Archives for category: Self publishing

In a response to my recent post on compiling AI books, Thad McIlroy sent a link to a paper on the topic, “Generative AI floods and dilutes the market for books” by Tuhin Chakrabarty, Xinyue Liu, Jane C. Ginsburg, and Paramveer Dhillon. He has himself written about it, less than flatteringly, at The Future of Publishing.

Chakrabarty et al assure us “When a large supply of cheap-to-produce titles enters a market faster than the sales, revenue, and top ranks available to absorb them, the return and attention left for the average title both fall. We call this dilution.” Mr McIlroy’s most potent point is that it’s hard to see some new thing flooding a market which is already flooded. “In 2025 there were roughly 475,000 new adult fiction titles published, from a total of over 4 million new titles and editions. That’s 71 times more new fiction titles and editions as were published in 1995.” 

We do need to acknowledge that the use of the computer in the identification of AI content is already a disputedly imprecise art. Even the best program, Pangram, manages to register false positives. However let us just accept the authors’ estimates: they say that 20% of their total 14,419 universe of books show substantial AI involvement* and 17.1% light AI content.

In this paper there’s a whole lot of hairy looking mathematics being applied to Amazon’s sales data, derived Mr McIlroy reveals from Kindle Unlimited data — which data is page-rental, not sales data. But hey, it’s hard to get these sorts of numbers! I’m sorry, but all this analysis just seems to me to be busy work. It’s what academics do. But the beam in their eye which they are of course overlooking is the fact that books are not like gravel. One load is NOT just the same as any another load. A gain in sales of this book does not mean a loss of sales for some other book. Book purchases are individual purchases, and I suspect that market dilution just isn’t a factor in the book business — that requires some sort of equivalency between the objects being sold. There is no equivalency between The Elements of Hydrodynamics and Shy Girl. I am perhaps willing to accept that some such equivalency may be available in some genre markets — though even there I’d need to see more real evidence. The authors are, to be fair, focussing on genre books plus what they call General/Contemporary Fiction. See their diagram for the categories they use in their analysis of Amazon “sales” data.

They claim to have detected a drop in sales of real books after the addition to Amazon of lots of AI books. “The study rules out the conclusion that the average revenue per book fell merely because the catalog filled up with a plethora of low-earning AI books, cumulatively depressing the average without deleteriously affecting the sales of human-written books. In other words, bloating the catalog with poorly selling AI books is not akin to the effect on individual students in a college class when weaker students enroll: there, the overall grades fall, but the weaker students’ results do not drag down the stronger students’ grades. Rather, this study shows that books with ‘no AI text’ (the good students) also, as a category, earned less than human-written books had earned before AI-content books began to compete with them. AI books’ entry into the market on a vast scale does not just overpopulate the bottom tier of sales; it allows at least some of the entrants to vie for the higher echelons of revenue generation, to the displacement of human-authored content. We note that this study does not attribute a reason for competition through scale; it merely documents the phenomenon.” . . . Our authors slide effortlessly from sales numbers (which are actually really numbers of pages read) to dollars earned. Just what “earned less than” really means is not clear. — Do bear in mind this is just a pre-print, an academic paper which has not been looked at (and by implication approved) by other scholars in the field. It may all end up having to be revised or abandoned.

Do not panic about market dilution — if your book’s any good, chances are it’ll find its market. Twas ever thus. A fixation on AI is once again obscuring the important issues.

_________________

* Substantial is defined by Chakrabarty et al. as greater than 25%. One might argue about the substantiality of 26% if one really knew what counted as evidence of AI use.

To use AI or not to use AI, that is the question: whether AI be plutonium or if it be more salt. “Plutonium is deadly in the smallest quantities. Salt can kill you if you use too much of it, but in small quantities, it makes things better.”

In this Publishing Perspectives piece, Michael Tamblyn, President and CEO of Rakuten Kobo, gives us that neat distinction, but in the end of his sober discussion of the issue of whether or not it’s “right” to sell AI-generated text, he ends up, to my mind, painfully straddling the fence. It’s fine and dandy to say you incline against the idea of allowing AI into your on-line store, but if you have no means of enforcing the policy, it seems scarcely to matter. To be fair, Mr Tamblyn knows this, and muses about ways of overcoming the problem. I suppose in the end it doesn’t really matter one way or the other — if you make the effort to keep mice out of your kitchen, the world doesn’t come to an end if you meet an occasional mouse beneath the cooker. It’s a bit like my attitude to the new mayor of New York, Zohran Mamdani, — it’s surely better to want to fix problems in the system, even if everyone knows your chances of success are slim, rather than to run on a platform of “It’s too hard to solve these problems, so we’re not going to waste time trying”. So rather like Barnes & Noble, Kobo will be selling many books written with the assistance of AI, as long as they are not grossly, obviously AI-generated. Down the road we can all argue about specific cases, once (and if) specific cases become an issue.

As in all this discussion it’s the edges that make the difference. Something awful is awful regardless of how it’s written. If you’re just reading stories to kill time, it probably doesn’t matter a hoot who or what wrote them, as long as they entertain, and fill the hours. As long as the bookstore stocks only fairly good stuff, we should be alright. Of course, this does seem to mean we expect our bookshop buyers to read the damn books — it seems every week brings us more and more evidence that the publishers aren’t even reading their manuscripts — see for example Shy Girl.

As always in the end, caveat emptor. Keep that money in your pocket if you have doubts as to whether this or that book is worth buying.

The New York Times has an interesting article by Kashmir Hill, about her reaction to discovering a 90-page AI biography of herself on Amazon. Here’s a link to the piece at The San Francisco Examiner. To say the biography wasn’t very good understates the case: just a collection of a few disparate facts and fictions cobbled together by AI. What makes a person make such a book? Ms Hill tries to locate the author of her biography but manages only to find another “writer”, Bill Jones, who publishes ten AI-generated books a month on Amazon (the most they’ll allow) — he now has 445 books available on Amazon. He is quite content to sell few copies of any one of them, even in several cases, no copies at all. Indeed he claims if he gets a really negative review of one of his books he takes the offending volume down. “I don’t take [the reviews] personally” he says, and as Ms Hill comments “Why would he? He didn’t, technically speaking, write the book.” He compiles his books as a pastime in his retirement “I wouldn’t characterize them as flying off the shelves” he said, “but it’s bar money at least.”

Inevitably these AI-generated books are not very good — though I suppose just like those monkeys typing Shakespeare, it must be at least theoretically possible that an AI-generated book could end up being a worthwhile book. But that’s not the point, is it? Nobody is pretending these books are “good”. Even Amazon, which is happy to overlook the “problem” for most of the time, will take down a book which is heavily criticized by customers. The point is, people will buy this sort of stuff. If you stand on a street corner waving dollar bills for long enough someone will soon enough appear willing to take the money even if they don’t have anything of value to swap for it. We keep getting bombarded by articles telling us that nobody reads any more, yet crap like this is being bought every day. Obviously people are willing and eager to read — and, sorry commentators, they clearly are reading, even if some of the stuff they’re reading isn’t great. (I suspect there is a fair bit of value judgement in those articles about how hoi polloi can’t read any more — only we, cultural commentators that we are, are smart enough to grapple with War and Peace!)

In so far as there’s demand for AI-generated slop, producing it is surely a perfectly legitimate endeavor. Moral judgement is beside the point. If some book buyers don’t care, why should not the enterprising entrepreneur satisfy the demand? It’s not as if the existence of bad books somehow imperils the quality of good books. You read what you chose to read. If you just want to buy a football-fan friend a birthday present of a book which you suspect they’ll never read, what does it matter that Bill Jones’ College Football Dynasties is “littered with mistakes throughout and the author tends to ramble on and be very repetitive”? It’s the thought that counts. Of course Mr Jones probably never rambles: ChatGPT’s the one who’s guilty as charged. But so what?

For more ranting on the ethics of AI writing, please see also “Shy Girl” fiasco.

Printing Impressions has a piece pushing POD as an addition to a printer’s armory. They keep focussing on customization, which may become an important feature of book making I guess. Not sure however that I’d like to receive a book with a line printed in the front saying “specially printed for Richard”. However the technology exists to provide just this “service”.

There is however no reason to question the POD impulse. As I suggested in a recent post, the print industry has been late to the party on print-on-demand, having long left it to specialists. Of course this was at least as much a result of publisher’s caution in the face of a new technology as it was printer prejudice*, but now we are all on the home stretch. Ultimately book printers will provide whatever it is that publishers will buy, and it seems we are all beginning to wake up to the fact that improved margins will result from an almost total elimination of waste in inventory management.

__________________

* Printing well by offset lithography was always a bit of a challenge, and printers who conquered the technical difficulties were always justifiably proud of themselves. Having mastered a difficult task, you would tend to be reluctant to throw it all away (quite apart from the sunk capital aspect) and change over to a less challenging technology.

Joel J. Miller points out that Serious Reading was Always a Minority Sport.*

His thesis is basically that, after a period of mass book buying, we are reverting to the norm. After all, book reading has for most of history been a minority activity (governed by access to sufficient funds to participate). It was only briefly, he argues, that books became cheap enough for the mass of people to join in the frenzy and buy books for entertainment in their idle hours. Now that so many other options have been invented, time no longer needs filling, he suggests that the job of reading has reverted to the literati.

He does quote some apparently worrying statistics. “In 1995, annual unit sales [of mass-market paperbacks] in the U.S. hit 532 million. But the wave has since crested and ebbed. By 2024, mass market sales were down to a fraction of that figure, just 21 million, a 96 percent drop.” This looks terrifying until we recognize that publishers were actively abandoning the mass-market paperback during those years. After all if you can sell 10,000 copies at $20, why would you waste your time trying to sell 15,000 at $10? Sales numbers are always hard to obtain, but I bet that sales of non-mass-market books have remained steadier — certainly book publishing is not in any sales crisis at this time. Indeed I might even like to suggest that the reason mass-market paperbacks dropped out of favor was because of a huge improvement in the book-buying market. After persuading people that your product is worth $20 why would you muddy the waters by offering it at $10?

Another striking statistic is his pointing out that “At the tail end of 2024, the National Center for Education Statistics released the 2023 findings of its Program for the International Assessment of Adult Competencies. It’s not a pretty picture. Average literacy scores in the U.S. dropped by twelve points since 2017, and the share of adults performing at Level 1 or below rose from 19 percent in 2017 to 28 percent in 2023.” Twelve points is quite a lot, but we were rather expecting something of this sort as many kids only turned up at school intermittently during the pandemic, and immigration continued apace, so it’s not altogether clear that this is any real kind of long term trend. Things are already improving.

It always seems to me quite impressive how we are able to persuade ourselves that statistics which confirm our fears and prejudices must be accurate, while we remain quite willing to question other numbers. On 8 June The New York Times published a nice opinion piece by Brian Bannon, Chief Librarian and director of branch libraries and education at New York Public Library. It is headed “Make America Read Again” — OK Mr Bannon may not have written the headline. He starts off telling about a large crowd of people turning up in January for a reading party in the library (they had to turn many away). He continues “This was not an anomaly. More New Yorkers are borrowing books from the New York Public Library today than 15 years ago; borrowing is up 27% since 2010”. His next words, however, are “And yet America is facing a book-reading crisis”. The evidence for this crisis is the usual survey or two about young people reading for pleasure. He allows himself to be conned into writing “The diagnoses keep coming. Screens. Shrinking attention spans. A culture losing its appetite for books.”

Mr Bannon starts his penultimate paragraph “The reading crisis is real”. And to the extent that “the reading crisis” is a fashionable trope, a bug infecting the book-loving brain, this is true. As far as the real world is concerned, it is (apart from the blip in school reading performance caused by the pandemic) nothing more than a comforting fiction cooked up by people who love to worry that things they love are in danger. Crisis? Book publishing is booming. Record numbers of books are being sold. Self publishing has created a large new market. New bookstores keep opening. Library borrowings are up. The encouragement of more reading is of course a desirable outcome of such lucubrations, but I’d be happier if such campaigns were more reality-based.

________________________

* Thanks to John Samples for the link.

At Small Potatoes* Paul Bloom speculates on how to repeat in the book business the Moneyball baseball trick (which basically means paying close attention to statistics and making decisions based on that analysis) and concludes that you can’t. Book numbers are fuzzy numbers. He ends up suggesting that there’s really only one (OK, two) royal roads to financial success for a book writer: “Be the sort of person who gets invited to go on Oprah”, or “Write something people really like to read”.

Moneyball-style statistical analysis works because every baseball game is essentially the same as every other baseball game — obviously individual success or failure varies, but the structure is the same, everyone does the same things. This makes comparisons potentially useful. In book publishing every book is different — OK people open them and read words, but that’s not quite the same thing as three strikes and you’re out. A baseball team has one product — baseball, and is governed by a rule book detailing the way things may happen. A publishing house has hundreds or thousands of products, each one unique, and what worked with one of these products is in no way guaranteed, or is in no way even likely, to happen with another. What sells a book? People wanting to buy a book. What makes people want to buy a book? — too many individual reasons for any analysis to mean anything useful. Back to Mr Bloom’s two tricks.

In book promotion publishers have a list of steps you need to go through with each book. We all know that doing these things may not make any difference to sales in this particular instance, but we do them anyway, partly because you’ve gotta try and that’s what we’ve always done, partly because authors expect us to, and partly because every once in a while it actually does work. (See Book promotion and Promotion and marketing for some of the mechanics.)

“The social media companies have, essentially, tricked us into making content for them, with the promise of exposure”. Complains Carol Crampton as quoted in Helen Lewis’ The Bluestocking. Exposure is necessary, but unfortunately it’s not sufficient. Nevertheless, something’s afoot — the American Booksellers Association membership grew by 19% in 2025, with 605 new brick-and-mortar, pop-up, and mobile store members. The organization has grown 151% over the last six years. CEO Allison Hill says “the majority of stores also reported year-over-year increases in sales for 2025 over 2024.” Of the 605 new stores in 2025, 399 were bricks-and-mortar, 77 mobile, 106 popups, and 23 online. Total membership now stands at 3,417. Might this bookstore growth be because books are hot nowadays — to a large extent because of social media and the fashionability of many book influencers?

__________________

* Thanks to John Samples for the link to the Paul Bloom piece.

Thus Ben Parker at Public Books in his piece entitled AI=B+. The virtue of this piece is its grounding in reality, its willingness to look at actual evidence.

Sam Altman, head OpenAI, may well claim that ChatGPT is able to spew forth words at “a legitimate PhD-level expert in anything”, but we should possibly take a look at that axe ready for grinding that he’s hiding behind his back. Professor Parker, of Brown University, has checked the evidence, and reports that “the essays it produced in mere seconds are quite plausible as the last-minute work of a rushed undergraduate. Its prose is at once eerie and banal, its analysis unswervingly trite, but full of flat-footed assertions and irrelevancies.” He assigns these papers a B+ — not awful, just not good.

I have seen a blog post written by AI for a blog meant to mimic this one, and while the output was OK, it wasn’t great. It did contain one assertion I couldn’t agree with, but even that wasn’t any kind of fatal “error”. I know, I know, you’re going to argue that AI’s going to get better — but will it? Why should it get better at writing? I suspect the main focus of development may well be on areas other than the ability to write decent prose, like driving cars, writing code, designing proteins, making diagnoses — there’s just more money to be found there. Isn’t what it’s able to do now in the way of text generation quite enough to fulfill whatever needs we might go to it for? So it can write acceptable advertising copy, an undemanding novel, an answer to your latest question — why would we invest more in training and in computing expense in order to automate “great writing”, even if we could agree what that was?

The Authors Guild have updated their Best Practices for authors. They point out “Among the most important legal considerations: AI-generated text is not copyrightable, and knowingly failing to disclose AI-generated content in a copyright registration application can constitute fraud on the Copyright Office. Many book contracts also include warranties that the manuscript is the author’s original work, meaning undisclosed inclusion of AI-generated text may put a writer in breach.” The New York Times has sent an email forbidding contributors from submitting “any material for publication that contains content generated, modified or enhanced” by generative AI. 

Not sure I think these scare tactics are quite the right approach — after all what constitutes AI-generated text? As I asked the other day does it count as AI use that when I typed the word “need” the screen suggested I might like to add the word “to”. Well obviously not; but neither do I think it should mean the generation of a first draft of which 90% of the sentences have been edited in revisions, initially by finer grained AI prompts, and finally by a human. The Copyright Office, book and journal publishers, and writers’ organizations are going to have to think harder. There’s no real benefit in using these scare tactics: at best they’ll discourage writing, and at worst promote deceit. After all it’s pretty much impossible to avoid AI altogether if you are writing on a computer. As well as those annoying word prompts, you get red dotted underlines for things it disagrees with your spelling of, and blue ones when it’s silently amended what you typed. If you Google some fact and then read the AI-generated response, does your work become uncopyrightable? (The word uncopyrightable is showing as red underlined in my draft — no doubt the word doesn’t really exist — but if I do anything about it, does this become an AI-generated text?)

I have often suggested that making students handwrite their essays represents a positive reaction to AI. Of course the students could just slavishly copy AI output, but in all probability they’ll make one or two changes, and at least they’ll have read the thing. (Slavishly copying a printed book is a long-standing traditional way of dealing with your weekly essay; equally frowned upon of course.) But I’m not sure why I should believe that looking something up in a hefty printed encyclopedia volume is axiomatically more virtuous than going to Wikipedia: but the Authors Guild and others who share their defensive crouch in the face of AI do seem to think it is.

Knee-jerk reactions are often crazy. The New Publishing Standard has a nice example of misfiring anti-AI prejudice.

Mr Lechowski posted a Monet Lilly pond picture which he said had been created by AI. Folks laid into it, pointing out how awful it was, how unworthy of Monet — then Mr Lechowski revealed that it was actually a real painting by Monet. I don’t think this event carries as much weight as Mr Williams wants it to bear, but it is amusing.

Well, we all know it happens, and we hope that we are able to avoid harm by remaining alert and skeptical.

We have been hearing over the past couple of years about authors, specifically, being targeted by online or email scammers. Robert Gray investigates the issue with one author and one publisher at the May 7th Shelf Awareness issue.

Johanna Ingalls, editor at Akashic Books in Brooklyn instances an e-mail they received from a hopeful author, who was checking that the e-mail they had received from Johnny Temple, publisher of Akashic, was indeed legitimate. It wasn’t.

Newbie authors are an easy target because their eagerness to be published may well push them into deals which turn out to be undesirable deals, and of course they’ve no idea of what a “real” publishing company might look like. The ultimate aim of the phishers is of course to extract a fee (or even better a credit card number) from the author in return for the service of bringing their book before the public. After all many a first-time author will strike a (perfectly legitimate and respectable) deal with a publishing services company offering editing, design, and even publishing services. But obviously getting an email from Publishing Services Company XYZ is a different cup of tea from getting a fake one from an imitation HarperCollins, Macmillan, Akashic or whomever. Vigilance is called for, but it’s almost not enough.

Here, from WriterBeware are accounts of scam invitations to authors to participate in book festivals, or to be interviewed on radio — both chains ultimately getting round to the point that of course you need to pay something for the honor on offer. (Link via Nate Hoffelder’s weekly Morning Coffee email round-up.)

I do suspect that the “enshittification” of the online world will probably end up meaning that we all eventually desert it. It’s getting to the point — maybe it’s already there — where we look on almost any email or message coming from an address we don’t recognize as a potential malign weapon. Let’s hope (assume) we’ll come up with something else which works better. The decay in our current system is being greatly helped along by AI-assisted nonsense. AI is already involved in compiling these phishing emails to aspiring writers. Surely there will come a moment when we say to ourselves “All this hassle over authenticity is just not worth the efficiency of online communication. I give up.” Maybe by then someone will have come up with something better than good old-fashioned letter-writing (if the postal service still exists by then — apparently they just cancelled it in Denmark), but even that we might come to love. Why did we tear out all those analog phone lines?

See also Scam sites.

In Wes Enzinna’s piece at The Baffler about author earnings, author Wallace Stegner (1909-93) is quoted as describing the job of self-supporting novelist as “the profession that doesn’t exist”.

The fact that authors mostly don’t earn a fortune can’t be news to anyone who is interested enough in the book business to be reading this blog. This is not any plot by plutocrats to defraud innocent idealists: it’s just the fact that not many books sell very many copies, and if you’re getting a couple of bucks off each one, when you add it all together you’re not going to end up rich. If you have the knack and the luck, and work at it, publishing multiple volumes each year, you can definitely make a decent living, but real wealth is the province of few traditionally published writers.

Mr Enzinna slips in a bit of knife, telling us “People at all levels of the publishing industry, meanwhile, are mostly mum on money matters, perhaps even more so in private than public.” Come off it! Remarks like this suggesting subterfuge and concealment are just the result of lazy thinking. Surprise, surprise, publishing workers regard their authors’ sales, hence incomes, as confidential information, and even if they know what author X is earning they’re not likely to blab it to a reporter. Almost everyone in every profession is “mum on money matters”; nobody goes around disclosing what anybody else earns.

The plot of a recent episode of The Brokenwood Mysteries (is it a mystery why I’m watching a New Zealand television series?) features an author who’s presented as successful, but is revealed to have, years earlier, stolen a plot from a local writer. Now of course I suppose there are plots which could be valuable, but you’ve still got to write the damned book! Even the most exciting story wouldn’t be worth enough to become a motive for murder, so after the author is found with a paperknife in his back, the detectives needn’t have bothered with that line of enquiry. But of course everyone knows that successful books are wildly profitable, so of course the Brokenwood script writers knew that it could have been a motive.

It is undeniable that every now and then a book gets made into a successful movie resulting in a big payment to the author. I’ve no way of knowing, but I doubt if this is an annual occurrence for any individual novelist: I’d bet it’s more of a once in a lifetime event. When it happens, we hear about it, and hearing about it we internalize the “fact” that authors make a lot of money.

The other relevant “fact” that we may learn about is when an author lands a big advance against royalties for their upcoming book. We never hear about the thousands of authors who get no advance, or only get a small advance — but this group forms the vast majority, and when I say vast majority I’m talking about 99.99% of authors. We never see headlines to the effect that “XYZ’s latest novel attracts a modest advance”. The only advances that get reported are the biggest advances — ‘cos someone involved wants to boast — so, following our news sources, we all know that advances are always big.

Unsurprisingly it’s not possible to know just how much the average author makes. It’s not possible to know what the average publishing employee makes. (The definition of “average” is in any case almost impossible.) Surveys get made — of publishing salaries too — but as we all know the outcome depends on the input. You can be pretty sure “the average author” didn’t respond to the survey even if they got the form. Mr Enzinna presents as evidence “According to the Authors Guild’s most recent income survey, which queried 5,699 book authors in 2023, the median book-related income for traditionally published trade authors was between $15,000 and $18,000. When combined with other writing-related income, the total climbed to a measly $23,329. Fifty-six percent of the respondents relied on side jobs to survive.” (I think we can probably assume that most of the other forty-four percent relied on a gainfully employed spouse, an inheritance, or a full-time job which required them to write in the evenings.) Again, no way to know, but that “measly $23,329” seems high for an average, unless you’re publishing multiple books every year. I think we can assume low-earning authors tend not to share their income with survey takers. I certainly won’t be building any economic arguments based on a sample of 5,699.*

In a post entitled author incomes last year I quoted Thad McIlroy’s estimate for the median for traditionally published authors at $6,000-$8,000. Makes $23,329 look a good deal less measly. But of course, one could die of exhaustion saying that the average earnings of writers is impossible to know. We just have no data: maybe it is a profession that doesn’t exist.

Numbers from the self-publishing business are hard to find, but the rate of financial success is no doubt a bit higher here than in traditional publishing. But here, I would argue, the job of author shades into the job of publisher, so comparison, even if we had data, would not be relevant. (It would be relevant if, as an author, you are motivated primarily by money! And of course, if you have the talent — and are up for the publishing mechanics side of the enterprise.)

________________

* In 2025 there were four million individual books published. Now more than one of these may well have been written by one author, but I think we might all agree that 5,699 is a pretty measly sample size.

At Publishing Confidential Kathleen Schmidt floats an interesting idea. “The crux of the issue is that too many books are published, and there are not enough places to cover them or enough people to read them. Scarcity is not a thing in book publishing. I had an idea a while ago: Create an imprint that operates in the realm of ‘drop culture.’ In other words, publish books that are digital-only and available on a particular day. Make a limited quantity available so that, when it sells out, readers must wait for the next drop. If it’s wildly popular, produce a deluxe hardcover edition. I know it would be hard to execute, but it’s not impossible with the right partners.”

Drop culture originated in the clothing business. According to Subverse “The roots of drop culture can be traced back to the streetwear scene of the early 2000s, where brands like Supreme began releasing limited batches of apparel to maintain an air of exclusivity.” Wikipedia tells us “Supreme releases two collections each year. Instead of offering the entire line at once, the brand releases small groups of products online and in-store from the current season’s collection every Thursday.. Supreme invented this retail model known as ‘drops’, which is now widely used in the industry.”

This seems to me to be an idea worth thinking about. Obviously there are large swathes of book publishing where demand is pretty much inelastic. Academic publishing for instance. You kind of know that, if there are 750 members of The Society of Nuclear Plant Managers, the maximum sale for your latest offering, Advanced Nuclear Plant Management, may be somewhere south of a thousand. No amount of marketing is going to make anyone else want it; sorry. But maybe you could create a frenzy by restricted distribution of your latest romantasy novel. Maybe even if you offered 100 copies on the first of the month, with another 100 available on the first of the next month, you could keep this going for a while, finally breaking it all open when the Pavlovian response was ingrained. Of course getting the numbers right would be critical, as it always is in publishing. A hundred’s probably too few, but would five hundred still represent scarcity?

And of course there’s always the problem of people lending books. Easier to control with ebook distribution perhaps, but frenzy is frenzy.

Trouble is that we don’t really live in a world where this is doable. How do you prevent Kindle, say, from overstepping the limit? Ms Schmidt does posit ebooks but unless the publisher is the only retailer making sales, how can you control numbers sold? Publishers, quite apart from our love of the printed book, are extremely unlikely to ever agree to publishing a book in any format which is not available to the entire trade — we depend on them and can’t afford to (wouldn’t dream of) doing anything which might damage or offend them. We’re so conditioned to making the book as widely available as possible that we’ve constructed a supply-chain in which restricted offers are impossible.

Tantalizing as Ms Schmidt’s plan is, because I suspect it might work, I can’t see how traditional publishers could ever talk themselves into testing it. Maybe a self-publisher might be able to try. Maybe someone already has.