Thursday, March 31, 2011

The blind leading the blind

A screen reader is a piece of software that converts written words to a computer-generated voice. The technology has improved over the years, and I guess some of these programs are at least satisfactory, even if listening to a book read by a screen reader will never be as fulfilling an experience as listening to an audio book recorded by a skilled voice artist.

For a publisher who is trying to accommodate people with visual impairments, producing a version of a book that works with screen readers is much less expensive than producing an audio book or a Braille book. And now that a large number of books are being packaged as electronic books (e-books), it should be no trouble to do this. At least that’s the theory.

There are some snags, though.
First, the popular e-book readers have visual controls for navigating the text. This makes them unsuitable, and therefore the main e-book file formats (.mobi for Kindle and .epub for everyone else) don’t solve the problem.

Instead, the preferred format is something called accessible PDF. This is a PDF file in which Alt tags have been embedded so that there is a verbal description associated with every picture. The PDF also has a document structure set up so that the order in which the screen reader reads the elements on a page can be controlled.

That’s the theory.

In practice, books undergo frequent revision in the last stages before they are released. So it would be very inefficient to have to start with a regular PDF after every revision and reenter all of the Alt tags, then reimplement the structural arrangement for the screen reader. Therefore, ideally, one should be able to create the Alt tags and the reading order in the original document before creating the PDF.

Now all of this—creating the book layout and generating the PDF and making it an accessible PDF—is done entirely within the integrated Adobe Creative Suite. So the programs should communicate with each other correctly.

Further, because this is Adobe’s technology, Adobe ought to have good documentation and support for the process.

Not so much. On both counts. The documentation for the process omits a great deal of critical information, and the actual output process does not work as advertised. At best the solution is partial.

This is all pretty new, though.
It’s so new, in fact, that Bowker, the company responsible for assigning ISBNs in the U.S. and for publishing the Books in Print database, have never heard of the accessible PDF format before. They learned about it today from a client of mine who is the person on whose behalf I had to disentangle the production process in the first place.

Bottom line.
I’ve now figured out what is needed to produce a proper accessible PDF that works for people with visual impairments using screen readers. Bookmark this post. If you need to produce such a book in the future, get in touch with me. What I’ll charge you for a consulting fee to walk you through it is much less than what it will cost you to figure it out on your own.

Saturday, March 19, 2011

How does that work, Daddy?

When I was a little kid (before about three quarters of people now living were born, in other words), technology was macroscopic. I could ask my father how something worked, and most of the time he could answer with what, in retrospect, was reasonable accuracy. (I could ask my mother, but her answer was consistently “Ask your father.”)

The reason he was able to do this is that an observant person could see how something worked. If you stood and watched a machine for a while, you could figure out its theory of operation. You could, if you pondered a bit, see why a particular piece was shaped the way it was. A tour guide in a factory could point to some part of the operation and explain, so a six-year-old could understand, what was happening.

This is no longer the case.
Yesterday, UPS delivered a new camera my wife had ordered. Because she plans to use it to take short movies for her next DVD, she also ordered a spare memory card. Now the camera—a high-definition model—is about the size of a pack of playing cards. The spare memory card is small enough that a spy might swallow it to avoid its discovery. And while this particular card holds 8 gigabytes of data, for a little more money the same size card could hold 32 gigabytes (256 gigabits).

There is nothing in either of these devices that the average dad can explain to the average six-year-old. Yes, I’m sure Wikipedia has articles that explain the technologies and even provide schematic diagrams. But I defy you to visualize—in a way that reflects the reality of how it’s done—how 256 billion bits of information are stored on that memory card or how a factory can produce such a device as reliably and cheaply as it does.

The same can be said for much modern technology. I know how a rotary telephone dial works. I’m okay with a Touch-Tone keypad. But my Droid does things I would not want to have to explain. (Do you really know how GPS navigation works or do you just sort of wave your hands and say plausible-sounding things about satellites whizzing around and the general theory of relativity?)

Now I say this as someone who started programming in numeric machine language in 1962 and who minored in physics and who is familiar with modern programming languages as well. It’s not that I don’t know how to make a computer do what I want it to. But that’s different from satisfying the need to see parts moving in visible space, making things.

In the natural world, we’ve gone from distinguishing organisms by their appearance and criminals by their fingerprints to distinguishing both groups by their DNA—a molecule whose structure was unknown when I was born.

Every generation faces the struggle to keep up with a changing world.
And every generation looks back at the changes it has seen. My grandparents were born before automobiles or electric lights were invented and they lived to see men walk on the moon.

I’ve heard pundits pontificate about the changes my Boomer generation has witnessed. We grew up with television and were shaped by it. Computers have moved from the periphery to the center of our lives. The Internet has truly changed the way society functions. Social media are enabling whole populations to liberate themselves. Yada yada. All of that may be true. But I think the biggest change in the way we human organisms interact with the world—what has perhaps pushed us further from our cousins in the animal kingdom rather than closer to them—is this shift from seeing the world as filled with things we can see and touch to seeing the world as filled with technology so advanced it is indistinguishable from magic, as Arthur C. Clarke put it.

Friday, March 04, 2011

The HarperCollins gambit

HarperCollins, one of the huge publishing conglomerates that dominate the commercial book world, announced that libraries “purchasing” their e-books will be able to lend such a book only twenty-six times before the book expires. The Twitterverse exploded with accusations of corporate greed and the absolute entitlement of libraries and their patrons to not only cheap but probably free books from publishers, on the theory that libraries don’t have enough money to pay for books in the first place. This view was supported by some authors who want their words read, and hang the royalties.

I suspect part of brouhaha was sparked by the injudicious use of the word purchase. E-books are software. As such, they are licensed, not purchased. When you “buy” an e-book, you are licensing the right to read it. There is nothing physical that you own. When you buy a p-book, you own the physical object. One person can read it at a time. In the U.S., when a library purchases a p-book, the author collects the royalty on one sale, and that’s the end of the transaction. In many countries (the United Kingdom and Australia among them), authors receive payments based on how many times a library lends a book. Not so here.

Another confusion arises over copying. You cannot take an in-copyright p-book to a copy shop and ask that copies be made of it. Copy shops know that’s illegal copyright infringement. But copying a digital file is trivially simple, so some people think they should be allowed to make unlimited copies of e-books and they castigate publishers for trying to protect e-books with digital rights management (DRM) technology. Yet copying is still copying is still copyright infringement. The rationale of the infinitely entitled—that unlike a p-book, which costs the publisher money for ink and paper, an e-book has no production cost—exhibits a profound misunderstanding of what publishers do and how they spend their money.

The worldview of authors (principally academics) who are secure, salaried professionals and publish books for the purpose of sharing information is very different from the worldview of authors who seek income from writing. (Yes, I know there are some academics, authors of popular undergraduate textbooks, who make a nice supplemental income from royalties; but they’re the exceptions.) One group would like libraries to get all their books free and devote their limited budgets to promoting those books to readers. The other group would like to receive royalties every time a book is checked out. Neither group seems to see any role for publishers or to think that publishers have any costs to recoup.

My point in this discussion of the HarperCollins e-book gambit is not that twenty-six is the right number or that charging more for an e-book than for a p-book can be justified. My point is that there needs to be some calm discussion between groups with differing worldviews so that a rational pricing model can emerge. This is early days in the evolution of e-books, and I have no idea where we will end up.

What has irked me about the tenor of most of the comments I’ve seen is a sense of absolute entitlement on the part of librarians to pick the pockets of publishers and authors—the people who put dinner on the tables of editors and book designers, I should add, just so you know where my interests lie. Libraries cost money to run. They are a public benefit. We, the public, should pay their costs and be happy to do so. Publishers should be free to price their goods in whatever way maximizes their profits in a competitive marketplace, and libraries should be free to buy or not buy a given publisher’s books as a result. But libraries are not entitled to a free ride at publishers’ expense just because the public feels entitled to pay too little in taxes.