DRM Is Not Evil

You’re of course free to choose whether you want to publish books with DRM, but its use is based on two false assumptions: The first is that an illegal digital copy is equivalent to a lost sale. There is zero evidence that those who consume illicit material would instead purchase if the illicit option were unavailable. It’s is incorrect to assume that preventing piracy prevents “lost sales”. (See the seminal Piracy is Progressive Taxation” for a detailed argument on that front.

The second false assumption is that the underlying value is in the content itself, rather than in the delivery, packaging, and convenience (in other words, the service). O’Reilly customers are a tech savvy as anyone else’s, yet have demonstrated a willingness to spend millions of dollars for books and subscription access to content that we make openly available for free on our website.

If selling without DRM was an invitation to lost sales, then I would expect our ebook sales to be flat or shrinking, rather than growing at a double-digit rate — again, especially given the technical nature of our audience.

Again, your books, your choice. But it’s important to acknowledge that there are many publishers out there seeing very strong growth in sales of digital books without using any DRM — choosing DRM or lost sales is a false choice.

>>>Firstly because paper is a form of DRM.

Really? Paper is digital? When did this happen? Puhleeze.

Good article. Tell you what, I’ll give you DRM if you make sure all of your titles:
a) read properly on a wide variety of hardware
b) have working tables of contents
c) have working indices, references, tables of figures, etc. (where appropriate)
d) can be searched
e) work with dictionaries for lookup of unfamiliar words
f) treat ligatures and special characters correctly on all devices

You manage to do these in paper books. Give it a shot in ebooks.

Skeptically,
Jack Tingle

Social DRM is very different: there are no hardcoded technical limitations in the users files, it just helps the legal pursuit of unfair usages. That makes sense.

I greatly suspect that ‘interoperable’ DRM would be rather ineffective. If it can be placed on any machine, it can also be emailed, and presumably copied multiple times.

I sometimes wonder if a form of digital escrow might not be a better approach: a site at which anyone who buys digital material can sign up for free, and into which every digital purchase is automatically placed at the moment of purchase. Like a virtual safe, if you follow. The user is then in charge of their own DRM. They can make infinite copies of files stamped with identifying information, for their own use, at any time, without extra cost. But all copies are ultimately traceable back to them. The site would need to be run by a body independent of publishers, and it could hold any type of material – music, games, software, books.

Hmm. Maybe someone should tell Google.

Michael, you are brave to poke your head above the parapet. So please do take my comments not as an attack on you, but as an attempt to examine a couple of the strands of your argument.

DRM is a bad thing for two reasons:

1) It simply does not work. It is unenforceable, and as you yourself have pointed out, it brings annoyance to the person who has purchased the product whilst not actually protecting the creators of that product in any way. why pay and also add a ‘cost’ to the purchaser for something that does not work?

2) (This is the big one) the very mechanism by which ANY DRM works is one that turns your purchase into a rental. It may be yours until you upgrade your digital device, or until the drm server ceases to validate your file, or until the DRM controller decides to change the rules of the game (read Amazon EULAs for for scary information on what rights they have over your devices in order to ‘protect’ the DRM.

A print book isn’t a form of DRM ( I get your point, and the fact that it’s not digital is irrelevant to the point your are trying to make) for the simple reason that you own it. You can do what you want with it. Your purchase can stay with you and be handed down the generations of your family. You can re sell it. Nobody can come into your house and remove your rights to read that book. Nobody can stop you reading the book out loud to your kids or even a group of kids. Nobody can stop you reading the book because you are now in a different country and the eyeballs you use aren’t rights cleared for that location.

As for your suggestions of different grades of DRM, I say this: How on earth does one get the message across to the readers that this is anything other than a nickel and dime operation to enable them to obtain the rights for their purchase that they are already expecting to have.

If you argument is that digital is different (agreed) and that the use is different (Ok still onboard with that) and therefore the terms of use should be different, then I would say – why are the pricing models for these digital objects built on the purchase model of old.

If you wanted to be brave you could put DRM on eBooks via an explicitly described rental system: Pay-Per-View. Sky Box Office, Netflix/Blockbuster ‘downloads’, the iPlayer – nobody expects them to be permanent additions to their digital devices, so DRM works just fine there. But not if you’ve bought it.

A rental model for eBooks isn’t a bad idea…

I had this argument with Chris Anderson in 2004: http://www.boingboing.net/2004/12/29/cory_responds_to_wir.html — Every word of it is relevant to what you’ve said above. Moreso, actually, since Amazon won’t even tell the public (or publishers) how its DRM works, what flags are available, whether the DRM-free versions are portable to competing devices, whether or how DRMed files can be revoked, etc.

I don’t mean to be dismissive, but the above reads like special pleading blended with denial and wishful thinking:

“- interoperable DRM is a must. Seriously, until we have decent interoperable DRM then it will always be a huge and unnecessary barrier to adoption of new technologies. Getting this in place should be a priority for everyone in the content industries.”

So, why isn’t there interoperable DRM? As a former member of several DRM interop working groups (BPDG, DVB CPCM, XrML OASIS, etc), I can tell you: *DRM vendors don’t want to interop. They want to sucker companies like Pan-Mac into being locked into their technology*. See for example Amazon, Apple, and Audible (and that’s just the A’s).

Efforts to establish rightsholder DRM standards generally fail because *rightsholders want crazy things that CE companies would be crazy to adopt.* For example, the MPA’s rep at DVB CPCM wanted a flag that let disable entire outputs after previously approving them (“Selectable Output Control”) so that parts of your home theater would go dark depending on which show you were watching. When CE companies are present, they violently reject this, and the negotiations dissolve into ignomy. When the CE companies are absent, these things sneak in and are never adopted.

What’s more interop is *hard*, even with open standards that don’t assume that device-owners can’t be trusted. Tried to get a Bluetooth keyboard running on your netbook lately? Gotten a render-bug in Firefox or Explorer?

Add in a bunch of security gunk and you end up with a bunch of incompatible “interoperators.” See, for example, Microsoft’s failed “Plays for Sure” certification program, which they had to pull the plug on as a bad job.

On top of that, there’s the fact that interoperable DRM looks a lot like illegal collusion to antitrust regulators (that’s why every CPTWG meeting opens with a ridiculous benediction that purports to say that nothing there should be taken seriously, it’s just a casual conversation). Once at a DVB CPCM meeting someone asked what the pricing was to license a patent in a cipher under consideration and the antitrust lawyer turned green and rushed in to say that this could not, under any circumstances, be discussed.

You should try serving on a standards committee — DRM or real standards, take your pick — and come back and report on the dream of “interoperable DRM.”

(I’ve written at length about this: http://www.informationweek.com/news/personal_tech/showArticle.jhtml?articleID=201000854 )

“- more flexible DRM. I should be able to lend my file to people – just not torrent it at will.”

Who decides what’s flexible? Fair dealing and fair use vary from country to country and often require judges to interpret them. Without solving the hard AI problem, how will your “flexible” DRM know if I’m loaning my file to my sister, my friends, or a total stranger? See, for example, the DVB CPCM spec on the “authorized domain” (and see if you can make any sense of it and whether you’d put up with its rules) for how difficult this really is.

Here’s some use cases to consider for your notional flexible DRM:

* You put your elderly father in an old folks’ ward. His device is now networked to 10 other OAPs’ devices, the number changing as some pass away and others move in. Can he share his books with his new “family?” Can you explain to him how to restrict the sharing? Can the sisters on the ward resolve problems?

* You and your wife divorce and you share custody of your daughter, who goes from one house to another every week. How many times can she associate and de-associate her media library from each house before she trips a “content laundry” ceiling? How does she resolve it? Does she have to call an 0800 number and explain to someone in a distant call-centre the circumstances of her parents’ divorce?

* Your son goes on a gap-year trip before college. Every night he sleeps on the sofa of a different Facebook friend. How many of those people can read his books while he’s staying over?

“- more choices and granularity of DRM available. As a publisher we don’t always want to slap the heaviest DRM on all our titles. Yet this is what we have to do. Some titles could have lighter- or no- DRM while others have more restrictive controls.”

How do you communicate these choices to the public when they make their purchase decisions? It’s already the case that buying a track from iTunes over an iPhone requires that you read *26,000* words of license agreement — that’s for the “simple” DRM. Creative Commons has a mere five licenses and these confuse the hell out of punters (I speak from experience). How will you offer readers a fair deal that they can understand in a world where combinatorial possibilities number in the hundreds or thousands?

“- more social DRM. Watermarking and the like could be very effective, but as far as I am aware this technique is not yet widely used.”

Robust, imperceptible watermarking is technically discredited and has never worked. An imperceptible watermark can never be robust, because if a watermark adds no perceptible information to the “signal,” then the signal will not be perceptibly changed by its removal. Such removal is trivially achieved by comparing two different copies of the same book or other file. (Watermarking a text-file such as an ebook is even less plausible than watermarking a bitmap, audio file or moving image — it’s a long string of text. Comparing long strings of texts is a solved problem and has been at least since the Unix “diff” utility was written in the early 1970s).

“- an acknowledgment of the different uses and situations people might find themselves in. This means recognising that an inherent give in the system will make peoples experiences better.”

Which is to say that the DRM will be designed not to do its job. Try this one on at the standards committee and report back on the response of the security engineers present.

“- giving something back. If we are going to use DRM then we have to make sure that what we are offering really is great. This means harnessing digital delivery to add content and experiment with new forms of content to really make the offering attractive.”

You can do this without DRM more easily and cheaper and it will be even more attractive. DRM doesn’t make your products attractive. No reader woke up this morning and said, “I wish there was a way my ebooks could do less. I know, maybe those nice people at Pan-Mac have come up with something!”

“This might not make everyone turn round and start liking DRM, but it should make life easier for the most important people of all: our readers.”

Your readers don’t want DRM. What’s more, the DRM will only restrict the legitimate uses of honest readers who don’t pirate the books (remember, DRM is always cracked). The lesson, after the first time someone runs up against broken DRM, will be, “Only suckers buy media and get locked up with DRM. People who steal their media get to escape these shackles.” You drive your readers away from being your customers.

As to this business; “mass culture relies on a mass business model undermined by piracy. An argument against DRM is that the web will engender a liberation and proliferation of culture free from the corporate bonds currently suffocating it; get rid of the suits and we end up in a grass roots web driven artistic utopia. This might be true. However in this scenario there will be no more Hollywood blockbusters, huge epoch defining albums and tours, door stopping bestsellers and all the other accouterments of mass culture that rely on a company infrastructure.”

I suggest that you call up Tor, another Macmillan division, and ask about the sales figures on my non-DRM Tor Teens New York Times bestseller: nine hardcover printings (so far) totalling nearly 100,000 copies. Sold into 17 markets in 16 languages (UK rights went to HarperCollins). Film rights sold. Award-nominated theatrical adaptation just concluding in Chicago, with sublicensor queries from the West End to every major US city. Multiple bidders on graphic novel rights.

Indeed, the past year has seen a wealth of DRM-free, copy-native blockbusters, from Madonna’s deal to sell her contract to a concert promoter (copies increase ticket sales) instead of a label (copies reduce CD sales) to Radiohead, David Byrne and NIN.

I know that you’ve got writers, literary agents and internal people who’ve been sold on fairy tales about DRM. Giving in to these people (rather than setting them straight) just reduces your firm’s relevance to 21st century publishing and increases the chance that you’ll end up locked into a retail channel the way the labels are to Apple (if you think Tesco’s and Amazon exert undue influence on your operations, imagine what life would be like if your readers couldn’t change booksellers without throwing out their collections!). Don’t pander to their fears. The sooner publishing rids itself of vendor-hyped delusions about DRM, the sooner it can get on with adapting to the technical reality of the 21st century.

I’d ask one simple question – how much will you spend on DRM? Not just the one off cost this year – but every subsequent year? Every upgrade? Every customer support call? Every meeting? Cap-ex and Op-ex? What’s the TCO of DRM?

Take that money that you would be spending and plough it in to giving a free ebook reader to whoever wants one. With the remainder, invest in authors.

What’s hurting ebooks isn’t piracy – it’s the fact that a decent reader costs upwards of £200. For that kind of money I can buy 100 great books from a charity shop – or 50 from Amazon. £200 is the cost of a year’s worth of reading before I’ve actually bought a book! Be like a mobile phone company (disclaimer, I work for one, but don’t speak for them) – subsidise the hardware and charge for the content.

T

As an author I have zero business acumen. All I know is I hate DRM. I hate the way Apple made it so damned difficult move my iTunes when I changed Mac. I hate the fact my Sony ereader is essentially unusable on my Mac because of the weird and incomprehensible Adobe DRM stuff. I hate the inconvenience and I hate being mistrusted. I paid for these things? Why pick on me?

For all the above I think we need to trust readers, not punish them. Sure some will rip you off. But most readers are happy to pay the small price a real book costs, and understand that authors need to make a living too. We don’t try to block second hand book sales. We don’t scream when someone passes on a bought copy to a friend for free.

It’s long seemed to me that DRM hurt most the people who were doing the decent thing and buying the book, not those who were ripping it off and probably wouldn’t have paid for it in the first place. But I should make it clear I may well be a lone voice in these things. There were a lot of discussions about DRM in New York here last week and I suspect the majority of authors are very wary of having their work turn into the equivalent of Napster. I’m happy to go DRM-free and see how it goes. I hope it works. But if I’m ripped off right, left and centre – and ripping off of ebooks is occurring already, even with my DRMed copies – then we’ll have to look at it again.

DRM can be made commercially viable – This is already happening on the Kindle and the iphone, despite the anti-drm protests.

Your point that the Anti-DRM crowd overplays the evil of DRM is an important one. What O’Rielly, Cory Doctoroow, and, I suspect, Lawence Lessing all have in common is that their revenue isn’t exclusively (or even primarily) tied to book sales. I suspect each of them makes more in speaking fees each year (or organizing conferences where they rail about the evil of DRM) than they actually make selling books. Authors who are not inclined to license themselves to the conference circuit should think twice before following the techno-utipians into the DRM-free future.

And Andrews’s points above may be true for his business, but it’s a mistake to assume that just because O’rielly traffics in booksthe consumer is the same as those at Pan Mac.

Orielly is a technical publisher, not a publisher of trade books – the markets, and consumers, have very different buying patterns. One suspects that a large portion of Oreilly sales go through “professional” clients who are billing their company. Readers seeking entertainment, and paying out of pocket, will have a different approach to whether something that can be had for free is also worth paying for. This is very similar to the software industry, where personal users don’t think twice about using unliscensed software, while professionals do.

18

If vendors do away with DRM they have more money. Hence can digitise more product.

They will sell more books, and they don’t have to pay pipe-dream peddling technologists. How about that? They spend less time dealing with this, and the support issues. They’ll also take less abuse.

Will also likely avoid the scenario of being reviled like the music industry if a fair percentage of content is digital in the future.

Or, put it this way. I can get Nature DRM free, but mediocre novel X that hardly anyone reads needs complicated, painful protection? Let alone a celebrity biography.

20

Cory Doctorow is smarter than multiple you and mes put together and knows way more about this stuff.

23

A publisher that has actually done some research.

Something most of them don’t seem to have too much idea about. Because of they had done it, and there was evidence for their pro-DRM position, you think public relations would suggest ‘consumer consultation’ with the information. Ignoring the ‘they may have done it and found out the DRM idea is codswallop and hence ignore it’.

My argument here is simple: if we want Harry Potter- the books, films, computer games, the whole phenomenon – then DRM has a role.

You chose kind of a bad example, actually. We’d love Harry Potter e-books—but because she’s concerned about piracy, J.K. Rowling won’t make them available. Even DRM isn’t enough protection for her. The odd thing is, she’s only hurting herself with this attitude, because people who would have been willing to buy the e-books were instead downloading the pirate versions within hours of the print versions being released.

Louis @19: [Paraphrase: DRM might not be all that great now, but it could be better in the future.]

There will never be unbreakable DRM, at least as long as general-purpose computers are available to consumers. No matter how well the DRM-makers try to hide the key, hackers will be able to extract it. That flaw is inherent in the very nature of a DRM system.

> As to the “success” of DRM on the iPhone, ORLY? (as we technoutopians say).
> DRM was dropped from the major catalogs’ music offerings in iTunes years ago.

Cory, to be fair to the otherwise obnoxious JEB he didn’t say it was a success – he said it was commercially viable. It plainly is: iPhone users continue to buy millions of DRM’ed applications and music. Apple proved that you can force people (in the short term at least) to buy DRM encrypted files. Ditto Amazon and the Kindle. Personally I don’t think that’s a good thing, but hey.

I might be getting the wrong end of the stick but my understanding of Michael’s position was:

* people pirate DRM free content

* nobody has a viable, profitable new business model for (DRM free) digital content that compensates for the sales lost through that piracy yet

* without a profitable business model nobody has enough cash to invest in Harry Potter movies

* DRM is shit but if publishers make it less shit (improving interoperability, flexibility etc.) then maybe it can give the industry breathing space to experiment with new models AND keeping Harry Potter alive until we come up with something better

If Andrew in (2) is right and you don’t actually lose sales through piracy of DRM free books (I think he’s right but only half agree with his logic – I reckon you *do* lose sales but you also pick up new prospects by exposing more people to your material & in the long run you win out) then DRM is a red herring – the only thing publishers do by adding it is aggravate customers.

Are there any numbers to back that up for trade publishing though? Suggesting that we should go all DRM free is a fairly weak proposition to bring into a boardroom otherwise – the rest of the entertainment industry is complaining that billions of dollars are being lost every year (they might be making it up but they do at least have figures), people complaining about DRM are vastly outnumbered by people using it (happily or otherwise – again, see the Kindle and iPhone), the margins on books are too tight to encourage much crazy experimentation.

IMHO it’s an unpalatable but not unworkable suggestion to try and make existing DRM a bit better and then live with it until the industry finds the right way forward with DRM-free files.

If that’s what Michael is suggesting then what’s the problem?

Research has already shown that any action that could be characterized as “the average/estimated capacity for lending a paper book” will by its nature exclude people with print disabilities, including blind people, from reading the book. In the electronic context they have to go to unrealistic lengths to (often illegally) break DRM just to read a book. And in some cases that’s after they were handed that E-book as an accessible alternative to the paper book, which they couldn’t read in the first place.

“Averages” and “estimates” always leave out groups with existing legal rights, including print-disabled people, scholars, and librarians and library users, to say nothing of people with other rights, be they recognized in law yet or not.

There are instances where DRM can make some sense. I hate DRM with a passion, but even I have to admit that in some cases it can theoretically serve a purpose. I say theoretically, because in practice it’s a pain in the bottom, but let’s ignore it for now. One is the “rental” business model, whereby you get access to a large collection of titles for an affordable monthly fee. The other one is when the content is not available by any other mean, so it makes sense for the distributor to try to punish its customers. The only example I can think of is high def movies on Blue Ray; you can’t find them on any other format, so if they block that one from being copied, it can’t be copied at all.

What is completely idiotic is to put DRM on stuff that’s available without it.

Two examples:

- I wanted to listen to Bob Woodward’s “The War Within.” Available in castrated, badly compressed proprietary format from audible.com for $25. I can’t even listen to those since I use Linux. So instead I ordered it in CD form from an amazon.com affiliate for $10, shipping included. Mind you, I’m in Europe so that meant shipping half a kg of plastic across the ocean, and the delivery took 3 weeks, but that’s less than half the price and there is no shackles! And I can resell it on eBay when I’m done listening to it.

- Arte is a French-German publicly funded TV channel. You can record its programs in high definition over the air, or stream (and rip easily) all of its programs for a week after it’s been broadcast. They have a VOD service. Not only is it *very* expensive (something like €5 to download a year old 15 min show), it’s also protected by the craptastic Windows DRM (incompatible with Macs and Linux). What are they trying to protect it from? Are they afraid that someone who just paid them good money would then make copies for their friends or put them on p2p networks? Well, guess what, geniuses, they can already do that with *all* the programs you brodcast in the first place.