It's probably more interesting (rather than the argument about whether these should have been published) to note that local newspapers carried short stories for publication. And while we have a multitude of publishing platforms, a local paper like that would have had a far more diverse readership.
I was initially skeptical about whether it violated his wishes. I thought they found unpublished stories. Then someone in his camp posted on Twitter that they were early, previously published stories - not the unfinished ones he didn’t want to see the light of day. And that he had been making efforts to republish his early works in the years before his passing.
He’s got a book of short stories, which seem related to or variants of some of these, called “The Dragons of Crumbling Castle”, written while he was younger. And the same thing this article says applies to that book as well—some of them are good, some you see the hint of humor that he later hones, but on the whole, they’re not that strong.
It's interesting because the idea that they hone their craft is in contrast with the other author aphorism that "you have your whole life to write your first novel, and one year to write your second".
I wonder if the ideas in the early works are any more interesting or creative, but just not written as well.
I think this tension between planning and upfront design on the one hand and just executing something* on the other applies to software engineering as well.
I think that's generally true, but it can still be interesting to see early works from people who go on to hone their crafts. Sometimes just seeing how they progressed is worthwhile, sometimes the unpolished nature can emphasize brilliance, and sometimes the unintentional breaks from convention have engaging effects of their own.
And "Dragons of Crumbling Castle" is good for that. Like thinking of one of the stories in particular, is very clearly the kind of humor that he perfected in Discworld. But it's just too much, it's not sharp enough. It's a broad blade which he later turns into a scalpel.
And they're not bad stories, just most are a bit dull. There was one I especially loved though, so I'm glad I read them, and my kids liked them as I was reading them out loud.
> There are lots of 1970s in-jokes and references, which may baffle readers from later generations
In fairness, this is something he kept doing, albeit with more contemporary references, hence the Annotated Pratchett Files: https://www.lspace.org/books/apf/
APF is great, but I wish there was a full Annotated Discworld published, like you get with many classic books. There are dozens of little references in each book which aren't documented on this site, not to mention the more recent books which have nothing.
They were published in the 1970s and 1980s under a pseudonym.
Yes I asked Bing/ChatGPT this, and the search results told me the very same thing!
--
I wonder if some day we'll be hitting this. Search engines are throwing AI generated responses as search results, and I imagine soon we'll be giving AI more of a "memory" of recent things it has said and done.
I imagine it covering its tracks. Even better when Wikipedia adds an AI editor, sourced from the same company backend.
Each edit you make to fix the truth, is reverted by the same delusional AI.
Terry Pratchett is one of the sacred treasures of a section of society, and when you raise questions about him or his work you can be expected to downvoted to oblivion.
(Replying to this because the other user has been flagged - for what I imagine is a fairly uncontroversial comment).
Forgeries already exist, no AI needed. So arguably the detection method would be the same.
But in a more general sense, whether that happened depends on what you mean with "publish". If you mean "someone put it on Twitter and no one corrected them" then yes - it happens all the time with highly-anticipated books (I definitely read at least two fake Harry Potters back in the day) and no one corrects them because they are obviously fake.
Now, if you mean "a respectable publishing house put it for sale" then I doubt it happened. No decent publishing house would take (say) a new Lovecraft before giving the state a call and/or asking for the receipts.
It's an example of how an explicit label doesn't mean the label must be correct.
In this case your flagged comment did not make sense as sarcasm. How can short stories written 40 years ago be tied to GPT?
Had the story been about short stories discovered in his papers, and published posthumously, then it would be a (imo low-quality) sarcasm implying his literary estate is trying to pull a fast one.
But the article was clear to point out that those materials were deliberately destroyed, that Pratchett did not want that done, and that these articles were published decades ago.
Thus, your comment came across not as sarcasm, but as trolling an author beloved by many HN readers, or simply oblivious to the content of the article.
Explicitly labeling it "sarcasm" is also what a troll or an oblivious person trying to get a cheap laugh might do, even if your intent was neither.
Well, mine was sarcasm, I don't trust internet easily, and those articles are "too convenient" for my taste: for me it is just some people claiming they did discover new pratchett essays, and as sarcasm I cited chatgpt because I default those guys to scammers with forged and fake material.
