A book on the web: Sherlock Holmes and the never completed text

Richard’s post on the difficulty of completing a “factual book” when new facts emerge got me thinking about the notion of completedness, and how it relates to that thing we call a book, and its publication medium. Could books published online – especially fiction – differ more from the physical texts we’ve read for centuries?

Part of our understanding of what a book is includes the notion of it being finished. Novelists research, draft, write and edit until they create something that is ready to be produced and published. Although they may feel there’s more work to be done, and that certain bits aren’t as good as they could be, decisions are reached and time is called. Manuscripts are made into an artefact, either a physical item or an eBook, which is then published, canonical and complete.

The first thing to note is that similarity between physical and digital fiction books. They are complete things, to be distributed and owned. In both cases, the web is serving the same transactional function – I order the book online in some way and initiate its delivery, whether that’s from a warehouse via a van, or directly from a server to a device. The book itself exists independently from the internet, even if it is constructed using a form of HTML.

What intrigues me is the question of what a novel or short story could look like if its being was a part of the internet? If it was published, distributed and formed online?

You can read books as web pages already (take, ahem, The Adventures of Sherlock Holmes) – and they can provide an excellent, device agnostic reading experience, requiring no distribution beyond visiting a URL (nor the author receiving a payment). But they’re mainly an already completed text, or intended to be published as such.

What would the The Adventures of Sherlock Holmes look like if Conan-Doyle was writing a text for the web today? For a start, our notion of completedness isn’t as important when publication and distribution is achieved at the click of a button. There’s no separate artefact to send to publishers, or the Amazon store.

Conan-Doyle famously resurrected Holmes after toppling him over the Reichenbach Falls in 1891 by publishing a new collection of stories, The Return of Sherlock Holmes. If he’d used a website, he could have presented alternative endings to The Final Problem (perhaps suggested in below the line comments), or even edited it two years later so Holmes didn’t die. There’d still be a canonical version of the text, maybe at https://conandoyle.com/holmes/the-final-problem; the difference being that the contents of the text could change at any time.

The edits authors make needn’t be so drastic as resurrection. Perhaps Conan-Doyle could have tightened up the final scene, changed the tone of some the passages or even introduced a new character. More cynically, writers could update sections that didn’t look quite so good in a modern light.

And yet the novel has remained fundamentally the same thing for centuries. You could argue that nonfiction has developed into something very different; after all, a website providing information is a nonfiction text, but then we still have resolutely traditional nonfiction books, such as Richard’s work on Darwin.

The reasons for this seem fairly clear. Firstly, the web makes it difficult for authors to get paid as it is an inherently shareable and copyable medium. The answer to this “problem” isn’t to create a weird ledger where we can claim ownership of artefacts such as ebooks. Instead, web authors would generally have to rely on their readers’ largesse, or a somewhat more dramatic social change where the fruits of our labour were owned and valued differently.

And maybe we don’t want to lose that notion of a completed, fully formed artefact. Whether the modern concept of a book originates in the development of the means to produce, copy and sell such a thing is a moot point; it’s been a successful and venerated concept for centuries.