kindle’s business model makes Amazon the default beneficiary of any added value from the “reader” feature.

    It doesn’t seem quite so open-and-shut as you suggest.

  53. If owning a device capable of committing copyright infringement makes you guilt of copyright infringement, then the publishers who own printing presses are guilty of copyright infringement. So why not just shut down all forms of publishing, so that nobody commits copyright infringement? That’ll solve the problem.

  54. @44 Text readers for the blind are already granted legal exceptions. This has nothing to do with that. The blind reader argument is a strawman.

  55. Blount is a modern day John Phillip Sousa, railing against “those infernal recording machines”.

    Righteous screed, Mr. Doctorow.

  56. The text below is a copy of a letter to the editor written this am to the NYT’s.

    I got home last evening to find my new Kindle reading device delivered by UPS hours before.
    I ordered it to replace my existing Kindle because it had two features, clearer text resolution and atext to speech function. I upgraded because as a person with low vision I am constantly on thelook out for tech tools that support my deep desire to read and stay connected to the world.

    But earlier in the day I was not so pleased. In fact yesterday morning I was extremely upset
    and angered by the op ed piece published in this paper and written by Roy Blount. Mr Blount is the
    President of the Authors Guild (a writers union) and was using the pages of the Times to
    express his Union’s distress over the fact that Amazon had made this text to speech feature
    available to consumes without first negotiating with his union. Wait you say, the consumer has to
    first purchase the electronic text, yes? Yes. The union’s argument is that millions of new
    consumers will buy Kindle’s, and then buy the electronic text’s, but then somehow the author is still being cheated because there is added value, the new text to speech function, that is not accountedfor in the existing royalty agreements.

    What evidence does the President of the Union, Mr Blount, site for such an assertion? Well, he
    tells us, I have a friend, yes, Mister Southern Good OL’ Boy, Mr Wait-Wait Don’t Tell Me, I have a
    friend who once worked for IBM and he tells me that these computer voices are darn close to the
    real thing! That’s his evidence. Mr Blount is playing on his public persona as a trusted souther
    gentlemen and he references his father at the beginning of his editorial, to communicate to the
    public that, well, he’s just on the side of the poor writers struggling against the big bad corporation, Amazon,out to extract every dime of surplus profit that it can. This is the essence of the editorial.

    Well let me tell you and your union Mr Blount that this is not another David and Goliath story. You have no idea how upsetting your editorial was to me and I’m sure to many others who struggle every day to read. If there is a David and Goliath story embedded in the use of the Kindle it is the story of those who struggle every day to read and thus stay connected to the world, and those who take sight for granted.

    I will get to the substance of the argument in a moment, but I first want to let
    Mr Blount and the Times know how this article made a lot of us feel. Awful. And there was
    absolutely no consideration shown to this fact yesterday either by the New York Times or the
    Authors Guild. I made several attempts to discuss my feelings about this piece with the Editor ofthe Op Ed page, to no avail. I called the Authors Guild in New York City and spoke with theExecutive Director having been told that Mr Blount was not available AND that no one at the
    Guild knew how to get a hold of him! Needless to say my anger was not assuaged.

    Mr Blount, you have no idea how your article, your attemtp at collective bargaining on the Op Ed
    page, effected me and I am certain many others who depend on a variety of devices to keep us
    connected to your words. You should be ashamed. There is not an ounce of awareness or
    generosity in your words.

    Your argument about the text to speech function on the Kindle is illogical and irrational. You areplaying the populist card, and also playing on the ignorance of the general public. Most people have never consciously used such afunction, ever, except for the thousands of times each year when we make a phone call and hear the usually female voice tell us the dialing menu for the phone company, the cable company, city hall, the doctors office, the retail outlet, or any of the hundreds of other opportunities afforded us to interact with a computer voice. Based on absolutely no scientific evidence you’re claiming that these “voices” represent a breakthrough in the human auditory experience such that when people use a text to speech function you should now be paid extra. Your friend at IBM tells you that why it’s not much different than hearing the real person apeaking. Is it live or is it Memorex? And because of this you and your union want all the prospective low vision users of the Kindle to register with a national data base for the blind in order to be able to use this function without charge.

    Well Mr Blount anyone who actually uses these devices because they have macular degeneration,
    retinities pigmentosa, glaucoma, diabeticho onset blindness, or any of the other myriad eye conditionsthat make reading difficult or imposible will tell you just what they think of your ridiculous argument. The Executive Director of your Union yesterday was totally dismissive of my maincomplaint, that neither your nor anyone at the Times had any appreciation for how the visually impaired would respond to your piece, again tried to make the argument that his union members would be economically impacted in a negative way because people would BUY THE E VERSION OF THEBOOK, but if they happened to use the text to speech function that value would not accrue to theauthor.

    He went on to claim that many people would use this feature, even those who are not blind,
    and that the people who do use this function because this is the only way they can “read” the book,well, they can register with a national data bank, and then get permission to use the “text”
    just purchased, without paying the additional fee.

    So, all you folks with low vision excited about the prospects of using your text to speech functionon the device that you bought for several hundreds of dollars, probably for this purpose, the Authors Union wants you to register and get permission to use the device.

    Nice of them. Mr Blount, for years I was a member of the NationalWriters Union, now an affiliate of the UAW. I respect unions, but on this one you folks are dead wrong. If people want to listen to an audio book they’re downloading the file from iTunes and you know it. This is a disengenuous argument made in a prestigious national publication; you are bargaining in the media. And in the process you have upset potentially millions of people for whom this issue is not just another part of collective bargaining. It is our lifeline to the world of words and you and your Union should apologize.

    And by the way Mr Blount, I’m sure your father taught you something about kindess and geneosity. Perhpas that’s what you should have remembered.

    Russ Eckel

  57. All they have to do is create files for Kindle, etc. that disable read-aloud features for authors/publishers who don’t want their works read aloud by a machine.

  58. Robert:IMHO, the argument boils down to format shifting: is it allowed, or not? NO. That is what the Authors guild want’s you to believe. If this WAS about the creation of an audio copy (which would constitute a derivative work) they would have a leg to stand on. But I believe that no copy is “fixed” by the TTS software. Therefore, the TTS merely “performs” the text. Said performance is infringing only if it is public. Recording said performance WOULD be infringing, but if the Kindle has no capability to do so, I’m not sure that the existance of TTS software is relevant.

  59. As a ‘content creator’, I have mixed opinions on this…regardless of the law, I believe there is a moral right of any artist to protect his works in any way they feel fit — even if the consumer of that product doesn’t like it…or it makes no gatdamm sense at all because we are created it and that should be enough.

    On the other end, yeah…I’ve bought several ebooks over the years, and as I have a few 12+ hour trips in the near future (starting next week), I’ve ripped the DRM from several of these to transfer to my iPhone, and then realized after hearing about this bitching last week, I thought “Hey I Can Convert These To Audio”…doing a history on my terminal window:

    say -v alex -f /Users/ccmarsig/Desktop/mccarthy_theroad_text.txt -o theroad.aiff

    Sounds like crap, but it works for what I need (i.e., motion sickness on planes keep me from reading for more than an hour)…but what about the future? These speech to text apps can also run with a score, changing inflection and voices per the speaker. What if others start selling companion scores that rival the VO artists? Would that fall under fair use? The vocal proponents of FU would say so…then again, there is little they won’t say is. Same as the opponents…there is little they claim is actually covered by it. It is a shame these discussions aren’t being conducted by the rationally minded as opposed to partisans willing to wage out and out war without even realizing how far to the fringes they are.

    Anyhoo…I see the Author’ Guild’s point and it is a legitimate complaint…but so is the need to move on with reality…

  60. @ Fuzzy

    thanks for the answer.

    i did read the wikipedia entry, the wired review and the official kindle page at amazon but for what i understand this thing gets content exclusively over the air.

    maybe it’s the hottest feature but since it does have a USB port surely you can upload text (regardless of format)from a pc, right? if not it should be called the Swindle more than the Kindle.

    i was considering buying one but i am not interested paying 10$ for an ebook (at least i would not buy that many to justify the hardware), and i have several gigs of ebooks that i would like to read on something bigger than my compaq pocket pc.

    Sorry for the OT

  61. All legal arguments aside for a moment, isn’t it just a little bit sad that, assuming the war on copyrights is successful, the day when one can no longer make a living as an author seems very, very close?

    Perhaps in 20 years or so, only the independently wealthy will write books. Or, I guess, there could be books “commissioned” by modern-day Medicis, but probably darn few, since there would be no profit in it. Unless, maybe, books will start to have a lot more product placement? Interesting times.

  62. I think the Author’s Guild is within their rights here, and the NYT article certainly isn’t the sort of hysteria you’d expect from the RIAA. (It’s perhaps worth noting that the lovely, decidedly un-venal folks at indie press Small Beer, publisher of Kelly Link and Benjamin Rosenbaum, came down on the side of the Guild. These are wonderful people, strong supporters of Creative Commons, etc.)

    The fact on the ground here is that an ebook is more versatile than ever, potentially subject to new commercially significant derivations through technology such as TTS. My stance is that authors aren’t wrong to raise this issue when negotiating with their publishers over electronic rights. If I manage to sell my novel, I’d vastly prefer to take the same route as Cory or Kelly over at Small Beer — give the ebooks away. I think that’s smarter. That said, authors aren’t unreasonable to treat ebooks as a dynamic medium and to expect reflective, dynamic compensation.

    However. Going after Amazon is misguided (BoingBoingers helped persuade me of this last time the issue came up), and Cory’s right that any wrangling with the technology itself is ridiculous. This is something for authors and publishers to hash out.

  63. I think that this will start (kindle perhaps? :P) the development of the first AI.

    If this sounds far-fetched then think about it: They will now start developing better and better text-to-speech systems, and to finally compete with audiobooks (because that IS what they are aiming at, even if they don’t know it yet themselves) the computer will have to understand the text on a certain level. This will, in a way they don’t see themselves yet, lead to the development of the first true AI. From making a “darned perfect” text-to-speech system that is capable of actually understanding the text (and if you really want it to sound like a human they will see that they can apply the base system of their text-to-speech system to understand other types of semantic sets. They should just stop trying so hard theoretically, and start somewhere practially, and the Kindle 2 is now the ideal starting point. It’s how captialistic technological advancement works, and it will work this way. Looking forward to talk to a computer in the year 2045!

  64. For those of you who are arguing that the Kindle is — in the legal sense — a performer:

    You are wrong. Under US law, only a human is a performer. Tools lack the legal standing, as capable of making choices, to be classed as performers. They’re tools.

    The Kindle is a tool. The human using it is producing the text-to-speech performance. They are the performer.

    “Software which reads your email does not compete with an audio book. Software which reads your books on a Kindle does compete.”
    –Wrong, and irrelevant besides. The software does not compete – the human using the Kindle “competes” with the audiobook.

    “The Kindle is in part sold on the ability to read to you. That means it is ‘performing’ the work.”
    — No, it means it is converting text to speech at the behest of the operator, who is the performer.

    ——–

    Cosmicdog had an interesting question i want to address: “But how will software understand the emotional context of a scene, when to whisper, when to shout?” –

    I will address it by rewriting it: But how will text on a page understand the emotional context of a scene, when to whisper, when to shout?

    The answer is neither of them understands the emotional context of a scene, neither practically nor in a legal sense (standard disclaimer: I am not a lawyer). Human beings read text off a page — squiggly glyphs arranged in a geometric relationship — with their eyes and use it as the basis to construct a (sometimes to-them-imaginarily-audible) linear-time narrative that in turn is used to construct imaginary people, events, lands, notions, technology, clothing, architecture, and on and on. The listener infers the meaning of — exegesises — the text in their brain, not in the visual stream, not in the glyphs, not on the page, not in the book. There are today — and may be more tomorrow — plenty of people who can be read to by an uninflected or machine-produced voice and use that as the basis transmission from a storage device to their comprehension.

  65. Since this guy is suggesting laws to prevent an eventual use of superior technology, why stop short?

    Better outlaw the upload of books directly into the users’ brain, because in 100-200 years that will be possible and it’ll ruin the ebook industry.

  66. I think the only way one can accuse a technology such as speech-to-text of copyright infringement is if the entity using said technology was an actual legal entity.

    Perhaps if Mr. Blount starts a robot rights movement, and successfully changes law. Then he may be able to successfully sue any company that manufactures robots that can potentially read his book.

    You are downright silly Mr. Blount.

  67. You are wrong. Under US law, only a human is a performer. Tools lack the legal standing, as capable of making choices, to be classed as performers. They’re tools.

    Let’s check back in 50 years when we have the first AIs which developed in the way I described above, and then we’ll see whether they will be classified as performers (i.e. as beings equivalent to a human being in their rights). :P

  68. Blount has obviously reached the “you kids get off my lawn” part of his life cycle. It seems that he (and his supporters) have some fundamental misunderstandings about technology.

    But let’s assume — for the sake of argument — that TTS could render a performance just as good as a professional reader. As a content creator myself, my response is: So what? Technology moves on, it creates new art forms even as it diminishes others. That’s history. Maybe human-performances of audiobooks will become as common as original plays in iambic pentameter.

    I don’t believe that the law has a responsibility to ensure that my chosen art form is economically viable. At some point, the Author’s Guild might just have to give up on the idea that they can monetize audio recordings.

  69. Until the Amazon sells and the Kindle supports something long the lines of “voice packs” so you can listen to the ebook you just purchased read to you by someone that sounds like a real person described in the voice pack, Mr Blount can just shutup.

    Hmm maybe that is an idea you author types can take to Amazon….you know a Kindle 3.0 feature :P Do a deal and get a bigger piece of the pie, rather than just sit around and complain.

  70. You mean like professional writers can make a living now? You mean professional writers aren’t considered upper class now?

    Writing has and always has been the occupation of the rich and idle. Most writers have to slave at wage a day rates as security guards, teachers, etc like every other artist in town.

    Copyright isn’t going to change anything. You think that extra 300$/year from royalties is REALLY going to bring you into a new tax bracket? Think again.

    Eliminating copyright and mass copying means that your work is more easily distributed and you can bypass publishers. Patronage for the arts hasn’t changed since the age of Medici. You’re still spitting out derivitive works vetted by upperclass nobility. The only difference is that their titles fall more along the lines of “Chairman” or “Shareholder” rather the Don or Count.

  71. @Rich Gibson:

    Software which reads your email does not compete with an audio book. Software which reads your books on a Kindle does compete.

    […]

    The Kindle is in part sold on the ability to read to you. That means it is ‘performing’ the work.

    The legal question is at best ambiguous, but as I read it his point is pretty strong.

    The value in audio books over just the text of the book comes from the performance of the actor reading the book, not from the writing of the book. If you’re not utilizing the services of the professional reader, isn’t that fair competition? Sucks for the pro that his job was replaced by a machine, but there’s a long line to wait in if you want to complain about that…

    (It’s amusing to me that you try to make this argument to Cory, though, since he seems to make a good living through live “performance”. I suspect you’ll have trouble drumming up sympathy from an actual working artist when proposing a protectionist system that lets people rest on their laurels.)

    That is fine. Nobody has a guaranteed business model, but under our legal system authors have certain rights. Kindle 2 does things with text which most authors did not agree to when they licensed their work.

    That’s the same argument as the DRM argument that led to the DMCA. The problem with that argument is that the authors didn’t hold that right to restrict in the first place. The public graciously grants them the right to restrict public performance. This is *not* public performance. Not even close. Let’s hope they never expand the law. It would suck to have to pay royalties to teach our children to read.

  72. While we are at it let’s be sure to ban lightbulbs,
    which clearly infringe on the rights of the candlemakers guild.

  73. I haven’t finished reading all the comments here, but I REALLY need to jump in on the “Perfect text to speech = computer animation replacing live actors” thing before my head pops off with outrage.

    That’s a false analogy.

    Computer animation replaces one performer (an actor) with one or more performers (the guy who sculpts the computer model, the guy who skins the model, the guy who animates the model, the guy who voices the character represented by the model… any of which could be a team, not a single artisan)

    This is how you can replace an actor with a computer animation – HUMAN BEINGS (often many) use their creative energy to breathe life and believability into the character, to enact and react to the script realistically. The only reason why replacing all of hollywood with computer animation could even be considered is because, for the moment, none of these people have “star power” and therefore can’t charge the same wages as Tom Cruise. If you eliminate the actors, the public will probably start paying more attention to the animators and voice talent, and studios will start marketing them just like face talent, and lo and behold they’ll be expensive again.

    The most perfect text-to-speech program in the world, like the most realistic computer rendering in the world, will still need a human being to tell it when to be tense, when to be worried, when to be happy, when to pause dramatically, when to froth with rage and chew scenery…

    None of which is in an e-text.

    Until we get computers that can not only fully comprehend natural language, but ALSO human psychology, AND THEN can learn to act with the same skill as a professional actor, it’s just not going to happen. This is not happening Real Soon Now or even in my lifetime, no matter what the singularitarians say.

    When it does happen, it will still be a while before it’s cheap enough to be worth jamning in an ebook reader, presuming the debate over audio performances are still going on in that distant time.

  74. I’m disappointed that the strong supporters of the Authors’ Guild position (Rich Gibson and eicos mainly) seem to have dropped out of the discussion silently after their arguments were decisively rebutted.

    I think there’s the potential for some genuine learning and synthesis here, and it’d be a shame to end the discussion simply because one’s initial position turned out to be untenable.

  75. I think that people are really missing the most interesting part of this argument, although it has been referred to. I am surprised that more media and other people are not focusing on the ADA issues that could be addressed here.

    If Kindle’s TTS could eventually be “performance-equivalence”, then why should a sight impaired person not be allowed to listen for the same price as everyone else’s book.

    Currently, the price for an unabridged audiobook with professional readers through Amazon for the Harry Potter 7 book is $50 on audio, $10 paperback. If a non-wealthy, sight impaired, person wants to buy a copy of the Kindle version for $10 and then listen to it on that device, then why shouldn’t she? Why shouldn’t she have a really good TTS reading, instead of the mechanical voice? What’s wrong with that, Author’s Guild?

    This seems like an issue that various ADA groups would want to address. Also, in the event that some horrible “registration” process comes into play like someone in a previous comment suggested, there will definitely be an increase in “blind” people and “ripped” books :)

  76. Very odd argument. Is he also railing against screen readers for accessibility purposes? As was stated above, there is a HUGE difference between an computer generated voice and a voice actor.

  77. “What is this “book” thingy you’ve “written” here with MY Iliad and Odyssey supposedly somehow stuffed inside it??!!! I never meant for anyone to trap my words with dirty, physical objects! Ye can all bloody well learn them by heart and recite them properly or I’ll have the dogs on ye! Now get off my damn agora!!”

  78. Bardfin: You’re right, the Kindle isn’t a performer. But using it IS a performance, and does not, in itself create a copy, and therefore does not create a derivative work.

    I think that it is important to allow content creators to set up tollboths. But just because changing technology enables people to bypass barriers that used to enable them charge seprately for audio and text does not grant them new rights under copyright law, anymore than it automaticly authorizes people to violate copyright law. Yes, TTS DOES mean that authors should reconsider what they charge publishers for electronic rights. And there is argument for changing the law to reflect the state of technology. But those don’t seem to be the argument that they’re making here.

  79. @takuan

    Homer was blind so maybe he would had hated the dirty ink but i believe that he would appreciate the TTS feature.

  80. This “copyright infringement” bugaboo is misleading, and the Author’s Guild is not trying to silence the Kindle 2. Rather, the Author’s Guild has a true mission of ensuring that authors still get paid for the audio version of their work.

    A theoretical program that has audio-book-quality TTS software is a risk to an author’s ability to sell audio books of their work — Amazon effectively gets to make an audio version for free. Framing it as a copyright debate is immensely boneheaded on the part of the Guild, because this is more of a market issue. High-quality TTS in the future might rob an author of their ability to get paid for an audio version of their book.

    Is that a good thing? A bad thing? That’s a more prickly issue. The Author’s Guild isn’t interesting in shutting down the Kindle 2’s feature. They’re just interested in making sure the author gets paid for the audio version of the book — Amazon right now doesn’t pay for that version, just the e-book text, and the Author’s Guild is mostly concerned about making sure that Dave Eggers (for instance) gets paid for audio copies of WHAT IS THE WHAT.

    And this is coming directly from the Horse’s Mouth, so to speak — my roommate is currently employed at the Author’s Guild (she’s probably the one fielding your angry phone calls and e-mails) and from their position it certainly sounds more like Amazon trying to get audio books without having to pay writers for them rather than a luddite crusade (they actually don’t want to cancel the audio feature, they just want Amazon to pay up for the audio rights that they don’t actually have).

  81. It sounds to me as though Mr. Blount is worried that he won’t get paid twice – once for the audio book, once for the “text” book. If he performs his own audio books (and it looks like he has done half of them), or picks a good performer to do his audio books, that should be incentive enough for some of his readers to also become listeners. The people who buy audio books likely would not use the text-to-speech feature on the Kindle in the places and for the reasons they’d buy audio books.

  82. @ #35 posted by thekevinmonster

    Someone mentioned that animation is already removing actors from movies. That’s true, sort of. However, note that while animation is removing actors, nothing is removing voice performance yet.

    Jerril beat me to this in #73 but I wanted to reiterate: Animation does NOT replace human performances with mechanical ones. Animation pools the creative talents of many human beings into one performance. Computers are wonderful tools but actors they are not.

    It saddens me when I see someone I otherwise admire (Roy Blunt Jr.) making such asinine arguments. Even if we did accept that this technology falls under the category of “performance” it would qualify as a PRIVATE performance, no different than if I hired Morgan Freeman to follow me around and read me books from my library whenever I felt like it.

    Oh shit, is that illegal too? I’d better let him go just in case.

  83. Forget the Kindle– I want that IBM software Blount mentioned that tells you to STFU if you interrupt it. Somehow, having a program that you can annoy into a fit of anger seems highly entertaining.

  84. @#80
    “It sounds to me as though Mr. Blount is worried that he won’t get paid twice – once for the audio book, once for the “text” book.

    Well, that has been the way a writer makes a living for the last 80 or so years: they write, and they get paid when their writing is used.

    Actors make money every time their names run in the credits, even years after doing the performance.

    Musicians make money for recordings of their songs, and for those songs to be played live, and every time those songs appear in a commercial.

    The market is going to change, but you can’t really fault the AG, in my mind, for wanting writers to get paid for an audio version of their book — whether recorded on an MP3 or just read by a high-quality TTS program that makes the MP3 irrelevant.

    I think the AG is kind of backwards on this, myself, but they do have an slightly understandable position. They aren’t the RIAA. They’re a tiny little nonprofit operating to try and prevent authors from being taken advantage of when companies (publishing companies or Amazon) try to use their work. That’s an admirable goal, even if, in this instance (and in others) it might be dunderheaded of them to frame it as a copyright issue.

  85. During Bezos’s interview on the Daily Show Monday he said that the DRM on the books was only if the publisher wants it. Not sure if that extends to Audible.

    He also said the text to speech sound like a weird robot, so they won’t be burying the audiostore with this version of the Kindle.

  86. if one is trying to make a living by writing, what is the better investment of time: writing more, or chasing what you can squeeze for what you have written?

  87. “High-quality TTS in the future might rob an author of their ability to get paid for an audio version of their book.”

    Ya know what? I’m starting a guild. The Bard’s Guild. Membership is open to anyone who’s orally told an original story or sung an original song to anyone else, ever, in their lifetime, from out of their own imagination.

    Our first order of business will be to make the talk circuit, where we demand that every extant text and recording be accounted for, and royalties paid on them, or they be destroyed.

    Our next order of business will be to sue all scribes, dictionarists, and typesetters for competing with our business model.

    Glyphs on a page! Those can be used to create a private performance of a copyright-protected derivative work of our original works of story and song! Someone might take notes regarding our original performances!

    What Kind Of Bard Can Live While UnTrained Common Folk Have Quills and Ink?!?! The Poetry They Might Compose! The Quality of Discourse! Won’t Someone Think of the Quality of Discourse? Our Livelihoods! We Shall Enact Laws!

    — Once upon a time, in the British Isles, there was a class of performers, known today collectively as bards. They alone learned the rules of composing spoken and sung and music performances. They alone knew and set the laws of the land. They were semi-aristocratic and existed outside of the rule of royalty. They were privileged. Eventually, scribes learned lettering, printing presses made publishing capable, and people learned to read. Their privilege dissolved. Technology democratised the expression, utility, and enjoyment of expressed human thought.

  88. Takuan #86,

    if one is trying to make a living by writing, what is the better investment of time: writing more, or chasing what you can squeeze for what you have written?

    In the usual common law framework of copyright, authors don’t have a natural right of ownership over their work. They get a limited exclusive right, at the expiration of which the work returns to the public domain which inspired it.

    All authorial rights, therefore, are the products of negotiation. Mr. Blount may be a tactless negotiator, but he’s trying (I assume) to do well by his constituents. I’m sure he’d rather be writing for pleasure, but he took the job in good faith (I assume) and that’s the spirit in which I take his complaint.

  89. I’m sorry, those of you who have children, but you have committed an illegal act in creating something that is capable of violating copyright.

  90. “if one is trying to make a living by writing, what is the better investment of time: writing more, or chasing what you can squeeze for what you have written?”

    Ordinarily, the former. But nowadays the latter, since the very concept of being paid for one’s work is threatened. Volume of work won’t earn you a dime if anyone can get it for free. “Making a living by writing,” at least in the traditional sense, will not be viable for much longer.

    Whoever said above that writing is already Medici-driven or whatever is correct that it’s mostly a patrician occupation, but is still fundamentally incorrect in that publishing is, indeed, a business and and not a charity.

    (Now, mind you, I’m not siding with the author’s guild, here. And I absolutely agree that the law should not guarantee that any particular artform has a viable business model, or for that matter that any type of work remain profitable.)

  91. I’ve been producing audiobooks for 5 years.

    It will be a long, long, long time before text-to-speech can replace real live people with the ability to create characters, parse complicated sentences (phrases can mean very different things in different contexts), pronounce words correctly (lots of words are pronounced differently in different contexts), and generally convey the emotional quality of writing. I’m not worried!

    BTW, audio rights are usually controlled by the publisher… i.e., the author, being a poor artist, sells all the rights for a lump sum (hardcover, paperback, audio, film), then the publisher auctions them off — meaning, usually (I’m talking about the mass-market industry standard here, not indy stuff) that authors don’t get any royalties from audiobook sales. Nor do the narrators who add value with their performances. This is something that might be examined and changed.

  92. #87 Takuan
    “what’s Harlan Ellison said?”

    Judging by just about every video I’ve ever seen of Harlan Ellison, something very assholey.

  93. Daedalus @83: You’re making a false analogy.

    What if I started a service in which I read your book aloud to you from your library? Would the Writer’s Guild expect me to have to give them a cut for each reading? What if it’s you’re spouse?

    It’s more than fair for everyone involved in the production of a book, or in the production of a work derived from that book (including the author of the original book), to be paid for their work. What’s not fair is to sell someone a book, tell them they have the right to read it any time they want, and then later say “oh, but if your spouse reads it out loud to you, you will have to buy the book again in the official ‘spouse reading format’.”

    I’m sure you’re already thinking of performances, but this scenario is a very different thing from a public performance. The audio book you’re buying isn’t a production of the author – it’s a production of the narrator, whose work is derived from the work of the author. The narrator gets paid, and the author gets a cut because the thing the narrator was selling – ok, it’s complicated by the fact that an author has moral rights to be paid for performances for which the performer was not paid, but that’s because what makes the derivative work a derivative work is the act of interpretation.

    The Official Spouse Reading Format Hardcover is not a derivative work, in that you and your spouse aren’t getting any added value above and beyond the standard non-spouse-reading hardcover.

    The principle here is that you only pay more when you get more. If you get a hardcover instead of a paper back, you get more. If you get an illustrated edition, you get more. If you get an audiobook performed by a human narrator, you get more. What additional value do you derive from the Official Spouse Reading Format Hardcover?

    The argument being made by the WG is that the text-to-speech instantiation of the book is creating a derived work, for which someone is getting paid, and that they deserve a cut of that derived work. The reality is that the text-to-speech instantiation of the book is actually a direct product of the book itself, with no interpretation or other intervention – it is merely a mechanical instantiation of the text. The text to speech in this case has the same mechanical relationship to the text that an Itty Bitty Book Light has – it helps you to experience the text in a place you wouldn’t otherwise have been able to experience it.

    Sure, Amazon is profiting from the feature – in the same way that the Itty Bitty Book Light folks are profiting from the books being purchased. They are not profiting directly from any individual work, and they are not adding any new creative content for which they are being paid.

    So, should the library have to buy a special edition that includes licensing fees for multiple readers? Should a coffee shop have to pay a fee for leaving books out on shelves for their patrons to read? Or are all of these, like the Official Spouse Reading Format Hardcover, included in the rights you obtain by purchasing a book?

    And should the Writer’s Guild be extracting a royalty from the Itty Bitty Book Light Company for every book that’s sold in a store that offers Itty Bitty Book Lights?

  94. I don’t know if weighing in now makes a difference. But, it would seem that there is broad agreement here that contemporary copyright law and conventional agreements with publishers and distributors are flawed, out of touch with contemporary technology, and failing in their mission of ensuring that creators are fairly compensated for the work they do. (I probably could have left that comment on 15% of the posts here.) In the face of this, there will be plenty of graceful and graceless efforts by those in the culture industry to make sure they get their share in the future.

    In the sense that Mr. Blount is arguing that authors should get a fair share of money made from their work, I agree. That’s how we roll in Capitalism. But the nature of this argument casts him in the role of a luddite.

    Ultimately, it is in our best interest to foster creativity and help the next generation of artists to build our culture. Given the liquidity of contemporary media, I suspect that per-sale royalties will be obsolete soon, if they aren’t already. We need to ask ourselves how our cultural consumption can be directed properly to support the next Isaac Asimov, James Joyce, or .

  95. I want to give Mr. Blount the benefit of the doubt here, but I’m having a lot of trouble. To me, as a visually-impaired person, it feels like he’s pitting his livelihood against my ability to access books. To his credit, it appears that he’s given at least cursory thought to accessibility. But his quick dismissal of the Kindle’s potential as an accessibility tool (“blind readers can’t independently use the Kindle 2’s visual, on-screen controls.”) shows me that he gave it just enough thought to craft a glib and facile dismissal.

    I admit, I have a considerable bias in this discussion, but I believe that accessibility trumps the other issues in this discussion. Businesses are required by law to make themselves accessible, even if it costs them money to do so. TTS may cost the Authors Guild or it’s members a piece of their income, but like wheelchair ramps and special parking spaces, that’s the price of doing business in a conscientious society.

    Instead fighting against this technology, I would encourage Mr. Blount and his organization to examine it for new opportunities. Improved technology has new costs and new benefits for everyone. To use a folksy turn of phrase (I’m sure Mr. Blount would approve), “When God closes a door, somewhere he opens a window.”

  96. #95. If he was argueing that the Kindle’s capability meant that authors should rexamine royalty rates for e-books he wouldn’t have been called a Luddite. If he argued that copyright law should be changed to reflect the traditional boundaries between audio and text heretofore enshrined in publishing contracts, he’d still be called a Luddite, but at least there would ba a basis for discussion.

    But by espousing the idea that copyright law DOES prohibit the TTS function of the Kindle, he shows himself to be willfully ignorant and deserving of our scorn.

  97. Greta, I agree, there is no danger of humans being replaced for quite some time. I’d give it as least five years. Things really do need to change with regard to contracts, because as you stated, the publishers are really acting greedy.

  98. @Daedalus

    Does an author have a right to be compensated in any of the following scenarios:

    * I do a statistical analysis of the words used in a book and create an abstract piece of art — one where color-coded squares correspond in size to how often a particular word appears, for a Kandinsky-like appearance.

    * I take a book, cut it into pieces, and re-arrange all the words into a new text using only those words originally in the book.

    * I use the title of a previously published book for my own book.

    * I am inspired by a concept mentioned in a book to expand upon that concept in a work of my own.

    It can be argued that all of these “derive” from the original work, but just because the Author’s Guild might want to pursue compensation in these scenarios doesn’t make it any less silly.

  99. PTROURKE #94,

    The text to speech in this case has the same mechanical relationship to the text that an Itty Bitty Book Light has…

    Does the IBBL harm the market for any book, anywhere? Blount’s somewhat hysterical point is that the text-speech function of the Kindle could reduce demand for traditional audiobooks. This makes sense to me, in the absence of proof either way.

    Obviously text-to-speech isn’t a good substitute for a real audiobook, now or in the foreseeable future, but it’s not hard to imagine that some people, for poverty or convenience, might just take what’s given rather than purchase a separate product.

    Just a few people, probably…but if Amazon is gonna sell this thing with an umbilical link to their marketplace they should acknowledge the point, and probably give the authors their due.

  100. It will be a long, long, long time before text-to-speech can replace real live people with the ability to create characters, parse complicated sentences (phrases can mean very different things in different contexts), pronounce words correctly (lots of words are pronounced differently in different contexts), and generally convey the emotional quality of writing.

    I predict that computers will start doing this well around the same time that they begin writing novels for themselves.

    Blount of all people should know that a computer reading is no replacement for a human performance. People line up all over the country to hear him read excerpts of his work in that sweet southern drawl.

    The only exception to this rule I can think of would be an audio book by (warning: tactless joke ahead!) Stephen Hawking.

  101. Hang on…isn’t Kindle horrible, though, because it’s just Amazon trying to get in on mass market ebooks early so it can have a nasty vertically integrated business model a la Apple with ipod/itunes?

    Or has this been addressed?

  102. This is stupid. Even with significant advances in text-to-speech technology, the Kindle 2020 still isn’t going to be equivalent to a good audio book, certainly not a reading of a work of fiction.

    A good audio book is a performance, requiring a skilled actor who can make each character distinct. If text-to-speech is improved so that the computer voice sounds like a natural voice, it’s still going to fall short, unless the text is modified somehow to provide much more information about how each part should be read. And this kind of modification would clearly create a derivative work, so the Author’s Guild can get royalties from that.

    But the blind and the visually impaired should not be held hostage to greed.

  103. just to be the devil’s advocate Mr Blount in the original NYT article ( http://www.nytimes.com/2009/02/25/opinion/25blount.html?_r=1&scp=1&sq=kindle%20blount&st=cse ) does not pass as a tech hating money grabber.

    IMHO legal issues apart i believe that his fear is that robo-voice reading will be “good enough” for some (possibly many) people and that will erode a well established market so he would like to renegoziate the terms of licencing which is understandable.

    i find it more disturbing that all the copyfighters here have nothing but love for this Kindle that apparently is a DRM infested closed platform that will not play your files. if it was made by MS or Apple it would be hell. i hate double standards.

  104. Joe @ 104’s observation:

    “If text-to-speech is improved so that the computer voice sounds like a natural voice, it’s still going to fall short, unless the text is modified somehow to provide much more information about how each part should be read.”

    Sounds a great opportunity for writers who want to capitalize on this technology.

  105. So if I want to read to someone out loud from a book, am I also violating some sort of stupid law? The human body is also a machine.

  106. Zio Donnie @ 105

    I don’t have much love for the Kindle’s DRM, but I do love it’s potential, especially as a reading aid for the visually impaired. Even though a visually impaired person can’t operate a Kindle on his own (another reason to not love the Kindle) a caregiver can help him choose from a huge selection of books (larger than almost any other outlet, I suspect). Then the caregiver can activate the TTS software and allow the visually impaired person to enjoy the book on his own. But this is exactly the kind of use that Blount is arguing against:

    “For the record: no, the Authors Guild does not expect royalties from anybody doing non-commercial performances of “Goodnight Moon.” If parents want to send their children off to bed with the voice of Kindle 2, however, it’s another matter.”

  107. The Kindle is a cool piece of tech, but a bad platform. I’ll wait for the inevitable Chinese knock-off which will be open format and cost half as much. Oh, and feature TTS, but it will probably have a Mandarin accent.

  108. Jeff @ 112 — yes! Only I think it will be a lot longer than 5 years for text to voice to catch up. To be a good narrator you have to be a good reader — to understand the subtleties and emotional nuances of a work of literature. Such things are meaningless to a computer until AI is a lot more developed and self-aware.

    Maybe AI will become self-aware when it reads Ulysses, or Great Expectations, or something… Until then, I (and most others, I would hazard) prefer to listen to people who understand what they are saying.

    That said, it’s very possible there are some things which don’t need subtle readings…

  109. Brainspore @115 wrote:

    “Obviously text-to-speech isn’t a good substitute for a real audiobook, now or in the foreseeable future, but it’s not hard to imagine that some people, for poverty or convenience, might just take what’s given rather than purchase a separate product.”

    What makes an audiobook a separate product is the narrator’s act of interpretation. That’s what justifies paying for the audiobook in addition to paying for the print book. To me, if there is no additional content, there should be no additional charge.

    The author is compensated for the sale of the original Kindle book, and he didn’t add a damned thing to audio book that wasn’t in the print book or Kindle book – unless he did the narration. If the Kindle is doing automatic narration, what are you compensating the author for – above and beyond his contribution of writing the prose, which is compensated for by purchasing the actual Kindle book?

  110. Just to clarify what is a “derivative work” under U.S. law:

    A “derivative work” is a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a “derivative work”. 17 U.S.C. 101.

    The courts have added layers of interpretation to this definition so a review of the appropriate cases might be in order…

  111. See, an audiobook is an actual copy of the text of a book, in audio format.

    In the case of text-to-speech, the text of the book is the copy and the audio speech is a real time interpretation. Just like a school teacher reading out loud to a class. The audio is not stored anywhere, and without the text you cannot listen to it. It does not really constitute a copy.

  112. @126

    No, an audiobook is not a copy of the text of the book in audio format. It is a different work altogether.

  113. Anyone who believes that we’ll see a real-time system that can read plain text out loud in a way that would threaten audiobook performances is unbelievably ignorant when it comes to technology.

    You would need a supercomputer capable of understanding the text, who the characters are, and what’s going on to a point where it could speak in different appropriate voices, use the proper intonations, and do all of this in real time. We’re talking about a computer with the artificial intelligence of well-educated adult (when it comes to literature). Even flesh and blood performers can’t do all this on the fly.

    Sure, you could do some sort of “emotional mark-up” to the text to fake all this, but you’d be creating and selling a brand new medium, which would clearly be covered by the law.

    In 200 years, when someone creates sentient artificial life, I don’t think their goal will be to take over the audio book market.

  114. Perhaps its time to dun parents, and the buyers of children’s books, since it’s obvious that they are violating precious copyright by ‘performing’ the books for their kids at bedtime!

    Nevertheless, I think the answer is simple–use a synthesized voice such as ‘Nervous’ on OS X, and then claim exemption as a Parody.

  115. I forgot to add my other suggestion: Blount and the Dead Tree Guild ought to propose a Fourth Law of Robotics:

    “Thou shall not read aloud copyrighted book text.”

    Ironically, that would put them in a bit of a spot, since they would be creating a derivative work from Asimov’s original work;)

  116. @PTROURKE #106
    “What if I started a service in which I read your book aloud to you from your library? Would the Writer’s Guild expect me to have to give them a cut for each reading? What if it’s you’re spouse?”

    Well, if you’re, selling that service of reading a book out loud to consumers, the Author’s Guild probably would ask that you have the correct rights to do that, just like if you used a song in a film you made and expect to sell, you should probably have the rights to it. Amazon doesn’t have the audio rights to the books they sell. High-quality TTS would be a way they could (theoretically) get around ever having to buy those rights, while still undercutting the channels that do have those rights.

    That might not be the Way Things Should Be in my mind, but it’s not insanity.

    “What’s not fair is to sell someone a book, tell them they have the right to read it any time they want, and then later say “oh, but if your spouse reads it out loud to you, you will have to buy the book again in the official ‘spouse reading format’.””

    The AG isn’t saying anything like that, though. They’re basically telling Amazon that if Amazon wants to make audio versions of their e-books, that’s cool, but the original writers need to be paid for the right to perform their work like that. Which is the same business model that most other creative enterprises have run on since the dawn of copyright, and hardly an untenable position.

    “The audio book you’re buying isn’t a production of the author – it’s a production of the narrator, whose work is derived from the work of the author. The narrator gets paid, and the author gets a cut because the thing the narrator was selling”

    Right. Where Amazon is in the wrong according to the Author’s Guild is when they don’t pay the author for the thing the narrator (the Kindle in this case) is selling.

    “The reality is that the text-to-speech instantiation of the book is actually a direct product of the book itself, with no interpretation or other intervention – it is merely a mechanical instantiation of the text.”

    The quality of the end product isn’t so much a concern right now — the AG and Amazon are both confident that time and tech will vastly improve TTS. The concern is that this gets Amazon out of paying the authors for using their work in a different way. This is pretty much what Amazon would LIKE to do with tech like this, it seems. It means they don’t have to pay out as much for an audio version.

    “So, should the library have to buy a special edition that includes licensing fees for multiple readers? Should a coffee shop have to pay a fee for leaving books out on shelves for their patrons to read? Or are all of these, like the Official Spouse Reading Format Hardcover, included in the rights you obtain by purchasing a book?”

    The thing is, shifting to auio is pretty explicitly NOT included in the rights Amazon bought to sell e-books.

    Unlike a private consumer buying a book off the shelf, e-books come with a lot of baggage (cough*DRM*cough). Amazon is trying to work around that baggage.

    @#111
    “If he was argueing that the Kindle’s capability meant that authors should rexamine royalty rates for e-books he wouldn’t have been called a Luddite. If he argued that copyright law should be changed to reflect the traditional boundaries between audio and text heretofore enshrined in publishing contracts, he’d still be called a Luddite, but at least there would ba a basis for discussion.

    But by espousing the idea that copyright law DOES prohibit the TTS function of the Kindle, he shows himself to be willfully ignorant and deserving of our scorn.”

    I’m totally on board with the idea of giving most of the AG a huge wedgie for casting this as a copyright issue. I’m just mostly trying to clarify the AG’s ACTUAL point, which is, basically, “Authors should get paid for audio of their books, and Amazon is trying to use TTS as a work around so they don’t have to do that.”

    The AG is not the RIAA. They’re not high-powered middlemen trying to change the laws of nations to better accomodate an outdated business model. They’re a nonprofit. They exist basically to help authors get fair deals for selling their work. They don’t hate the Kindle or the blind. They do kind of have an axe to grind against new tech, which I think is unfortunate, but a lot of new tech does happen to tread on compensating creators (it’s a mixed bag).

    @chris7crows #113
    ” I do a statistical analysis of the words used in a book and create an abstract piece of art — one where color-coded squares correspond in size to how often a particular word appears, for a Kandinsky-like appearance.

    IMO, this should be OK. But, as the Sheperd Fairey case shows, it’s more of a legal grey area than it probably should be. The AG would seem to only have an issue if you bought the rights to the e-book and then sold the abstract art from it — that’s not the thing you negotiated.

    * I take a book, cut it into pieces, and re-arrange all the words into a new text using only those words originally in the book.

    Same as above. Cool for me, but certainly grey legally, and the AG would only have a problem, it seems, if you bought the e-book rights and decided to sell the art.

    * I use the title of a previously published book for my own book.

    That’s more classic copyright zone: usually, this is a problem (unless the title’s in the public domain or something).

    * I am inspired by a concept mentioned in a book to expand upon that concept in a work of my own.

    I can’t imagine the AG would have a problem with you making an original work. The RIAA, to my knowledge, doesn’t go after cover bands.

    It can be argued that all of these “derive” from the original work, but just because the Author’s Guild might want to pursue compensation in these scenarios doesn’t make it any less silly.”

    Sure, it might be silly, but I’d just like the argument to reflect reality rather than some AG strawman that all of the internet’s unresolved nerd-rage over the RIAA becomes focused on these guys. Cory’s post misses the actual point in a lot of ways, and I’d prefer if the sound and fury signified something.

  117. @131: Point of fact–titles are not covered by copyright. (For example, there are now two books called The Forever War, and Joe Haldeman can’t do a thing about the second except hope for some accidental spillover.)

  118. @132 Brainspore, sorry about that, my eye slipped down to the wrong line, and by the time I noticed it it was too late.

    @131 Daedalus, if you’re describing what the Kindle does as a performance, you’re never going to accept what I’m saying. The text to speech on a Kindle is the product of a device – it is a format change – not a performance.

  119. GRETA, that was a little joke. I’d like to think we’re on track with AI, but even so, it’s going to have to be raised like a baby, and we aren’t going to use it to read books. As I mentioned before, the Diamond Age had humans interacting with humans. A well done audio book is acting! And acting takes imagination, and it’s a craft that needs to be practiced. I wish more sf had more full length, well narrated audio books.

  120. 134. Well most of us DO regard what Kindle’s TTS software does as a performance in the legal sense. While there is no specific mention of a fixation requirement for derivative works, I don’t think that that it consitutes some sort of logical leap by us. And it is this lack of fixation, not the degree to which a device is used that leads us to regard the use of the TTS software as a performance rather than a copy/

  121. So, if I buy some sheet music, and then buy a program that can play music from a score, am I infringing on someone’s rights by having it play for me alone? Doesn’t seem like I should be.

  122. @138 Performing from sheet music or reading book aloud in your own home is not in question. It is legal. If you bought my sheet music, had it performed (by man or machine), and then tried to sell the recording, then we’d have a problem. Also worth noting is that publishing rights for music and mechanical rights for recordings are more thoroughly defined than for books.

    My thought experiment is this: Imagine Amazon hired a bunch of non-speakers of whatever language to phonetically read the text of any ebook sold on the Kindle, and then bundled a recording of that reading with each ebook, for free. Clearly, Amazon would be selling audiobooks, albeit crappy ones. And clearly, they would have to pay royalties.

    Essentially, the user’s experience of the current Kindle is the same or similar. But since the performance is generated on demand, by a microchip, when the user presses the “read” button, the legal situation is very different. And no matter how long it takes, the machine reading will eventually get better.

    Now is a perfect time to ask whether the copyright holder gets a piece of the sales of that performance.

    Back to the music world, imagine someone selling toys that had a microcomputer playing the Star Wars theme music, for example. Do you think that wouldn’t trigger legal action? Would they argue that the user was performing the music by pressing the button?

  123. I think Amazon and its supporters have done a great job of framing an argument that the Authors Guild ISN’T making. And, to an extent, the Authors Guild is doing a lousy job of articulating the case.

    1. Calling something copyright infringement is not the same as barring a use. No where have I read that the Authors Guild seeks to stop text-to-speech. It’s pointing out that if you do not have permission or a license to produce a derivative work, THAT is infringement. Basic copyright law.
    2. No one is saying audiobooks as we currently know them are exactly the same as text-to-speech. They are effectively two SEPARATE types of audio. An conceptual analogy would be paperback rights versus hardcover–they’re BOTH text rights, but they can, and often are, licensed separately and for different compensation structures.
    Copyright holders own ALL derivative uses, now known or developed in the future. Basic copyright law again.
    3. Licensing a new derivative right is not propping up for a failed business model. It is addressing the licensing of a new work within a successful, accepted and existing model that the entire creative works industry and copyright law is based on. You can legitimately argue (as Cory often does) that copyright law has gotten out of whack. That is not the same thing as saying that a creator does not own all derivative rights of a work, including ones not envisioned yet (or currently not successful enough to sell). If that were true, creators would not have audio, movie or performance rights at all (copyright laws go back centuries).
    4. The use of the Kindle by the consumer is NOT the issue. The use by Amazon–and attendant revenue–is.
    5. Dismissing the legal aspect of the argument because the text-to-speech software stinks is inherently foolish and detrimental to creators. Improved versions are not as far into the future as one would assume. I am not yet 50 and have been alive to see the rise of audiobooks, photocopy anthologies, VCRs, DVDs, CDs, etc.–technologies that have become major revenue streams for some people–and all technologies that companies (publishers among them) tried to grab from creators without compensation because they either weren’t envisioned when a contract was signed or a licensee tried to creatively reinterprete contract language (case in point: Peggy Lee sued Disney in 1988 over contract language from 1952–just 36 years previously. At issue: Disney did not want to pay her royalties for VCR tapes because her contract granted rights to “make phonograph recordings and/or transcriptions for sale to the public.” Disney made $90 million from that one film and tried not to pay Lee a dime. Lee won. To press the point about: it had nothing to do with how many people you sat in your living room watching the tape or preventing its distribution).

  124. @ mzed #139:

    My thought experiment is this: Imagine Amazon hired a bunch of non-speakers of whatever language to phonetically read the text of any ebook sold on the Kindle, and then bundled a recording of that reading with each ebook, for free.

    That is a very poor analogy for what Amazon is doing. A better one would be sending a non-english speaker to your home to read the text of any book you own, free with the purchase of a Kindle.

  125. markdef-Copyright holders own ALL derivative uses, now known or developed in the future. Basic copyright law again.

    Here is the root of our disagreement. Copyright says NOTHING about “derivative uses“. A simple text search will confirm that the phrase appears nowhere in 17USC. Rather it grants to the copyright holder the exclusvie right to create derivative works. The contention by many of us is use of Kindle’s TTS software doesn’t create a work since there is no fixation.

    I do however agree that the audio quality (or lack thereof) is a distraction from the central conceptual question.

  126. “Back to the music world, imagine someone selling toys that had a microcomputer playing the Star Wars theme music, for example. Do you think that wouldn’t trigger legal action? Would they argue that the user was performing the music by pressing the button?”

    No no no.

    If that toy was a keyboard that you could feed sheet music into and have it act like crappy midi player piano, and you also bought the Star Wars sheet music… then you might have an analogy.

    This is a accessibly tool, not a performance. If Amazon had advertised this feature accurately, even the AG wouldn’t have been stupid enough to start this fight.

  127. I think that the TTS feature is part of the licensing agreement for releasing a book on the Kindle.

    It is part of the negociation between Amazon and the right owner (which will often be someone non related to the actual writer of the book)

    It is part of the bargaining position of both parts (ie if you are nobody, it will be difficult to say no to a TTS feature, but if you are JKR, the balance of power will change)

    We can argue that the author’s guild stand point is conterproductive for the defence of small authors and is just empowering a little more big brass and big publishing houses, but that is an other debate.

    I agree that it will exist different level of performance : live lecture (stage) – recorded audiobook (human reader)- a accentuated TTS – a basic TTS

    I believe that accentuated TTS is closer than most of comments thinks.

    The solution is called [scansion] and is originating back to Antiquity when it was the art of scanding poems (they didn’t know how to rhyme at that time)

    It is a kind of coded signs. One can imagine a scansion track with embedded codes (an Sml (Scansion Markup Language which coud even be a standard allowing all TTS to read it)

    The business model will be the quality of the voices, the quality of the scansion, the number of different voices …

    Of course, some jobs will diapears just like internal combustion engine killed horse powered vehicles

    But, new jobs will be created with their own stars : a talented scanding (ie coding a book) artist could be revered as much as a famous conductor :
    a cory’s book coded by Mz. Y and played on Kindle 6 would be a top selling as well as a beethoven symphony conducted by karajan and recorded by deutsche gramophon …

  128. anon@142:Rather it grants to the copyright holder the exclusvie right to create derivative works. The contention by many of us is use of Kindle’s TTS software doesn’t create a work since there is no fixation.

    The former is a semantics game: if one has an exclusive right to create, no one else does without permission.

    The latter is bait-and-switch: swapping the discussion of audio rights to performance rights. There is no fixation when a band performs live or a theater group performs a play (unless recorded)—but if they use someone else’s song or play script, they need to have permission (and possibly pay a performance royalty fee) because they do not own the work. (and to head off/clarify—venues often have licenses to perform live music and such-it’s not the performer’s duty, per se. The point being, a license has to exist somewhere).

  129. UPDATE:

    http://www.latimes.com/business/la-fi-kindle28-2009feb28,0,133240.story

    “Amazon lets publishers and writers disable Kindle 2’s read-aloud feature”

    Amazon lets publishers and writers disable Kindle 2’s read-aloud feature
    The Authors Guild objected to device’s text-to-speech function, saying Amazon doesn’t have the right to essentially turn e-books into audio books.
    By Alana Semuels
    February 28, 2009
    Publishers and authors now have the power to silence the Kindle 2 e-book reader.

    Amazon.com Inc. reversed course Friday on the device’s controversial text-to-speech feature, which reads digital books aloud in a robotic voice. The company gave rights holders the ability to disable the feature for individual titles.

    The Kindle 2, which shipped this week, is a faster and smaller version of Amazon’s gadget. It can hold more than 1,500 books and has 25% more battery life than its predecessor.

    But the Authors Guild objected to the text-to-speech function, saying Amazon doesn’t have the right to essentially turn e-books into audio books. Guild President Roy Blount Jr., well-known for his role on the NPR quiz show “Wait Wait Don’t Tell Me,” wrote an opinion column in the New York Times denouncing the function.

    “They created a hybrid product,” Paul Aiken, executive director of the Authors Guild, said when reached by phone late Friday. “It was being used in a way they had not been given permission for.

    Amazon made it clear Friday that its reversal didn’t mean it agreed with that interpretation of copyright law.

    “Kindle 2’s experimental text-to-speech feature is legal: no copy is made, no derivative work is created and no performance is being given,” the company said. “Nevertheless, we strongly believe many rights holders will be more comfortable with the text-to-speech feature if they are in the driver’s seat.””

  130. MARKDEF 145-The latter is bait-and-switch: swapping the discussion of audio rights to performance rights. There is no fixation when a band performs live or a theater group performs a play (unless recorded)—but if they use someone else’s song or play script, they need to have permission (and possibly pay a performance royalty fee) because they do not own the work. (and to head off/clarify—venues often have licenses to perform live music and such-it’s not the performer’s duty, per se. The point being, a license has to exist somewhere).

    We’re just never going to agree here. You seem to be arguing that the fact that artists have to pay for public performace of a work is evidence that all performances constitute derivative works. If anyone, it is you that are performing a bait and switch here. And yet, you can buy sheet music without licensing it for public performance. I don’t believe that the legalese that comes with it specificly authorizes the creation of a derivative work. Or that anyone believes that people are buying sheet music to read silently to themesleves. Rather, the anticipation of everyone is that it will be used for non-public performance, an act that does not infringe on any of the special rights granted the creators of the work.

  131. The writers guild is within their legal rights to want control the copyright of their works. Check the laws…copyright holders control the rights to derivative works.

    Amazon is within their legal right to make devices that enable copyright violations. Amazon is prudent to let individual copyright holders licence Kindle book purchasers turn on the right to make text-to-speech works. Amazon is no more guilty of copyright violation than are axe-manufacturers of murders committed by people using those axes.

    Under copyright law, any author is free to license the creation of derivative works of their creation under the terms of their choosing. Prudent authors will allow a text-to-speech for one kindle, and not for resale.

  132. Anon@148: No, I did not say all performances are inherently derivative works—a) you’re trying to raise the “I can’t read to my children” argument, which is not a derivative RIGHT of anything b) you brought up the no fixation argument (i.e., performance) and c) both are beside the point.

    Again, your sheet music analogy supports my point rather than yours: when you buy sheet music, part of the cost is the royalty to the composer–just like part of the cost of a book pays the author. It filters back through licensing. You are free to perform (or read) it to your heart’s content. The moment someone wants to perform that music-whether fixed or not–a new license is required by law –just like taking someone’s copyrighted text and turning it into audio, which is self-evidently what Amazon is doing with no compensation to the author.

    Let’s be real: Amazon is doing this because it has a value and more so in the long run as the tech improves, not out of the goodness of their hearts. They want to make money and improve profit margin by eliminating costs (i.e., licensing fees). I find it incredibly bizarre that people seem to have no problem with Amazon making money on the backs of copyright holders, but somehow copyright holders are bad guys for wanting to make money for work that they create.

  133. The moment someone wants to perform that music-whether fixed or not–a new license is required by law –… Creators are only granted exclusive rights to public performances. You are free play music privatly in your home whether you have bought sheet music, or just have musical bent, and figuret out the chords yourself.

    I’m not even sure that this sufficiently derivative to qualify as a derivative work. After all, if it is, if the copyright holder authorized one person to use the TTS function, that person could get copyright on the derivative work. Because the creator of a derivitive work does receive copyright protection IN THE NEW WORK. Unlike reading aloud, or playing music, I don’t think that pressing one button is sufficiently expressive to qualify for copyright.

    Normally, the degree of expressiveness the choices made by the person doing it, is the border between slavish copy and a derivative work. (see Corel v. Bridgeman) Certainly the photographers for the Bridgeman Art library put had much more input into their photographs than the user of Kindle’s TTS feature. And make no mistake, Amazon isn’t creating a derivative work anymore than Fender Guitar is. It is the user who is performing/creating a derivative work.

    Of course this case is moot now. I think we agree that the writers of the author’s guild, and their publisher’s are under NO compunction to license texts to Amazon if they don’t like the TTS function. Indeed, Amazon has agreed to enable the TTS function to be turned off at the request of the publisher.

  134. anon@151: I think I see where we’re talking past each other. I am not contending that the CONSUMER is creating the derivative work. I’m contending AMAZON is. I’ve said this, yes, but when you raised the fixation issue, I was still thinking about Amazon and (i think) you were still talking about the consumer.

    I would argue that there is a fixation by Amazon–the text. The text is the basis for the produced sounds, just like an audio file on a CD produces the sounds of an audiobook. In effect, the text itself becomes an “audio file” when used in a text-to-speech program Just because one can read that file coherently because it is in a comprehensible text language, doesn’t make it different from the 1s and 0s in an audio file of an audiobook. One can’t hear an audiobook unless a button is pushed either. That doesn’t mean the source of the sound doesn’t exist until the button is pushed.

    We do agree that there is no compunction to license to Amazon—but that’s true of anyone. That attitude is basically “No, really, if you don’t like me stealing your apples, just don’t let me sell your oranges.” It doesn’t settle the question.

    If I understand you correctly, Amazon by default is enabling TTS, i.e., always on. The default position should not be “Stop enabling”, the default position should be “Go ahead, enable”—i.e., it should be incumbent upon Amazon to ask permission to do so, not the copyright holder to tell them NOT to infringe.

    That situation is analogous to EULAs that you have to “opt out” of. A lot of digital ink was spilled about how unfair that is to an end-user. Why should copyright holders have to “opt out” of Amazon’s TTS to protect their ownership rights?

Comments are closed.