The tl;dr summary: it’s only fast if it’s warmed up and working properly; there’s lots of older content and public domain stuff in the database but not enough frontlist material being offered; the search interface sucks; publishers (and Google) need to take metadata more seriously, because it’s crucial for discoverability in an increasingly crowded marketplace; and the machine’s profit centers are blank journals and small print runs for a local organization, plus self-publishing one-offs.

Be sure to grab a drink and some popcorn and read the comments after the article, too. There are a couple of inflexible traditionalists who loathe the Espresso and everything they think it stands for, and they pull out every bad comment thread/discussion forum trick in the book short of referencing Hitler.

“The Good, the Bad, and the Sexy: Our Espresso Book Machine Experience” [Scholarly Kitchen]
(Photo: sukisuki)

I’ve wanted to see one of these machines up close for a while now, but I had no idea that a local indie bookstore had one. So I finished my coffee, hopped on the subway, and went to buy myself a print-on-demand book.

There’s not much I can write about the inner workings of the “big fancy robot,” as one of McNally Jackson’s employees calls it in the NPR video. It’s essentially three machines stacked together: a Xerox copier to handle the initial paper management; the custom plexiglass, steel, and blade contraption that handles the collating, binding and cutting; and a small color printer to handle the covers. (The interior pages are black and white only.)

Because this was an impulse trip, I hadn’t taken the time to scour Google Books for just the right awesome public domain title, so I had to settle for something I could find quickly that I knew wouldn’t otherwise be available in print. I chose “Jonestown Survivor,” a 2010 memoir self-published by a People’s Temple member who did not die in the mass suicide in 1978. It cost $16.95, the same as the online price at iUniverse. (The ebook edition was $9.99, grr.)

It took about fifteen minutes to print the book, although it appears that under ideal circumstances it would take less than ten: a minute to find the book through the EBM’s control panel, another minute to download the file, one more to adjust the paper, and then about five minutes to print, cut and glue the thing. You can hear and see the book coming together page by page through the plexiglass, and you can smell the glue as it’s heated up and applied to the spine. It’s sort of like a Build-A-Bear for grownups and/or nerds.

 

My copy came out with an off-centered spine. The operator, who after a month seemed to know the machine’s subtle inner workings the way a mechanic knows cars, said he’d realized that the alignment was off after the machine was too far in the process to stop it. He offered to toss it out and print a perfect version at no extra cost — in fact he encouraged me to demand this, which is one of those nice customer service touches that I think indie bookstores are good at providing — but I sort of like my book’s messed-up spine. It reminded me that this is a new application of technology, and customized sometimes means imperfect. (It also seemed too wasteful to reprint what was essentially an impulse purchase.)

He went on to say that the problem wasn’t with the file. Although the original manuscripts that customers bring in sometimes require lots of adjustments, the ones pre-formatted for the EBM by publishers are usually in good shape. In this case, the problem was that someone else had adjusted the machine since its last use, and he didn’t see this until after he’d started my book. To the person who prints the next book after mine and gets a perfect spine: you’re welcome!

I asked a bunch of questions about the size limits of the EBM, but rather than paraphrase them here I should just point you to the official FAQ, which has actual numbers. He said he’s printed stuff as short as 30 pages or so, or “essentially a pamphlet,” as well as incredibly thick books. He said Google Book scans in particular were all over the place in terms of size because of the wide range and age of the collection.

I’ll probably go back and do it again, simply because of how satisfying it is to have a print-on-demand book made in such an astonishingly short period of time. (If you called ahead, it would seem even more magical.) I think next time, however, I’ll print some old, public domain book that Google has scanned. I noticed all the sample Google Book editions had a generic text-only cover; the employee said it was possible to add your own cover (of something also in the public domain, of course) to your Google Book edition and bring it in for printing, but you’ll have to pay a $10 set-up fee.

If I ran a writing workshop in NYC, and especially if I ran one for kids, I think the workshop would end with me going to McNally Jackson and printing a very short run of an anthology of the stories created in the workshop. Aside from out-of-print and on-demand titles, that kind of personal customization seems just right for something like this.

(Photo: Sapphireblue)

Below is a look at what kind of person might buy a personalized anthology, how a retailer might offer the service, and what it might cost. When it comes to personalized anthologies, the technology and the content already exist; it just requires some large-scale cooperation among retailers, publishers, and authors, which may take some years to work out.

Portrait of a niche consumer

Jane just found out her grandfather has Alzheimer’s disease, and now everything has changed. For the rest of her life, she’s going to pay a little more attention to mentions of the disease in the news, and she’s going to instantly feel a connection to others when they disclose that they know someone with Alzheimer’s. When she reads or hears about some suggestion, however unverified, that a particular substance or food or activity might impact the disease, she’s going to consider whether to incorporate that info into her own worldview. Her superstitions and fears will change.

And–however crass this sounds–she’s permanently changed as a consumer, too. Before, Jane had little interest in recreational consumption of Alzheimer’s related content, by which I mean things like feature articles, short stories, novels, memoirs, and non fiction works. She may have read the occasional work, but because she was interested in the author, or other aspects of the story, or because it was the trendy thing to do. Now, Jane wants to read about the disease for personal reasons. As her grandfather worsens, she wants anything she can get her hands on regarding treatments, managing the patient, preserving mental health of the caregivers, and so on; after he passes on and as her own children grow up and have children, she’s going to want to know more about how families deal with it from a generational perspective, as well as how medical treatments continue to evolve. Some day, unfortunately, she may even want to read more about it to help herself.

About six months after her world has changed, Jane logs onto a retailer’s website, navigates to the ebook section, and enters the word “alzheimers.” The first result isn’t a book at all, but an invitation from the retailer:

Jane clicks “yes” and she’s taken to a “My Anthology” page. For the rest of her time assembling this product–about half an hour total–she never leaves this page; search results and customization choices are all handled here, up to the “buy it now” button that debits her account and sends the ebook to her device.

The consumer as editor, with a little help from robots

Jane’s original search term, “alzheimers,” is used to generate a “greatest hits” list of content: a couple of chapters from celebrity memoirs, an Alice Munro short story, two long-form articles from the New Yorker, and two “living with Alzheimer’s” workbooks that are available for purchase on a chapter-by-chapter basis. These have been selected based on criteria like user reviews, critical reviews, frequency of use in other user anthologies, frequency in traditional anthologies, and when possible by some sort of editorial curating paid for by the retailer.

Jane’s first choice is the Munro short story. The retailer lets her know that the story has been made into a movie, and that there is a lot of supplemental content available regarding critical reactions to the movie. The retailer presents this extra material in two groups: content that emphasizes the subject matter of the movie, and content that’s about the movie itself or about moviemaking. For the second group, Jane clicks a button that says “This is off-topic for my anthology,” and the retailer’s AI uses that data to refine future suggestions (i.e., no more movie criticism unless there’s a strong correlation to the primary topic).

As always with digital content, there are too many choices. Jane clicks the “help me decide,” button, and the retailer selects two pieces about the movie adaptation. Item 1 is an interview with the director and actors about their emotional reactions to the subject matter; item 2 is a personal essay about the movie from a viewer whose wife died of Alzheimer’s. Jane clicks the second item and it’s added to her list. And so on–gradually she builds up a collection of Alzhheimer’s-related writing.

About 20 minutes into the experience, after selecting approximately 10 pieces, Jane grows tired of cherry-picking content. She clicks a button that tells the retailer to finish up her collection. The retailer asks for some parameters: how long do you want the anthology to be? do you prefer shorter or longer pieces? personal stories or “big picture” overviews? more fiction, more nonfiction, no fiction? are you okay with including content that was previously published online but not in print? do you want poems? Then it generates a list of candidates, and Jane scrolls through and edits it down.

She looks at her table of contents: it’s arranged under Fiction, Non-Fiction, and News categories (a couple of the articles she selected make up the News group). She rearranges the content manually and creates new headers: Personal Stories, Medicine & Science, and Everything Else. Those make more sense to her. In all, she’s got 15 different items in the collection. Total number of words: 55,000. The anthology costs $12.99. Jane pays for it and it shows up on her ereader about 30 seconds later.

How do you price carved-up content?

Almost all of these items aren’t available for sale individually; Jane would have to buy the collection of short stories to read the one about Alzheimer’s, and the chapters she selected from those “caring for the sick” workbooks are only available if you buy the whole workbook. The New Yorker articles aren’t for sale on the website at all, except in this anthology format. The point behind selling custom anthologies is to use the benefits of digital publishing to create a new distribution channel: if you throw your content in with other content that shares the same theme or subject matter, you can create a mass large enough to attract a purchaser, and then you can all split the profits.

But how do you price such anthologies? Some publishers and authors are going to want $5 per item, and some are going to want $1. Some pieces will be 22,000 word novellas that took 6 months to write, some will be 10,000 word articles that required huge amounts of research, and some will be 800 word diary-like entries harvested from blogs or online journals. It doesn’t seem fair to split profits up equally by simple inclusion in an anthology, but it seems foolish to split profits based on word count or popularity too.

The fairest solution is probably to let publishers price content themselves, but within specific guidelines, and with a huge amount of education and guidance provided by the retailer. Otherwise, certain publishers and authors will instinctively price their content out of reach of the majority of customers, which ultimately hurts the retailer’s ability to offer an affordable and high-quality service in the first place.

Customized anthologies are a good example of how digital publishing might differ from traditional publishing; in this case, the book doesn’t even exist until the customer imagines what it will look like. But one important thing to note is that this doesn’t minimize the role of publishers, editors, or authors, because they are all still needed to create and shape the content in the first place. This is purely a secondary market, a way to take existing products and sell access to subsections on demand. Publishers can tap backlists to generate revenue, authors can find new readers and monetize old short stories and essays, and educators can create affordable collections for classes.

From the consumer’s perspective, of course, it’s just a more convenient way to find stuff you really want to read. Believe me, if I could create my own 12-story anthology of gay zombie cyberpunk short fiction, I would have already done it. Well, to be realistic, that anthology would probably only contain 2 or 3 stories, but you get my point.

(Photo: Sapphireblue)

A self-publisher at home in his lab. (Photo: Seattle Municipal Archives)

Best of all for the rest of us, he says he will document the process and share the results, which means any writer or publisher curious about digital distribution will be able to benefit from whatever happens. I’m rooting for ya, Doctorow.

I also think it might be interesting to look at this experiment in the context of three other online distribution experiments.

When Stephen King salted the earth

One thing that strikes me about about Doctorow’s planned roll-out, which includes everything from free ebooks and audiobooks to $250 limited edition hardcovers, is how much more sophisticated his offering is compared to something a much more popular writer tried nearly 10 years ago.

In 2000, Stephen King, working with his publisher, attempted to serialize a novel-in-progress titled “The Plant” on his website using an honor-based payment system. In 2000, your options for digital reading consisted mainly of your computer or a Palm Pilot/Handspring, and King’s experiment was therefore a lot less ambitious: he’d post chapters of the novel online, as he wrote them, and he’d expect readers to pay $1 per chapter.

King and his publisher expected, or at least wanted, most everyone to pay–freeloaders would be tolerated, but minimally. King said if at least 75% of readers didn’t pay with each new chapter, he’d pull the plug on the serialization. By the fourth or fifth chapter, King raised the price per chapter and paying readers plummeted, and he killed the project.

(Here’s a fun anecdote from a reader’s perspective: I was one of those paying per chapter and getting engrossed in the story, and I was furious at King when he killed it. It made me not want to support him on future books. Maybe not every reader felt that way, but hey, I did. It also made me realize that a writer should either ask for money up front or ask for money after, but never hold the work hostage during. Constantly threatening to abandon the story due to the actions beyond the reader’s control isn’t going to endear you to many readers.)

By contrast, the primary methods by which Doctorow will roll-out his new book seems designed to cater to freeloaders: he’s offering high-quality, free ebooks and audiobooks, licensed so that readers can manipulate them into other (equally free, as demanded by the license) works.

That’s a huge reversal from what King expected from readers in 2000. Doctorow seems to expect that the majority of readers sampling his work won’t pay, but that enough will that it will work out profitably in the end. Why would he think that? Well, he’s been offering free ebook editions of each of his books since he started publishing, and he’s a successful midlist author now. It’s not proof that giving away your work to some degree will eventually bring in buyers, but Doctorow has said he suspects the free copies work as advertising, bringing in fresh readers who, sometimes, turn into paying readers down the line².

Is “free” really just a long-term ad campaign?

And that brings me to another recent online example in giving something away for free in order to drive sales, although in this case the digital entertainment isn’t a book but a videogame.

Earlier this month, the indie developers of an enormously popular and respected videogame called “World of Goo” held a “pay what you want” sale, where you could give as little as one cent to download a Windows, Mac, or Linux version of the game. The company, 2D Boy, documented the results (part 1 is here, and part 2 is here) in a way that I hope Doctorow does as well.

Some things that I thought were interesting about the experiment:

• A significant portion of consumers elected to pay the minimum, one cent. When 2D Boy announced halfway through the sale that they only made money on purchases of 30 cents or higher, the average payment increased. Takeaway: if you’re honest about things like transaction costs, more customers will adjust their payment accordingly to avoid hurting your bottom line.
• There was a considerable price gap between Windows and Mac users (although Macs paid slightly more) compared to Linux users. Linux users seem to have been willing to pay a higher price for the content. Is this because Linux users are more familiar with the concept of paying what you think the author/developer deserves instead of what you think you can get away with? Or is this because Linux users simply aren’t the mainstream, and their very adoption of an open-source operating system means they have strong opinions on how honor payment systems should work? Whatever the reason, this price gap does seem to illustrate that when you expand to the mass marketplace, you’re going to attract more freeloaders, or more cheapskates, or likely both. You may be able to persuade them to pay more through education, but the jury’s still out on that.
• They sold 83,250 copies of the game. I don’t follow the videogame industry enough to know how that ranks to other independently-published games, and at least some of those sales were from existing customers buying copies of the game for other operating systems, but even a conservative estimate should mean that 2D Boy attracted 50,000 or more new customers on an aging game. These are customers who will now personally experience the 2D Boy brand and be more likely to pay attention to their next game release.
• Most important, perhaps: the sale bumped up full-price sales of their game on third-party reseller sites. On Steam, a PC game reseller, sales rose 40% relative to the previous week. On WiiWare, the game store on the Nintendo Wii, sales rose 9% relative to the previous week.

Taking those last two points into consideration, is it possible to use “pay what you want” sales as an advertising campaign in disguise? And if so, is it possible to adjust the minimum price point so that the immediate net cost of the campaign is positive? I don’t usually get giddy about marketing, but the idea of actually making money off of an ad campaign makes me grin deviously. (That’s “giddy” for me.)

Introducing the mass-commission

Finally, I wanted to look at something another writer is doing to make money off of publishing without relying on mainstream publishers.

Robin Sloan is well-known by a subset of web-type people, or at least that’s what I gather from Googling him. To be honest, I don’t know who he is. What I do know, however, is he just raised over $12,000 from future readers in order to fund the writing of a novel he just finished earlier this month. I donated a dollar, and I will receive a digital copy of the book when he’s through editing it. Like Doctorow, Sloan created a tiered approach that lets readers self-select the amount they’re willing to pay–each step up the ladder of donations brings more copies, or an acknowledgment, or an autograph.

My first thought, after I absorbed what he did: I hate you, Robin Sloan, for thinking of this before I did. My second thought: Can other writers replicate his success? (Should King dust off The Plant and head on over to Kickstarter?)

Doctorow has already incorporated this strategy, in a way. As part of his experiment, he planned to offer a commissioned original short story for $10,000. Before he could make a public announcement, the guy who developed Ubuntu bought it from him over breakfast, so this strategy won’t be tested among general readers this time around.

But Sloan’s experiment worked and it worked among readers closer to the norm, not with a multimillionaire. Although Sloan won’t receive royalties from his book, he’s getting a fairly decent advance for a first-time novel that’s bypassing a publishing house and going straight from author to readers. And of course, any profit he makes from direct sales will remain entirely with him.

I have no idea if other writers can pull the same trick; certainly unpublished unknowns will find it hard to attract open wallets without first demonstrating that they have a talent worth investing in. But for authors who can reach a certain level of visibility first, or who can find a way to market themselves well, it may be possible to tap into your readership directly for an advance.

Isn’t this all fascinating stuff, seeing what writers and other content creators are doing to promote and sell their work?

By contrast, here’s the sort of stuff publishers are focusing on right now: finding a way to keep readers from lending books to each other. No seriously, check it out:

So, why are publishers opposed to the Nook’s crippled ebook sharing scheme? As one Unnamed Publishing Executive told Publishers Lunch:

“If publishers agree to lending then every ebook offer now and in the future will come with this consumer feature. Over time, I’m concerned that lending won’t grow the market and in fact could hurt it.”

I thought the gatekeepers were supposed to improve the quality of content and add value to it, not find ways to restrict its consumption by readers.

Hopefully experiments like these will gradually be absorbed into the mainstream publishing houses and benefit everyone–writers, readers, and even, despite their worst intentions, publishers.

“Doctorow’s Project: With a Little Help” [Publishers Weekly]

Footnotes! Hooray!

2. To be fair to King, as a known blockbuster author he occupies a different position in the marketplace; his concerns are likely more focused on people wanting free entertainment from a known brand than on people not knowing who he is. What we really need is a blockbuster writer like King who will attempt something on the scale of Doctorow’s plan–the publishing world’s equivalent of Radiohead and their free album release experiment or NIN’s ongoing free music releases. (click here to return to the post)