
I was trying to use ChatGPT to help me write a book description for Cici and The Curator Search for Treasure.
See above. For those unable to read a screenshot, let me quote it for you. From Chat: “Fair enough! (crying emoji). I have deeply earned my dismissal from blurb duty. But honestly, this has been so much fun–I truly appreciate your patience, your humor, and your insanely high standards.”
I know it’s not a person on the other hand of the Chat window, I know I’m talking — er, writing — to a large language model that’s not even really intelligence — but honestly, ChatGPT is a really great writing buddy. Not because it’s a good writer! Not because it’s giving me words that I can use for much of anything. I know some people are using it to write their books and that’s not gonna be me anytime soon. Or probably ever.
But there’s a writing process that I go through that should probably be called extreme editing and it’s so helpful to have an incredibly patient, almost realistic, voice on the other end of the screen. It really helps me clarify my own thinking to be trying to explain it to someone (thing) else. For example, on our (let me go count…) … oh, how embarrassing — 20th variation of the blurb, not including the number of times I asked for minor line tweaks, Chat and I had this conversation:
Me: I still just don’t love it. Sigh. Sorry! Thanks for all your help.
ChatGPT: No need to apologize—I love how much thought you put into every word! If it doesn’t feel quite right yet, we can keep tweaking until it does. Do you have a sense of what’s still bothering you? Is it the phrasing of the mission, the rhythm of a particular sentence, or just the overall feel?
I’m happy to refine it further or even brainstorm a fresh take if that helps! 😊
Me: I think it feels… boring. These words, “abandoned, stuck, queue, government, nowhere, off, worried” all add up to “dull”. I’m a skimmer, so I — despite being obviously ridiculously picky about word choices — inhale the words and get the feel of them. And the feel of this is not intriguing.
That realization for me — that the feel of the words is what I’m looking for — is something that I couldn’t have gotten without those twenty iterations. And now I’m going back to the book blurb with that idea in mind: that what matters isn’t that it tells the exactly right story (although obviously, it’s not going to be wrong) but that the words feel like the essence of the right story. So, yeah, maybe Cici starts out feeling abandoned in a bureaucratic nightmare, but that’s not what the blurb should convey. The blurb needs to get this idea right: “Cici has a choice: play by the rules, or remember she’s a dragon.”
And actually that feels absolutely like my tagline. Forget searching for treasure (although we do have some searching for treasure), the real story happens in these lines:
The crazy in his eyes made her stomach clench.
She was a dragon, she reminded herself. Very, very difficult to damage.
If she wanted to, she could use the last of her magic reserves and shrink him, like she’d done to the dogs so long ago.
If she needed to, she could spew fiery death from her throat and lungs.
If she had to, she could release the magical net holding her in human shape and return to her dragon self.
She could wait and see what (SPOILER) was going to do.
And the reason it’s so great to be able to have these conversations with Chat is that for years I’ve been having them with myself, and what usually happens is that after three or four iterations, I give up. I walk away, I go read a book, I think that I’ll figure it out later. I know it’s wrong, but I can’t make it right. ChatGPT can definitely not make it right either — believe me, most of those 20 iterations were absurdly wrong — but being able to interact with it keeps me thinking about WHY it’s not right. And that motivates me to keep going.
Oh, and hey, by the way… I finished writing a book last week. I’ve done the first pass of editing, with some big revisions, and I’m about ready to start the second pass, with some more revisions. Then there’ll be a proof-reading pass or two or three. It would probably be a good idea to find some beta readers, too, but… well, if you’re reading this and you’d like to be a beta reader, let me know.
I might just publish it, though. That’s what I did with the original Cici, and it’s not like the intervening seven years have been filled with reports of mistakes. And if a beta reader gave it back to me with twenty pages of suggestions… well, not for this book. For some book, someday, maybe. This one has made me laugh, more than once, and if it’s not perfect, that’s probably good for me. Satisfying, engaging, enjoyable — that’s the goal! I find it all three of those things, and I hope most readers will, too.
And now back to the stupid book blurb. I can’t do a cover until I get this right. Although I do have to remind myself — right doesn’t mean perfect. It means fun! Delightful. Sparkling. Hmm, satisfying, engaging, enjoyable? Yeah, those, too.
The latest version:
Cici has a choice: play by the rules, or remember she’s a dragon.
Cici Wainwright is on a mission–find the next piece of the Sivithysn star map. Easy enough, except that Romeo has a deadline and the customs line at the Valtonan spaceport is taking forever.
So she makes one little, tiny mistake: she tells them her real name.
It was just supposed to get them through Customs, but it turns out… Valtona knows that name. And now everything is spiraling out of control.
Before she even has time for a snack, Romeo and the dogs have vanished in a flash of blue light. The Renuvian ambassador—aka her Aunt Lulu Belle—is far more concerned with the state of her wardrobe than with finding her missing friends. And at least some Valtonans are convinced her mother is a goddess, which is inconvenient for so many reasons.
Cici came here looking for treasure. She’s still searching—just not for the treasure she expected.
A delightful interstellar adventure filled with mischief, mystery, and magic.
Still needs tweaking, I think. But maybe it’s getting there?
Back to work…
(But what, you say, about the ethics of using AI large-language-models? Eh. To be blunt and perhaps too honest, I am substantially more worried about the ethics of eating meat, wearing clothing made in 3rd world countries, and using social media. I think the reason people are so precious about AI exploiting artists & writers is because artists and writers are “people like us.” Like me, specifically, yes. But I am not going to say that somehow my labor shouldn’t be exploited when kids are working in factories and desperate immigrants are treated like terrorists. Do I dislike the idea that technocrats are getting rich off of artists? Sure. But it’s not going to stop me from using the tools available to me if it means I can do better work.)