Charlie Stross6d ago

#WritersCoffeeClub #WritingCommunity

20th May 2026. If not your current genre, what genre would you write?

This is your daily scheduled reminder that genres are just marketing labels.

I write SF/F, and I had a breakout success in 2007 with a novel about crime inside MMOs. In the UK, it hit its fourth hardcover reprint within 2 weeks of publication and my subsequent book advances quadrupled!

Turns out it went gold because Waterstones systematically misfiled it under "crime", not SF/F …

Show threadHollieK6d ago
@cstross SF/F Crime, just emphasising the last part of the label.
Show threadCharlie Stross6d ago
@HollieK72 Yes well, Crime outsells SF/F by about 3:1.
Show threadMark Whybird6d ago
@cstross @HollieK72 Your experience is a supporting data point for my theory that SF (and also F to a lesser degree) is not a genre by itself but a sort of meta-genre or theme category, in which you can place almost any other genre (eg crime, mystery, political thrilller, whatever).
Show threadRay McCarthy6d ago
@whybird @cstross @HollieK72
You can have multiple "genres" in any novel. It's artificial to help advertising and sometimes makes for shallower picks by the publishers because they want it to fit a box.
Riders of the Purple Sage has romance, adventure, Robinsonade etc.
Dune and early Pern novels are far more Fantasy than SF.
Forbidden Plant is based on "The Tempest".
Caves of Steel is more a detective story than SF.
Show threadHollieK6d ago
@raymaccarthy @whybird @cstross You need it to fit into a box so you can shelve it in the bookshop. Otherwise you've got hundreds of sub-genres or everything shelved A-Z under Fiction.
Show threadHollieK6d ago
@raymaccarthy @whybird @cstross I deal with this issue on a small scale in a charity bookshop that contains approximately 3,000 books in total, with various categories and sub-categories, and a limited amount of space. Under Fiction, General Fiction has 16 shelves, Crime has 7 shelves, SF/F has 2 shelves, and that's it. C J Sansom is sometimes in Crime, sometimes in General Fiction, and sometimes in both at the same time.
Show threadMark Whybird6d ago

@HollieK72 @raymaccarthy @cstross Revised theory: in computing terms, the genres should be treated a “tags” rather than categories.

This doesn’t help much in organising a bookshop, unless you wanted to just go A-Z and physically put colour-coded tags on all the books.

Show threadRay McCarthy6d ago
@whybird @HollieK72 @cstross
An app for the customer in a bookshop is less bad than QR codes in a cafe / restaurant?
Then physical books can be in multiple categories with cover view, yet spine view only on shelf to save space.
App tells you which bookcase & shelf.
Show threadMark Whybird6d ago

@raymaccarthy @HollieK72 @cstross You’ve made me think: a crowdsourced database of (potentially) every book; each one is tagged with whatever people want; mainly genres but even stuff you or I might think irrelevant (gender of author for at least one person mentioned!); people thumbs-up or thumbs-down on the relevance of each tag to the book. (Goodreads could have become this before the enshittifiers bought it.)

But here’s the key: bookshops can subscribe to it; for that they get sort of “check in/check out” by isbn inventory facility. Premium tier gets price guidance and promoted results in the worldwide inventory search — OMG SEND DOCTOROW’S HENCHPEOPLE TO BEAT ME UP IMMEDIATELY!!!

Show threadCharlie Stross5d ago
@whybird @raymaccarthy @HollieK72 You seem to be unaware of the rule of thumb (in the book trade) that 80% of sales of a given book occur within 90 days of publication. There are occasional outliers, but it undercuts the value of your proposed crowdsourced database to booksellers. And who's going to run/pay for hosting it?
Show threadRay McCarthy5d ago

@cstross @whybird @HollieK72
Sorry, who suggested a crowd sourced database?

I agree publishers seem to have.the rule of thumb.

Any computerised bookshop already has a database of their stock. I was thinking of them leveraging that.
EDIT
Just up. I see Mark Whybird @whybird suggested Crowdsourced. That won't work well for various reasons.

Show threadHollieK5d ago
@raymaccarthy @cstross @whybird And non-computerised bookshops (like charity bookshops, for instance) don't have a database of their stock. The only database for our bookshop exists in my head, and only goes as far as me being aware that we might have a certain book in, or definitely haven't ever seen it.
Show threadHollieK
@raymaccarthy @cstross @whybird Sometimes I'd like a computerised database of our stock, so I could actually track which individual books sold, and which got culled, but I'm pretty sure we couldn't afford one.
May 21 at 7:41amWeb
Show threadRay McCarthy5d ago
@HollieK72 @cstross @whybird
The actual database can be free.
There are free applications.
The cost is the work of setting it up, which is why it's only viable for big shops / chains with POS tills and ordering all on computer.
I've done video library & college computer systems in the past. The customer had to enter everything!
Hence only my ebooks are catalogued as the database (Calibre) mostly gets the metadata from the downloaded ebook (Gutenberg, Standard, GlobalGrey, Kobo, Amazon, Humble).
Show threadHollieK5d ago
@raymaccarthy @cstross @whybird On the other hand, people can read the blurb on the back of the book, and decide whether it fits their requirements. That's probably good enough for 90% of the people 90% of the time (percentages plucked out of thin air in the best tradition).
Show threadCharlie Stross5d ago
@HollieK72 @raymaccarthy @whybird I'd agree 100% with that if not for the problem that book cover blurbs are marketing copy and can be wildly divergent from the actual material because their priority is to sell product, not accurately synopsize content.
Show threadRay McCarthy5d ago
@cstross @HollieK72 @whybird
Blurbs are getting worse. I've seem many now that are so wrong they must be LLM-AI.
Show threadHollieK5d ago
@raymaccarthy @cstross @whybird So is there anything that we can trust about the descriptions of a book? If marketing is going to lie in the blurb, it will also lie with the tags.
Show threadCharlie Stross5d ago
@HollieK72 @raymaccarthy @whybird To some extent it's also on the editors and the authors. I have limited insight as I'm with Orbit (in the UK) and Tor.com, who run cover designs and marketing copy past me for comment before publication, and if I spot errors I suggest fixes (and they listen). But I can see the times changing and other imprints or publishers being cut to the bone on all sides …
Show threadHollieK5d ago
@cstross @raymaccarthy @whybird Not to mention publishers that knowingly push AI-generated books who are going to be trying hard to convince people that they're selling legitimate products.
Show threadCharlie Stross5d ago
@HollieK72 @raymaccarthy @whybird I think AI-generated books are being pushed by unscrupulous grifters trying to game the Kindle Unlimited payments algorithm, and by Big 5 execs who have drunk the AI kool-aid and are too far removed from the actual publishers/production managers/marketers to hear how bad the product is. Once the AI vendors pivot to milking their customers and the token costs rise to reflect the real cost of providing the service they'll get cold feet in a hurry.
Show threadRay McCarthy5d ago
@cstross @HollieK72 @whybird
The KU model cheats legitimate authors and invades user privacy.
A real library has to buy a copy. Amazon get it free.
In some countries there is also a per loan small royalty. Anonymous and purely per borrow.
Amazon monitors which pages you view and how long. That's part of reason for cutting off older Kindles. They can't run sufficiently controlling and surveillance firmware.
Show threadRay McCarthy5d ago
@HollieK72 @cstross @whybird
Older blurbs (< 2023?) better than current. I've seen older republished books with rubbish new blurb.
AI can't even summarise. It shortens.
There is no substitute for a human reading the book and doing synopsis (which is too long for blurb and has spoilers) and blurb (which should not have important spoilers).
Show threadHollieK5d ago
@raymaccarthy @cstross @whybird Ditto for tags, etc. And who does all of this? The author, the marketing team, the readers? Who gets rid of bad blurbs, synopses, and tags?
Show threadMark Whybird5d ago
@HollieK72 @raymaccarthy @cstross My initial off-the-cuff idea was readers tagging and upvoting/downvoting accuracy of tags.
Show threadMark Whybird5d ago
@raymaccarthy @HollieK72 @cstross Nice!