A few points have been brought up in the comments to my previous post which make me realize that I left a few things out.

DRM’d Books

First, I need to point out that I’m not too worried about Amazon’s DRM’d Kindle books; DRM is a pointless exercise, for reasons that Cory Doctorow explains here. In short, he says:

Say I sell you an encrypted DVD: the encryption on the DVD is supposed to stop you (the DVD’s owner) from copying it. In order to do that, it tries to stop you from decrypting the DVD.

Except it has to let you decrypt the DVD some of the time. If you can’t decrypt the DVD, you can’t watch it. If you can’t watch it, you won’t buy it. So your DVD player is entrusted with the keys necessary to decrypt the DVD, and the film’s creator must trust that your DVD player is so well-designed that no one will ever be able to work out the key.

This is a fool’s errand. Because the DVD player has the key, it’s always possible that it can be extracted by academics, hardened hackers — or just kids who are in it for the glory.

The argument in greater detail is available here.

I expect the Kindle DRM to be broken within six weeks, at the longest, but I think I will wait until then before spending any serious money on their locked-up books. It’s not that I have something in mind that would be specifically thwarted by DRM; it’s just that the whole notion makes me uneasy. DRM ties my data — data that I am properly licensed to use — to a specific device or piece of software. When that device or piece of software fails, I effectively lose my license to use the data: and that’s nuts.

And DRM is artificially restrictive, too. The goal is to keep me from being able to redistribute the data to someone else, which is, I suppose, fine.

Note that my assessment that this is ‘fine’ is based in large part on the fact that Amazon is willing to sell this data at about a 60% discount over what they charge for the same data on paper. This is quite different from, say, the iTunes Music Store approach, where DRM’d Pixies albums actually cost $0.02 more than you’ll pay Amazon for the same data, uncompressed and un-DRM’d, on physical CDs.

It’s also based on the fact that the author and publisher need to be paid for their work, and that this isn’t going to happen if everyone redistributes the stuff for free. The argument you often see for music — that artists should make money by playing concerts rather than by selling records (which is how most recording artists make money now anyway) — doesn’t hold water in the case of books.

But there are all kinds of things I want to to do data that doesn’t involve ripping off the author or artist but that I still can’t do in a DRM regime.

For instance, I don’t have a DVD player hooked up to my TV. This is mainly because DVDs are annoying; not only do you have to keep track of the things, and make sure they don’t get scratched, and sit through the insulting copyright threats, and then screw around with their badly-designed menus using badly-designed remote controls, but also because, for reasons I’m not sure about, the average useful life of a DVD player around here is about nine months. This has been true both of cheap and expensive DVD players; so I’m done with ‘em.

I have an Apple TV hooked up to my TV, so when I get a DVD I rip it and watch it on there. Some DVDs and DVD sets are particularly hard to use as directed, and a pain in the ass to rip, so I just download them. In particular here, I’m thinking about the Monty Python box set and the recent Simpsons DVDs; they have multiple titles per disc, and particularly annoying and time-wasting menus. I own these DVDs, but I never watch them because even after paying for them it’s easier to download the content separately because this way I can watch them the way I want to watch them.

So how does this relate to machine-readable books?

Last night I read the Kindle-store sample of Steve Martin’s autobiography, and I was struck by just how clearly his voice came through in the text, and how the same voice was readily apparent in his movie scripts. With enough Steve-Martin source material and futzing around with Perl, this voice could probably be quantified and I could build a Steve Martin robot. Or at least a Steve Martin detector. That sounds silly, but it would be interesting to build such a thing and run scripts through it, seeing whether it could differentiate between bits that were written by Martin himself and bits that were reworked by lesser writers.

That would be kind of neat, and entirely within my legal rights. Neither Steve Martin nor his publishers lose any money if I build a Steve Martin detector; I’m just not allowed to redistribute the book myself. But DRM makes this use — a use that makes Martin’s book more valuable to me — impossible or at least more difficult than it needs to be.

I realize that most people are not quite like me, though, and have neither the resources, skills, time, or inclination to build a Steve Martin detector. And I couldn’t easily do textual analysis on the non-DRM’d paper book, either.

Amazon sells the DRM’d versions of books at a substantial discount. In practice I think that Amazon is subsidizing these sales, but it seems to make a good bargain. Amazon and the publishers have lower costs because they do not have to manufacture, warehouse, and move around anything physical; but they have higher costs in that relatively few people will copy and redistribute physical books.

I, on the other hand, have greater benefit from the purchase because I can get the book delivered instantly, and I don’t have to have storage space on the shelf for it; but I have less benefit because I can’t resell the book or lend it out. And I had to spend $400 to enter into the whole ecosystem in the first place.

The idiotic thing to do would be to say that Amazon’s higher costs and my increased benefit exactly balance out my decreased benefit and their lower costs, and sell the e-book at the same price as the physical one. I don’t think this is true, and in any case the market doesn’t think it’s true; people won’t buy e-books at those prices.

At 60% off, though, I feel like Amazon is splitting the difference with me. Is the gross profit on a $10 e-book greater than the profit on a $25 physical book? Almost certainly. The manufacturing of physical books — leaving out paying the author and editor and the shareholders of the publishing company and the PR people and all that, but just counting printing and binding the thing — is a surprisingly expensive undertaking, accounting for the vast majority of the cost of the book.

So I’m getting less, but I’m paying less, too. I still don’t like DRM’d data, but it’s less insulting than the approach the record and movie companies take, which is that DRM’d data is better for me than non-DRM’d data. I believe the Texas way to express the feeling is ‘Don’t piss on my head and tell me it’s raining.’

The Future Of Publishing

One commenter on the earlier post brought up the easy terms under which anyone can publish their own work and have it sold by Amazon on the Kindle store. I am looking forward to seeing what this produces, as some of the self-published books out there are pretty interesting if only for their eccentricity. On this very website, I have the complete text of one such: One-Sixth Of The World’s Surface, a self-published account of travels in the Soviet Union in 1931.

However, I also expected cheap and easy online distribution of music to bring on a new Golden Age Of Pop; there are no longer any gatekeepers with faulty assumptions to keep the market from making its decisions. This has not happened, largely because while the faulty filters of the record companies can no longer present an obstacle, no other filters have arisen to replace them. Satellite-programmed radio and the Road-Rules-ization of MTV seem to have killed off those venues as means of discovering new music, and people complain about Media Consolidation. However, it’s cheap and easy for anyone to set up what amounts to their own radio station these days either as a podcast or as an audio stream, chock-a-block with podsafe music. This hasn’t happened.

Why? I don’t know. Too many of the talented bands are still signing recording contracts with the major labels, I suppose. The only way I’ve discovered new music in the last few years is through TV ads and some random ‘mix tapes’ that people upload to The Pirate Bay — but because the TPB ‘mix tapes’ are of necessity pirated music, there’s a lack of continuity and comment. If it worked well, the guys who assemble these compilations would be, functionally, the equivalent of the program directors of really cool radio stations. As it is, they have to operate from the shadows, and everyone loses.

But. Publishing. I said before that the publishing houses are spectacularly inept at their gatekeeping function. Think of the giant sections of ‘Bargain Books’ near the cash registers in every Barnes & Noble. Every one of these books (almost: B&N prints up public-domain texts specifically for this section, too) represents failure on the part of their publishers. The publishers paid an advance to the author and paid the printers to manufacture the books. The books didn’t sell, and so are now remaindered.

They’re remaindered pretty quickly these days, too, thanks to the way unsold inventory is taxed; the publishers can’t profitably just rent a giant cheap warehouse somewhere, and sell the books as they are demanded. So one of their goals is to publish books for which they can predict a strong and immediate demand.

This is very, very difficult. One — and possibly the only — good way of being sure that a book will sell well is to print the words ‘by Tom Clancy’ on the cover. Or ‘by Stephen King’. Or ‘by John Grisham’. Or something similar. These books are profitable.

The trouble is that they’re not profitable for the publishers. What do publishers do? Edit, Manufacture, and Promote. When Tom Clancy walks into the Putnam’s offices, he’s going to drive as hard a bargain as he likes. If he wanted to, he can hire his own editor (for all I know, he’s already done this). His books don’t really need promotion other than a notice to the public that a new Tom Clancy book is available.

There’s no risk in publishing a Tom Clancy novel; all the publishers do for Clancy is arrange for the manufacture and distribution of his books — and I’m sure that they get only enough margin to allow them to not lose too much money on the deal.

The movie business is in a similar situation. Sure, putting Tom Hanks’ name on the poster will guarantee you’ll sell a whole lot of tickets; but Tom Hanks gets paid so much that all those tickets make very little for the producers. (This is why movies with Tom Hanks in them these days generally are Produced By Tom Hanks.) The blockbuster is a terrible bargain for the publisher, no matter what the medium.

The money comes from the mid-list and the unexpected runaway first-novel. The ideal situation, for a publisher, is to sign up an unknown author on favorable terms, and then to have that author’s book picked for the Oprah Book Club. This does not happen very often.

So what’s going to happen? Presumably, some of these unknown novelists will choose to not take the small advance from the publishers and will instead self-publish as a Kindle book on Amazon. Some of these novels will be very good.

But they’ll be lost among all the trash, the modern-day equivalents of One-Sixth Of The World’s Surface, things that make reasonably good websites — the vanity-publishing business has got to have taken a big hit from the rise of the web — but not very good books. Where’s the filter? I’m not going to read the first chapter of all of these things to figure out which ones are good and which ones are yet another internal monologue of a lower-middle-class college girl insecure about her looks and position in society.

As self-publishing rises in volume and diminishes in stigma, the imprimatur of a publishing house will count for more and more. It could be that self-publishing (or its effective equivalent, anyway) is just what the publishers need.

Network Capabilites

I only touched on this in the earlier post, and I think that that was a mistake. As an e-book reader, the Kindle doesn’t have any serious flaws that are not purely the result of the state of the available technology (That is: an e-book reader needs a bigger, higher-contrast, higher-resolution, color-capable screen before it’s no-excuses mainstream acceptance. These don’t exist as production models yet).

The most important thing about the Kindle, by far, is that it has a self-contained Internet connection that is totally transparent to the user. By ‘totally transparent’, I mean totally. Setting up EVDO on my Macbook Pro is pretty easy: it only involves plugging in the EVDO gadget and paying a monthly bill. That’s all. Apple has made this very simple by including all the necessary software in the base OS install.

But it’s not totally transparent; I still have to buy the EVDO thingummy and plug it in; it’s a capability that I add to the computer, and that I pay for when the bill comes every month. It’s easy, but not invisible, and software cannot be written that counts on that network connection always being there the way that software for, say, the iPhone can generally assume that there’s a network connection. The iPhone’s whole purpose is as a communications device, though, and it comes with a monthly bill.

The Kindle’s network connection is totally invisible, and peripheral to the device’s main purpose. The device itself is a bit more expensive than it would otherwise be, and it’s tied into a service that tries to produce an ongoing revenue stream, and it’s produced by a company with very deep pockets. But I think it has huge potential for the future of gadgets; until now, using a gadget that did anything at all over the Internet required that you already have a home network. It’s easy to forget this at Tino Manor, but most people don’t have a home network.

Imagine what this makes possible:

• A camera sold by Flickr (or someone similar; an online photo service in any case) that automatically uploads all of your photos to the website immediately. To make them public, you just press a button on the back of the camera. Press another button on the camera, and a printed copy shows up in the mail. My dad cannot figure out how to use his computer, and so his digital camera largely gathers dust. Pushing buttons on a specialized device, though, he can do.

• A computer that’s accessible for remote administration no matter the state of the network it’s plugged into. This would help me out a lot when my dad does poke at the computer and screws something up. This would be something like Back To My Mac, but because the manufacturer could control the network, it would actually work.

• Tivo. Currently, if you have your TiVo hooked up right, you can program it remotely. Hooking the thing up right is a pain in the ass for most people, so this is a marginal feature. With self-contained EVDO, though, every single TiVo in the world would do this.

• HVAC. Wouldn’t it be nice to be able to set your thermostat remotely, or to be able to program different behaviors for different times without fiddling around in the ill-lit hallway with those tiny buttons? You can, for about $600 and a lot of screwing around. Or Carrier could start to build a self-contained network connection into their furnaces: as an added benefit, it would alert them when something was wrong with the system, so they could send out a repairman. You’d probably be able to save enough in heating and cooling costs to pay for any increased cost.

There are a million other possibilities. The important distinction is that these connections would be an integral part of the device rather than an added and optional service. There would be nothing to manage, and no bills to pay: the bills would be paid by the manufacturer, as part of the cost of producing the product in the first place. Since most devices would use the network connection only sporadically, and would transmit only very small amounts of data (EVDO is lousy for upstream traffic; you couldn’t stream your TiVo’s content over such a connection, for instance), the cost would be minimal. We tend to think of wide-area connectivity as something that’s added on, and that has to be managed by the user. If that’s not the case — and the Kindle shows that it doesn’t need to be — there are a lot of new possibilities.