Comments on: Ex-default for Kindle 2 text-to-speech: Legal?

Company backs down from its original plan to let its new Kindle 2 e-book reader read aloud from any document on the device. Its former plan was ill-considered and arguably illegal.

[PelicaTempe]

Another comment:

One point of interest (to me at least) is that if Amazon were to structure a 2-tier payment system, charging a small premium for books that were "text-to-speech" enabled, I doubt very much there would be the argument of legality based on "performance media". I'm not suggesting that Amazon do that, nor that it would be popular with it's customers.

The whole issue evolved out of a perceived undercutting of the audiobook market. Since other text-to-speech mechanisms currently exist to support the visually impaired, which have to-date gone unchallenged, the argument of legality is essentially constructed to protect profits. If the device rendered in braille I doubt very much there would be a hew and cry over legality, because the lobby for the market of braille books isn't profit motivated.

The text-to-speech issue doesn't stand up in the case of my own documentations being read aloud to me, nor does anyone seem to be arguing that newspaper or magazine subscriptions being read aloud via text-to-speech pose an issue of legality. That's because there isn't an audiobook market for them.

The problem is that the utility of text-to-speech will be hard to deny for the sake of protecting the profitability of the audiobook industry. But the current structure of the publishing industry will make this a battle similar to the Napster music issues of a decade ago. The corporate side of the argument sees text-to-speech as "free audiobooks", while for the visually impaired or those with reading disabilities see the feature as another accessibility tool.

Perhaps, in the end, and in this new climate of change, it'll become an issue for the rights of the disabled and congress will ultimately decide. Just as they have had to do so regarding other accessibility issues.

[RaskTreve]

This is another example of greed by the publishing industry, and this article employs the same poor logic. On one had he says that the quality of the text to speech feature is poor and does not compare to audio books, and then in the same breath that it will undermine the sale of audio books.

I was seriously considering purchasing the Kindle for the text to speech feature, but now I will wait until it is restored. So much for all those book purchases I would have made with it. I just don't have time in my life to read other than when I commute to work.

I will now make a point to listen to podcasts on my iPod and NOT purchase any audiobooks.

[Drisanna]

Your contention that a text-to-speech program could possibly HARM the sales of audio books is beyond ludicrous. If anything, I could only see it boosting audio book sales as people get a little taste of how nice it is to have somebody read to you while you're driving, or doing some mindless chore. Having learned that, they would rapidly become frustrated with the less than stellar "performance" of any text-to-speech engine, and seek out The Real Thing. Such extreme short-sightedness on the part of the Authors Guild and others concerned can only harm the future sales of both e-books and audio books.

For an author's perspective on these sorts of problems, I suggest that people read Eric Flint's Introduction to the Baen Free Library found at http://www.baen.com/library/. I personally have spent over $300 on Baen books since I first found the Free Library and authors there I never would have otherwise read. My hat is off to Jim Baen, my he rest in peace. He was the sort of visionary the publishing industry needs.

[Peter N. Glaskowsky]

Fine, fine, you have an opinion. But the fact you have an opinion doesn't mean you get to force it on all the authors and publishers in the world. If they believe that text-to-speech technology can harm an existing valuable business, it's their right to act accordingly.

[HLStudent]

This legal analysis seems bizarre to me.
"First, the Kindle 2's text-to-speech function is certainly copying and transforming the original work into a derivative of the original, and performing this new work for the listener. That can be fair use, or it can be a crime."
17 USC § 101 says:
"?Copies? are material objects, other than phonorecords, in which a work is fixed by any method now known or later developed, and from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. The term ?copies? includes the material object, other than a phonorecord, in which the work is first fixed."

The speech produced by the Kindle obviously isn't a copy--it doesn't satisfy the fixation requirement.
So this doesn't violate the exclusive rights provided in 17 USC § 106 to the copyright holder.

And it isn't a public performance--it's clearly intended for private performances. 17 USC § 101 says:
"To perform or display a work ?publicly? means?
(1) to perform or display it at a place open to the public or at any place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered; or
(2) to transmit or otherwise communicate a performance or display of the work to a place specified by clause (1) or to the public, by means of any device or process, whether the members of the public capable of receiving the performance or display receive it in the same place or in separate places and at the same time or at different times."
Of course, I could go out and hold up my Kindle in Boston Common and have it read a book out loud, which would be a public performance. But I could also take a hard copy of the book and read it out loud in Boston Common, which would also be a public performance.

Depending on what Amazon's agreements with publishers look like, there may be a contractual issue. But a copyright issue? Definitely not.

[Peter N. Glaskowsky]

Certainly there's a fixed copy of the audio. That's how audio chips work-- they have buffer memories. From these buffer memories, "the work can be perceived, reproduced, or otherwise communicated." If an MP3 is a "copy," so is the audio being produced by the Kindle 2. It may be a transitory copy, but the text you quoted doesn't include a temporal limit. So your first argument fails on facts. Sorry.

Yes, the Kindle 2 is intended for private use. But so are all audiobooks, and those are certainly licensed separately. So that argument isn't enough to overturn the rights of the authors and publishers to decide how books are presented in audio form.

. png

[paulej]

Peter,

I agree with others... you are really going to far. You're nitpicking over what a computer does with bits and bytes. In most (or all) cases, it is also legal to make a backup copy of software one purchases, yet the computer had to create a temporary "third" copy to facilitate the creation of the second copy. Nobody in his right mind would consider that, for a brief moment in time, the temporary copy in RAM is a violation of the copyright law because it represents some "copy" other than a legitimate backup copy.

And I would dare you take this argument before a judge. Take all of the technical nonsense to court. Nobody in their right minds would agree that the extremely temporary "copy" that might exist in a DSP is, indeed, a "copy". That is absolutely way out in left field.

I agree with others here who do not see this as a violation of copyright laws. The fact that the material is presented to the user in visual form for the eyes or audible form for the ears makes little difference. It is about like saying that the use of a different font is a copyright violation, or the very fact that Kindle had to buffer part of the page for display while also storing the original in memory is a copyright violation. I do not believe any court in America would formulate such a perverted interpretation of the copyright law.

[reasonablysure]

Everyone, now here this. There are already devices that have been "performing" text-to-speech on regular old fashioned books for years. One such device: http://www.plustek.com/product/bookreader_v100.asp (and there are others). Are they performing? Give me a break. These devices are simply for those who have a disability. I don't think anyone is expecting the voice talent of James Earl Jones on a text-to-speech device, this is just another medium to get the message to your brain (such as braille, sign language, Morse code, etc.). The particular logic from this article is pretty weak.

[PelicaTempe]

In reading all of the original article's author's comments back to us, I feel a lot of defensiveness, anger, and a whole lot of lack of understanding and empathy towards those who really need the text-to-speech feature. He doesn't get it and never will.

[Peter N. Glaskowsky]

Jeez, I'm not angry. I'm just explaining why many of these comments are wrong.

[adamjk]

The statement that a computer has to copy to do anything is ridiculous. Yes, there is a certain amount of replication of binary data to perform functions, but that's inherent in the architecture. You could also make the argument that to read someone has to 'copy' the information physically to their brain. It's all an abstraction and a physical reality of the PHYSICAL world. The definition of 'copy' that IS important to copyright is a persistent copy that is used and reused. In this case, the copy that is made for the text-to-speech is physically distinct at each reading and is always derived from the same physical source and legal source.

[whoopingrane]

First, Mr. Glaskowsy, I admire your responding to the comments posted herein. Must be exhausting considering you're making your case to much of the lunatic fringe who like to randomly post online.

I think the copyright arguments are complex and nuanced in this matter. I appreciate your citing Sony as it's been a lot of years since I've paused enough to consider that case in relation to current technologies (I'm not an intellectual property attorney so will not offer an alternate or concurring argument to yours).

But from years of practice I do know that none of this is cut and dried. From the person purchasing a book and reading it to a crowd to the text-to-speech reader in Amazon's Kindle, if such matters were to develop into suits (I doubt a court would hear a case involving the former), the analysis would venture well beyond fair use and Sony.

For those of you that espouse tenuous arguments in your comments that you believe are fact, here's a list of suggestions:
1. Take the Law School Admissions Test
2. Get into law school
3. Study law for 3 years
4. Take and pass the bar
5. Pratice for a few years
6. Come back and comment on Mr. Glaskowsky's argument.

[whoopingrane]

"Practice" - I hate typos - especially my own...

[Peter N. Glaskowsky]

Thank you very much.

I don't think it's necessary to pass the bar in order to understand or comment intelligently on my arguments here-- certainly I didn't pass the bar before writing these arguments in the first place!-- but I do agree that many people confuse what they want with what the law allows, and it would be better if they could distinguish one from the other.

[Pala98]

My wife is blind and finds the text to speach feature invaluable.

[drbig9]

Mr. Glaskowsky,

It is unfortunate to be publically dead wrong. As a major purchaser of audio books for my own use, I can tell you emphatically that the Kindle reader is in a totally different category from the audio books that I buy. A computer voice is still no match for the trained voices of the professional readers, actors and actresses who perform the audio books. As one who listens to many of these books in my car, I can tell you that the Kindle reader would be dangerous to use on a long trip!

As far as the electronic voice reader getting better and using tone and inflection, we are a long way from that process being easily automated. I would suggest that eventuality can wait for negotiation in the future.

However, eventually some third party will devise their own text to speech build-on, or work in a software patch anyway, and Amazon will have absolutely no control. The point will be moot. Are there not "reader" voice software programs already in existence for some eReader books? (Stanza desk top for the MAC comes to mind)

Only a fraction of books that are expected to be successful enough, are going to have an audio version. This means that often there is absolutely no choice about purchasing ANY audio version.

As far as your lame excuse about this being a derivative copy, I would argue that it is just another mode of privately conveying the same information to the book holder. You could hardly expect public performances of the Kindle, and in its current format these books are even less likely to be lent to others, and even more unlikely to sold as used books are.

The far greater danger to competition and fairness is the proprietary .amz format controled by the world's biggest book seller, Amazon.com. I don't see any other E-book readers that will use it. That should be far greater to concern to publishers and authors.

[Peter N. Glaskowsky]

You say I'm dead wrong, then you virtually repeat what I said in your own words! Good heavens.

The fact that other companies offer text-to-speech products is irrelevant. This piece is about the Kindle 2, which is a closed platform. The only legal way to get a current best-selling book onto Kindle 2 is through Amazon, and Amazon only gets these books through its suppliers.

In _this_ situation, the Kindle 2's text-to-speech function could be illegal, and it was certainly a crazy thing for Amazon to implement without the permission of its suppliers.

. png

[arshield]

Peter says that there is no legal way to get top sellers onto a Kindle. And that is wrong. There are a number of books being sold without DRM in formats that the Kindle reads.

[Atagahi]

Your analysis is deficient in two respects. Your deficiency is grounded in your improper legal assumptions.

1. The Kindle 2 is a machine, not a human. A human being is required to give a performance. That performance can be recorded, as is done in audio books or videos, but ultimately, the performance is created by the person.

A machine is incapable of performing. It can only replay a performance. Therefore, there is no actual performance if the machine is simply generating sound based on the text. Generating sound based on text is not illegal under copyright law unless it creates an unlicensed copy of the creative, protected work.

In this case, the work is licensed to the end-user when the legal copy is sold. The end-user may do anything with the licensed copy once it is paid for that it wants, so long as the use is not illegal. Having a computer read the text of a properly licensed copy has never been found to be a contractual breach or illegal under copyright law. It is improper for you to say it might be illegal since a court has never found it to be so on anything remotely analogous.

2. There is no illegal copy being made by the Kindle 2. It is reading from a properly paid-for, licensed copy. That's not fair use, which is an exception to illegal infringement, it is general use of the licensed copy. It falls squarely inside the proper use of a copyright, not outside it. Because having a machine convert text to speech is not an illegal general use under copyright law for an electronic file of a copyrighted work, it doesn't matter if the machine makes many copies in its functions. It is a licensed copy.

That is not to say that the Authors' Guild could not win a new right under copyright. That requires a court and the subsequent appellate courts to agree with it. At that time and only then, would the AG have the right to say that this use is illegal. Alternatively, the AG could lobby Congress for a new right under the copyright statute.

Finally, the present agreement between the AG, its members and Amazon opens them up to legal liability under the Americans with Disability Act. Your comments here that the Kindle 2 is not designed for blind people is specious. The fact that Amazon offered the Kindle 2 as a device with reading capabilities opened up the ability for those disabled people who could find it useful to use it. Not all legally blind persons have absolutely zero sight.

Some can see enough to make it through the menu and select a work to be read to them. Some may have helpers who can assist them in selecting the work to be read by the Kindle 2.

The question is if the Kindle 2 is a device that falls within the purview of the ADA statute or not. If it does, cutting off all access to the text-to-speech feature by the publisher or author could land them in court by every disabled person who wants to have the Kindle 2 read to them. If the device falls within the ADA statute's authority, then Amazon as well as the AG and authors must make reasonable accommodations to the disabled persons or face potential damages and pay the attorneys' fees of the plaintiffs. One good case could financially wipe out the AG and the authors who refuse to allow the Kindle 2 to read the text. But there will not be one case, there will be many.

Simply saying that the disabled should buy the audio books is not going to work, either. Audio books are very often abridged versions. Most courts are going to find that denying the full work to the handicapped runs afoul of the ADA. Also, often audio books are more expensive than the ebook. That needlessly penalizes the handicapped, again making it likely to not be a viable defense against an ADA complaint.

The handicapped should be able to use the text-to-speech feature without having to secure special rights from the AG or the author. That would be burdensome and impractical, making the scheme unworkable under the ADA.

The AG is ridiculously short-sighted on this and deserves to be pounded into the ground with attorneys' fees as they defend themselves against the multitude of ADA claims coming their way. I have no empathy for them.

[PelicaTempe]

Bravo! Very well written!

[Peter N. Glaskowsky]

Atagahi, you wrote a lot there, but as far as I can tell, none of those claims are supported by law. If you think they are, quote the laws, and we'll discuss them. I quoted the laws and precedents underlying my conclusions. If you won't do the same, I think you're just trying to get something you want regardless of the law.

[njakcnet]

I disagree and here is why:

1) You would have to show that the Kindle 2 gain more commercial than the kindle, because the tts function alone, which is way easier said than done. Amazon could claim that the fact it was a new version of a already beloved protect is the reason the Kindle 2, would have a greater launch than the kindle.

Also amazon doesn't just sale the tts audio themselves and doesn't sale books with tts enabled at a high valued then un-enabled tts books. So, it can be show, in the way, that Amazon isn't gaining more commercial from the tts.

2) the nature of the copyrighted work is the sole of the work. The question is do people buy audio books just because they are read out loud or because they like the person reading them? If you answer just because they are read out load, then tts does infringe. But if you can believe that people enjoy hearing different character in different voice and the that they would a sentence which is a question to sound like a question, then kindle 2's tts can't infringe on the sole of an audio book.

Note: You make no reference to this in your four reason why statement. It's not illegal to sale a picture you took of a painting, because it is understood that that picture well never be as good as that painting.

3) The issue is if the tts function infringes on audio books. Has no percentage of the audio books are used with the tts function, so it does not infringe. Amazon would have already have been given the ok to convert books to a kindle edition by there copyright holders. So, they have the right to sale the digital copies. And, like I said above they don't sale tts audio.

4) This is not a Napster thing. TTS and audio books got their starts around the same time. Though, it can be said that tts was thought if first (if you really look into it). Because they start around the same time it would be very hard to show that tts was the cause of audio book sells decline (which you must also prove).

Also there were ways to join tts and books before the kindle 2 (though the kindle 2 is the easiest and cheapest). I can only assume that Omnipage 1 through 16 and every apple vision with a tts read also infringe on audio books. And that when I used SpeakIt 0.2.0.2 to read your article. It also infringed too.

Note: You also did not point out in the Sony of America v. Universal City Studios case that Universal City Studios also lost due that fact that they receive a large amount of money for the movie, in question, to be publicly viewed for free on a cable network.

Likewise, Amazon pays publishers large amount of money (can only guess) to allow Amazon to republish a book for kindle, which just happens to have tts functions.

[letmein1234]

Stand by for the next sensational stunt: Mr. Glaskowsky is going to venture where no man has ventured before and claim that playing a DVD in a DVD player is illegal. Bejabers, it *copies* bits and sometimes even whole bytes! And the user can actually produce endless derivative works by changing brightness and volume!

Mr. Glaskowsky, you led us to believe you were a "technology analyst" (whatever that means). What are your qualifications to interpret the law?

[Peter N. Glaskowsky]

I understand it.

What are your qualifications to disagree?

Let's have no more nonsense about "qualifications". What we say here has to stand on its own merits; nobody gets a free pass because of "qualifications."

[MrBoomshadow]

Wow. All the people who disagree with you seem to have touched a nerve, Mr. Glaskowsky. Funnily enough, I'm a voice actor--one who wants to get paid for performing audiobooks, and I still disagree with your overall point.

Whether a machine-language-interpreted text-to-speech is indeed a derivative or transformative work under copyright law is decided on a case-by-case basis, and in this case, the decision has yet to be made. Personally, I don't mind the Kindle's capability one bit, as there are several huge differences between audiobooks and machine-interpreted readings:
1) Audiobooks contain nuances machines can't yet provide, and probably won't be able to for 25 years;
2) Text-to-speech software can handle things nobody has time to perform, such as newspapers and magazines;
3) Text-to-speech doesn't need to cost more, and therefore, should not from a business standpoint provide more money to authors, since it's already built in, where an audiobook requires the paid services of one or more voice actors.

You may, as is your right, disagree, but please try to be civil about it. You're on the clock.

[Peter N. Glaskowsky]

I'm not being uncivil. I'm just refusing to bow to all these crazy accusations of bias and inaccurate claims about the law and facts in this case. If you believe that disagreeing is intrinsically uncivil, of course, that's your problem, not mine.

[andrewheller2]

ro?bot?
?noun
1. a machine that resembles a human and does mechanical, routine tasks on command.
2. a person who acts and responds in a mechanical, routine manner, usually subject to another's will; automaton.
3. any machine or mechanical device that operates automatically with humanlike skill.

According to the 3rd definition of ROBOT, it seems to me that by creating the kindle, Amazon has in fact designed and begun to sell a robot that can read aloud. How is that wrong? The owner has purchased the book and simply employed his robot (Kindle) to read it aloud to him. There is nothing wrong with that. Whether the robot is good or bad at its job is irrelevant. What if someone hired a professional reader to read a book aloud? would that be copyright infringement? i think not.

[Peter N. Glaskowsky]

You can think whatever you want, but the law decides, and under the law, paying someone to read a book to you would be a commercial performance in violation of the copyright.

[__chris]

I really don't understand what all the fuss is about.

If publisher are so paranoid about derivative works, maybe we should inform them of all the derivative works that are being derived during normal use of a kindle...
The digital book is stored on a medium, then those bits are COPIED into memory. Oh no, a derivative work! Then those bits in memory are copied to video ram which are displayed onto the screen. Then a person reads the page and yet another derivative work is interpreted by the end users brain. Then that user might use said works to inspire works that could potentially be linked back to this original work on the kindle. Better call some lawyers...

Yes this is a hyperbole, but I'm making a point. Authors/Publishers need to make money, but instead of squeezing the last dime out of every customer... why not make a better product or find new areas of revenue with newer technology?

Text-to-speech has existed for a LONG time. Speak 'n' Spell? the first Mac 128k had text to speech.

You can't hold back technology, as hard as you try. It will always find a way around.

[Peter N. Glaskowsky]

When a publisher licenses Amazon to sell a book for the Kindle, it is explicitly accepting how the Kindle works.

Amazon _changed the terms of these licenses_ out from under the publishers by adding a feature (text to speech) that didn't exist when the licenses were originally signed. Obviously the publishers didn't agree to that.

They objected. Amazon had to accept the truth of their position: only the publishers get to decide how their works are presented. So now the publishers can enable or disable text to speech for each work. That's their right.

People here can complain all they want, but Amazon did the right thing here. Eventually.

Your final argument boils down to this: technology makes it easy to steal without getting caught, and there's no point to trying to stop it. Sorry, that isn't a legitimate argument, and you should be ashamed of yourself for offering it.

[aelalfy1989]

Is the question here whether text-to-speech is legal or not? haha thats a joke...I find it a werid question if you ask me. whether its used for commercial, educational or personal uses, it should be legal. like everything else if its used in the wrong ways and in the wrong hands its should be laws against that. other than that text-to-speech is still not the best technology and isn't perfect but its really cool to use.

Thank you
Ahmed El Alfy>[CNET editor's note: Prohibited spam deleted.]

[spec-tater]

I would like to add to this discussion that I am a mom of a son with dyslexia and I recently purchased a Kindle 2 for my son and I am now discouraged to hear that the text-to-speech portion of this device may not be available. It is incredibly difficult to find the different books my son wishes to read on audio version. There is just not a huge selection, especially for kids/teens. Look at your local bookstore in the kids section or browse the audio book websites in the kids book. I pick an author he wishes to read and 9 times out of 10 an audio version is just not available. Not to mention my son (who does not suffer from blindness) needs both print and audio support. The audio support helps him learn to be a better reader as he follows along. Any reading expert will tell you this. It helps him learn pacing and to get involved in the story more without his disability disrupting the flow of the story. I right now have to find book that I can buy the audio version of as well (which really limits us) and buy both. He then follows along. I understand by having this text-to-speech feature publishers cannot cash in twice with us once for a paper book and once for an audio book but how pathetic is it to cash in on a twelve year old dyslexic boy trying to learn to read. Give this kid a break and someone provide this needed feature to those who really could use it.

[shelob001]

The US copyright office has already declared Text-to-Speech legal for everyone, not just the vision impaired. check out page thirteen of http://www.copyright.gov/1201/docs/fedreg-notice-final.pdf and get your facts straight.

Peter N. Glaskowsky you might be a technical analyst, but you're not a very good reporter, nor a legal analyst. As a matter of fact, not making at least one version of a copyrighted work available for TTS use is actually in violation of the above copyright / fair use laws. The publishers and authors trying to block the use of TTS should have their legal people check it out before the groundswell catches up to them.

[zadriel]

Okay. Let's assume that today a robot is made available to us that can hold an actual hardback book and, through the use of camera type eyes, and recognize words and read them to us. Using your arguments above, does it violate copyright law? If so, how? (By the way, there is such a device now.)

If it does NOT violate copyright law. How is THAT any different from the kindle2 "holding" the ebook?

And finally if it IS different, how about when we hand the kindle 2 to the robot and it reads from the screen of the kindle to you?

Lastly, how is an artificial brain any different from the real deal? If your brain is allowed to scan text, interpret those "copies of data" on your neurons.... then trigger your vocal chords to speak the words to your child... why would a synthesized human (android or robot) be any different?

Thank you for your thoughtful reply.