STRAIGHT ON PRINT ON DEMAND
By Paula Guran
This article is also available as a downloadable MSWord document.
Most everyone in the sf/f field values Locus and depends on it for
accurate information. Like any publication, however, it occasionally
needs some assistance in making sure that information is as accurate as
I am not an expert in the publishing field. I have gained a little knowledge -- especially about modern publishing methods -- in the last few years. After reading the "2004 Book Summary" in the February 2005 issue, I felt compelled to write the following. I have, additionally, asked some folks more knowledgeable than I am to look this over in an effort to avoid errant information. Any mistakes made are, of course, my own.
I have employed the journalistic "we," utilize my own opinion in commentary, and make suggestions just as Locus did in the original article. I have tried to isolate my comments here, however, to those most germane to the topic of "print on demand."
Due to the structure of that original article, the most straightforward way to address the information provided was on a sentence-by-sentence -- even phrase by phrase in some cases -- basis. I quote from the Locus article first -- boldfaced and within quotation marks -- then present my own comments.
[If you are interested in reading the entire article, Locus has back issues available through Locus Online.]
"Print on demand remains problematic."
We are not sure "print on demand" was ever problematic and are not sure why Locus sees it as such. We do understand, however, that confusion has arisen and remains on the topic.
"It is often hard to tell if a book is POD."
We contend that if Locus has difficulties deciding if a book is POD or is not -- then it really doesn't matter at all. Yes, when print on demand technology began, there were obvious differences in the quality of the physical books produced. But, as Locus acknowledges, those differences have become (for most consumers) a non-issue.
"The stigma attached to POD has most publishers trying to avoid the label."
Without asking publishers, it would be hard to say whether this is true or not. More importantly, a misunderstanding of terms has arisen and we feel any stigma still attached is due to this. We hope to straighten that out here.
"...smaller publishers may do a small initial print run and then use POD after those copies are gone making it hard to track re-issues."
First, publishers also can (and do) switch to larger runs via other printing methods if the market demands.
But, before going further, let's define "print on demand" as it has come to mean several different things.
Please understand that during the dot-com boom and early irrational investment in e-publishing, that many companies went into the business of what was previously termed "self-publishing," "vanity," or "subsidized" publishing: the author of the material being published pays a printing company to produce books with no benefit of editorial discretion. In order to make this appear more palatable to those they wished to attract and take money from, the euphemism "print on demand" (POD) was applied to their enterprises. In truth, of course, these companies were -- and remain -- vanity publishers who use a form of printing technology referred to as "print on demand publishing." This is not "print on demand" -- but it is the major reason for the stigma to which Locus refers.
The first "real" definition of the term is purely technological. It is applied to newer methods of printing that, unlike traditional processes, do not use printing plates or any other type of physical image carrier. The printing elements are controlled by software and hardware. These new processes makes it economically feasible for a printer to do "short runs" -- printings of relatively few books. A "short run" for some printers is anything under 10,000 units; for others it is under 5,000 or 1,000. Some printers will even now deal with printing less than 100 copies at a time.
Other terms for this type of printing are "electronic-aided printing", "digital", and "direct from disk". We imagine that Locus itself is printed by this method.
The other definition applies to a combination of printing and wholesale fulfillment. ("Inventory-free publishing" is another term.) At the present this is truly offered by only one company*: Lightning Source Inc. (LSI). LSI is a subsidiary of Ingram Books, the largest U.S. wholesaler of books. Booksellers and libraries can order books printed by LSI through standard distribution and wholesale channels. Publishers set their own retail prices and discounts. LSI will literally print one book at a time.
In theory, POD was presented to the public as working like this: The printer would store digital files of the complete book. Since the printer could now feasibly print just a few books, or even a single copy, a customer could theoretically purchase a single copy of a book (from either the publisher or a retailer). The book would then be printed and sent to the purchaser. The cost of manufacturing one book at a time would be higher -- traditionally, the more copies a publisher prints, the lower the per-copy price -- but every book printed would have been pre-sold so there would be no returns.
In practice, it works nothing like that right now.
Remember, only one source -- LSI -- is really set up to accept orders through standard wholesale methods, produce a book, and fulfill an order. Any bookseller or library can order a single book (or a few or many) -- just as they can for any other book. The individual book buyer cannot. Although online retailers like Amazon can (and do) order directly from LSI, for the most part they do not handle books from smaller publishers in this way. The majority of small publishers sell their product directly to Amazon through their Advantage or Marketplace programs.
In other words, this idea that a customer can decide, "Oh, I would like a copy of The Best of Locus Volume 2," and then go to an online retailer and order it, have the book printed, and then shipped directly to him does not yet exist. (Yes, some experiments are being attempted on a small scale, but they are just that -- experiments.)
Further, LSI cannot and does not fill all orders instantly, no matter what they claim. How exactly they determine which books get printed when involves a combination of factors not all of which are known by the public. Suffice to say, from personal experience, LSI can take from a matter of days to a few weeks to print and ship an order. A surcharge of 10% will get faster priority -- but again, just how fast varies.
The reality is that to fill direct orders for customers on a timely basis, to supply Amazon, to fulfill library orders through wholesalers, to ship to at least some bookstores, for review copies, etc., the small publisher must keep an adequate supply of books on hand. Since each publisher order of books incurs a $1.50 handling fee , they are unlikely to often adhere to "one book at a time" ordering.
Definitions now aside, this brings us to wonder why "tracking reissues" is something Locus wants to do. We don't understand. All publishers "reissue" books to meet demand. Whether a book comes from a first printing or subsequent printings makes no difference in the book or to book buyers. We imagine it once mattered to book collectors, but they are savvy enough to know that these days books in a "first edition" can be printed at various locations by different printers at various times (and thus many now collect Advance Readers Copies and bound galleys, etc.).
We also have observed that publishers once noted the printing/edition and even the number of copies printed on the copyright page. We just don't see that these days beyond a First Edition or First "Name of Publisher" Edition or First "Type of" Edition and occasionally the publication month along with the year -- if that.
Perhaps "tracking reissues" has something to do with what was left unsaid in the article. Locus may be referring to the number of books printed. To determine whether a specific book or publisher is more or less legitimate based on the number of books it prints (or sells) would, however, be illogical. Locus regularly notes and frequently recommends books that are printed in limited numbers. Locus, like its readers, realizes that the number of books printed and/or sold is no reflection on the quality of a title.
At the same time, Locus is covering a portion of the business of publishing and such numbers are part of that business. But most anyone in the publishing business, and certainly Locus is aware that publishing numbers announced to the public are frequently distorted for a variety of reasons. "Real numbers," even in this day of (the extremely expensive to access and controversial) Bookscan are often closely held secrets. (Bookscan has already made a difference in how over-inflated numbers of books printed or sold are used in the publishing business and will probably have a bigger effect in the future, but this is outside our topic.)
Obviously, this is an area that anyone must be diplomatic about. In our society, big numbers of all varieties connote success; smaller numbers may indicate failure.
But Locus is not in the business of ascertaining the validity of any publisher's stated number of units produced. As for number of units sold, Locus confines itself to published reports and, for its "Locus Bestsellers" list, data gathered from booksellers.
Still, one cannot but wonder if part of the "problem" Locus has with what they term "POD publishers" is that they feel that the numbers of books these publishers sell are inconsequential. If this is the case, Locus is incorrect in that assumption. In any publishing concern some books do better than other books. This is true with independent presses or major publishers no matter what their relative size or method of production.
We suspect the real problem here may have to do with a need, in Locus's view, to separate "the big publishers" from "the little publishers." Perhaps it is time to consider the number of titles produced in a year (or "Total Books Published" as Locus calls it) is now less relevant to what divides the "majors" from the "minors". There are many ways of setting such criteria, and perhaps the time has come -- for all of us who rely on Locus for accurate information -- to do so.
One hint that Locus gives us, later in the article, about this concern with "reissues" is that they are "hard to spot unless the price goes up or the cover changes; even publishers don't always know whether the book has actually been reprinted or just unearthed from a warehouse." Does Locus mean "reprint" when they use the word "reissue"? If so, then we suggest Locus determine a viable modern definition of "reprint" with which to work. One simple determinative would be ISBN. Each version of a book is given a separate ISBN. Even if released concurrently, the hardcover and trade paperback and mass market paperback and limited edition and audio CD and e-book of any title are each given a unique ISBN. It seems to us that any title printed and distributed after initial release with a new ISBN is a reprint.
If Harry Potter and the Secret of Locus goes back to print under the same ISBN for the 53rd time, it may or may not be indicated as a "new printing." But that does not matter -- it is still part of the first release as long as it retains the same ISBN. * * *
"Major publishers seem to be reserving print-on-demand for backlist titles, if that."
At least in the context of the first definition we contend that major publishers are using POD perhaps more than Locus believes. However, whether they are or are not has no bearing on the issues Locus has raised concerning "print-on-demand publishers."
"Original POD books continue to appear from small presses..."
"...or companies that aid self-publishers, many of which are perilously close to vanity presses."
If a company is "aiding" a self-publisher then they are, indeed, vanity presses. We do not understand what this clause has to do with the first clause concerning small presses.
"We try not to list books we know are from vanity presses, but it's gotten hard to be certain."
If you err on the side of a high quality but self-published book that deserves notice, we feel that is commendable rather than not, but we agree such books are few and far between.
Our suggestion, if Locus wishes to ignore them, is to ignore books from iUniverse, PublishAmerica, Trafford, Xlibris, Author House, Vantage Press, Protea Publishing, Little Leaf Press, etc. -- any "publisher" that does not act in an editorial capacity and asks the author to subsidize costs either directly or indirectly.
We realize that's still not an all-encompassing definition, but will leave it at that since "vanity presses" have nothing to do with legitimate small publishing.
"Because POD books don't really exist until a copy is ordered and printed, they don't really compare to books published in the traditional manner."
Books are books. It is hard to imagine that Locus truly thinks a book must be printed on paper to be a book, thus saying that e-books, audiobooks, and books on CDs to not exist. But, since Locus deals primarily with printed books we understand their prejudice. If they wish to define a book as "text printed on paper and bound between two covers", POD books, in both senses, exist because they are printed. How could Locus have a hard time telling them from other books unless they were holding them in their hands?
"We don't include POD publishers on most of our main lists..."
It seems that Locus does not know which publishers are POD and which are not, especially since several of the publishers they do list use digital printing to some extent. But, more importantly, they are trying to define publishers by the methods they use to print books. We see no publishers referred to as "rotary letterpress" publishers or "sheet-fed offset lithographic" publishers, or "web-fed offset lithographic" publishers. Surely Locus realizes the absurdity of this.
"...and limit our listings of POD books to those we've actually seen here at Locus."
Is this not how Locus limits all listings of all books? We assume that no major publisher's books make any Locus lists without having been "seen." If that is a prerequisite for being included in various lists and totals, then it should, of course, be across the board.
"Publishers are often reluctant to send us the relatively expensive review copies, especially for reprint books we won't actually review."
This statement has been used by Locus in the past concerning small press as a whole, not specifically POD. We think that publishers, in general, do not send Locus (or any specific publication) books of any variety because they feel Locus has no intention of reviewing them. In some cases, publishers see no need to supply review copies of books that have already sold out a limited edition.
Locus is important as an institution that attempts to keep track of what is published by whom and when. It is possible that newer small publishers do not even know the significance of this. (Perhaps some education is due in this area?) Others simply may not care. Personally, we feel that being included for the ages in the single most complete "record of genre publishing" is worth the price of the copies and postage.
[The Locus article then includes a paragraph on Wildside Press.]
Whether any information Locus provides on Wildside or any other press is valid is not for us to decide, but we see no reason that books from Wildside, Black Coat, Hippocampus, and others should not be included on whatever lists they belong on. If they put out more books titles than Tor or Warner and they send copies of those books to Locus and thus land on a Locus-defined list above "major publishers," then, so be it. (Of course, we've also already mentioned that what defines Locus's lists may need some modernization.)
No publisher should be penalized for being financially efficient. Printing technology and fulfillment systems that allow publishing business models similar to other businesses with "just in time" manufacturing and delivery should be commended rather than condemned.
Traditionally, publishers issue books on a speculative basis hoping to sell enough of them to make a profit. By allowing publishers to print only enough copies to meet market demand (or at least closer-to-reality) POD reduces the number of unsold copies eliminating waste of materials and reducing publishers financial risk and expense. Warehousing expenses, shipping, and inventory taxes are also reduced.
We will also comment on Locus's concern that "this year [Wildside] threw us a curve by adding Prime..."
Locus seems to have no problem pulling Ballantine/Del Rey or Bantam Doubleday Dell out of the Random House, Inc group and listing all three as separate publishers. Similarly they recognize Tor and St. Martins separately, although both part of Holtzbrinck Publishing Holdings. We humbly suggest that Locus, if it wished, could do the same with Prime. (Which, by the way, uses offset as well as digital printing methods.)
"Print-on-demand books have a particularly hard time getting into stores, thanks to high prices and the perception of low quality, so that they have to rely on the Internet..."
This generality comes from ignorance. We personally know of at least one publishing company Borders asked to supply titles to their stores. They use strictly POD via LSI. When any publisher sells more than a thousand copies or so of a title, those books are getting into stores somewhere. (Moreover, the point is, most likely moot, with the recent announcement of Wildside's distribution agreement with Diamond.)
And, of course, high prices and low quality can just as easily be the result of offset printing as digital.
"...--making it impossible for us to accurately count them, or get any idea of sales."
With all due respect, we must suggest that Locus has not tried very hard. By adhering to its own definition -- that Locus must "see" the physical book -- it should be easy to get as an accurate count as Locus gets for any other publisher.
As for sales - Locus has defined how they determine sales as well: via selected bookstore data and from published bestseller lists. When books produced by so-called "POD publishers" show up on those lists (and they have already), then Locus will report it.
"Some small presses with strong niche markets find POD publishing a workable solution, but it hasn't revolutionized the market the way some pundits predicted."
We don't know about Locus, but we stopped listening to pundits when they
were predicting the "e-book revolution." And, whether a revolution is
taking place or not is, again, not germane to Locus's reporting of the
facts as they currently stand. We do think that, if nothing else,
publishers are rethinking strategies concerning rights, production,
inventory, and shipping. Authors, too, are realizing the impact POD is
* * *
Five or six years ago considerable confusion came about when certain companies such as iUniverse and Xlibris adapted "print on demand" as a euphemism for "pay-to-publish." But the terms are no more interchangeable than "science fiction" and "genre trash". We are certain that Locus will, in the future, make the distinction. Locus, has not kept up with the rapid developments of the last few years in this area of publishing. We understand this difficulty and Locus is far from alone in this lack. But since Locus is making definitions and rules based on an inaccurate understanding of current publishing practices, we felt it was time to update them.
We also hope that Locus looks to its own wisdom as far as deciding what books to include in their "count." Locus has already defined what it considers to be a book they are willing to officially "list": The book must have been physically received by Locus.
Locus may wish to add: "The book must not be self-published unless the merit of the book is so obviously above the typically self-published book as to make an exception." (If Locus is still befuddled by what is and is not self-published, we believe we (among others) might help determine what is self-published. We would be happy to help make a year-end inspection of books received. Locus need only provide adequate transportation, accommodation, and a small per diem for us to assist them.)
We further suggest that Locus adjust its criteria for ranking publishers. The "number of titles produced in a year" (or "Total Books Published" as Locus calls it) is, as mentioned, now less relevant than in the past.
If this suggestion is not viable, then we suggest that Locus adhere to its already established practice of breaking some publishers out of the large parent company when needed.
In Conclusion: After having spent lifetimes believing that "more titles published" equals "more important publisher", we all need to be aware that this has not been a functioning definition for more than a decade. Most people in the industry, if not the public, already understand that "big announced print run" does not mean "real number of books printed." We have now arrived at the day when "real numbers printed" no longer means either 1) most copies sold or 2) largest profit margin.
It is very important to those specifically interested in sf/f publishing to understand how publishing is changing. Most importantly, Locus, and other providers of information and definition, must continue to strive to present the best information, the most appropriate definitions, and most well-informed opinion that they can.
Locus provides highly valuable information and commentary. For more than twenty years it has been the most comprehensive compiler and indexer of sf/f books, stories, and magazines. These are tremendous tasks and the upswing in the sheer numbers of titles now produced in a single year makes this undertaking more daunting. At the same time, we feel that a better understanding of new forms of printing and publishing may make all of Locus's responsibilities more easily borne.
* Amazon's recent acquisition of Booksurge
changes the picture -- to what degree we don't yet know. Previous to this development
Booksurge was trying to compete with LSI, but they don't have direct
ties with Ingram and they impose their own discount and pricing schemes,
unlike Lightning Source that lets the publisher set its own discount. A
third player, Replica, is a branch of wholesaler Baker & Taylor. We
haven't heard much about them since 2002, however, and they don't even
have an operating Web site. Since books printed by LSI and Booksurge can
both be ordered through B&T, we aren't sure if Replica is even a viable
alternative at this point.
Added 05.01.05: On reading this article, a publisher emailed me the following information:
To add to your information...Booksurge does offer single-copy consumer-direct purchasing from their Web site. [Titles on this publisher's] Web site are linked directly to them for immediate purchase. Booksurge also fulfills orders for Marketplace and B&N Online.
Note also that Booksurge guarantees all wholesale orders will be printed and shipped within two business days provided the order is placed before 2 p.m. Eastern (they're in South Carolina). We have found this to be the norm.
Booksurge charges no "handling fee." We pay wholesale plus shipping, and there is a $1.50/copy discount for orders of 20-100 copies whether of a single title or multiple. There is an additional discount for 100 copies and up. They recommend not exceeding a print run of 1000 copies.
Your information that Booksurge sets discount is incorrect. I suspect the confusion arises because they set their standard discount for vendors at 20%. However, the publisher can increase that amount by sacrificing a portion of their net. We offer 35% to bookstores and libraries.
...[By using POD] the small initial cost allows us to publish more books by talented writers who might otherwise go ignored....
Added 05.10.05: M.D. Benoit, one of Zumaya's authors, adds "...that companies such as Booksurge do print one book at a time, and their shipping time is very fast. There's also the added bonus that they are set up in the US, in Canada, in Australia, and in Europe, so that readers who order from them don't have to pay duties on the books they order."
Further, John Clute wrote me with bibliographic concerns which we discussed via email. Some of that discussion and some further thoughts about publication dates went on here: http://www.ttapress.com/discus/messages/1151/1248.html?1115135263.