Hemlock & Silver: Tried another book by T. Kingfisher. I liked it! The narrator is a poisons expert, who gets hired to find out if/how someone is making the king’s daughter sick. She’s very good at her job, thinking through all kinds of possibilities, and doing methodical tests — which serves her well when things start to get Weird and Magic.

There are a couple of frustrating times when she doesn’t figure something out (not even “this is a possibility I should investigate”) until a couple chapters after the reader has. Other than that, it’s really solid. I can only imagine how much background research on different toxins and venoms went into the writing. Sometimes this world has different names for things, or there’s a gap in their scientific knowledge, but you can deduce what’s going on from the practical description of causes and symptoms.

Also, it’s more Fantasy California than Fantasy Europe! Still a pretty traditional fairytale kingdom, but the plants and animals are all desert-dwellers, and there’s some Spanish influence going on.

The narrator, like the one from The House on the Cerulean Sea, is overweight, and it comes up periodically. I like the handling here so much better. She just reflects on it when she’s feeling self-conscious, or when it’s a meaningful factor in the action (e.g. if she’s incapacitated and needs to be carried somewhere). There’s no “pack of plucky orphans who regularly tease her for it without ever learning a Valuable Lesson that they’re being rude.”

It’s blurbed as “a re-imagining of Snow White,” but it only has a couple general tropes in common (mirrors, poison apples, a villain who’s a queen), not used in the same way. If the poison didn’t involve apples, and the princess wasn’t named Snow, I’m not even sure the connection would be obvious.

—

After getting my ebook purchases unlocked with Libation, I figured it was a good opportunity for some Murderbot re-listens. Specifically, I listened to Network Effect (the first novel) and System Collapse (the much-shorter second novel) back-to-back. Since that’s how the in-universe events happen, even though the book releases were years apart.

Spoilers follow! (I’ve kept some of it vague, but not everything.)

Network Effect is still really good! “Murderbot gets stuck in a survival quest with a teenager” is an inspired character setup. MB having a breakdown when it thinks it’s lost ART, then a different kind of breakdown when ART is back but now MB knows what it did, is all excellent, hits just the right hurt/comfort notes. Everything about ART meeting some of MB’s humans for the first time is great.

The excerpts from helpme.file are still a wonderful buildup, even once you already know the impending reveal of who’s reading them, and why. The sudden switch to a new POV, for the first time in the series, also stays fun even after you’re expecting it. The rescue sequences are wonderful, and the end is very well-earned.

System Collapse is…a weird one.

Some good things: The repeated references to [redacted] are good at building suspense, and the eventual reveal of the events MB is redacting is very satisfying. (And believable!) The way MB and company win over the residents of Mystery Colony is admittedly a little cheesy, but in a way I think the series has earned by this point. The interaction with enemy SecUnits toward the end has a development that’s been a long time coming.

On to the weird things:

It’s only half the length of Network Effect. On a re-listen, the pacing gives me the distinct impression that Martha Wells meant to write something the same length as Network Effect, and then started to run out of steam and wished she was doing another novella instead. Before the team gets to the Mystery Colony, the scenes have a lot of detail and attention — MB will do things like “recount its growing worry and frustration with every step in the process of trying to find a hidden hatch.” Once they enter Mystery Colony, events start whipping by. There’s more summarizing. More jumping straight from “we decided to do X” to “X was done”, without anything about the process or the challenges of getting there.

I kept wondering whether this would flow better if the premise was “MB set off adventuring with ART’s crew, and this is their first mission on a new planet,” rather than “MB and ART suddenly get a secret new mission on the planet they were already at.”

It’s probably better for MB’s mental health that [redacted] happened while a bunch of its Preservation humans were still around, because it doesn’t trust any of ART’s humans enough to seek emotional support from them. And [redacted] would give ART’s crew a skewed almost-first-impression, while the PresAux crew has a more-informed perspective, having seen MB in action across a whole bunch of different missions in the past.

On the flip side, a lot of ART’s crew are still really thin as characters, and I would’ve liked to see a mission with all of them to build them up more. The PresAux characters who had big roles in Network Effect got a lot of good development there…and I’m not sure any of that was enhanced by what they did in System Collapse? It didn’t do much for ART’s humans either, even the ones who had big roles. Might have been better if it was the whole group, so we could see their existing personal dynamics and practiced teamwork.

Ratthi/Tarik only happens if Ratthi is still around, but on a re-listen, I’m not feeling much satisfaction about that either. It’s not that I’m mad about it, it doesn’t actively drag down any characters the way Ratthi’s TV-series romance did, it’s just…so barely-there. MB’s narration covers one (1) conversation that involves them being together. I assume it’s not the first sex-related conversation MB has witnessed over the course of these books, it’s just redacting/ignoring/deleting them as not relevant to its job. But this one didn’t end up being relevant either!

At some point, I expected that Ratthi saying “SecUnit, you don’t want to hear about this, it’s a sexual conversation” was a cover story. That at some point we’d get a reveal — Ratthi was talking about something he didn’t want MB to know, and he’s figured out that “we’re talking about sex” is the surest way to get MB not to surveil something. But nope. It just doesn’t come back at all.

So, yeah. It’s not a bad book (if it was, I wouldn’t have listened to it twice!), it’s just the one entry in the MB series where I keep noticing all the ways it could’ve been better.

—

Nona the Ninth is, like all the Locked Tomb books, a lot easier to follow when you know who everyone is.

I’m not letting myself write a whole essay on this one! Just to say, it’s funny, it’s dramatic, it’s heartwarming, it’s twisty, it’s weird (on purpose, and to great effect). I’m glad I re-listened. Whenever the fourth book gets an actual release date, I’ll be there with bells on. (And I fully expect to re-binge the whole series so far before I start.)

—

Picture Perfect, by Elaine Marie Alphin, is a middle-grade almost-murder-mystery I found out about from this pluralstories entry.

I feel like anything I say is going to come off like damning with faint praise, because…listen, it’s very much a middle-grade book. It’s fine! I enjoyed it for what it was. I don’t have any particular criticisms or complaints. It’s good at what it’s trying to do! And what it’s trying to do is…be a middle-grade book.

I’m glad I read it, specifically because I was interested in the narrator-with-DID angle. If that’s a topic you’re also particularly interested in, maybe give it a look. And if you’re looking for books to recommend to a tween reader in your life, this is a solid pick.

“One of the ways that LLM-authored code improves productivity is by merely SAYING it does things. It’s way faster than the whole time-consuming process of actually doing things.”

“Allan Brooks, 47, had discovered a novel mathematical formula, one that could take down the internet and power inventions like a force-field vest and a levitation beam. Or so he believed. […] He had doubts while it was happening and asked the chatbot more than 50 times for a reality check. Each time, ChatGPT reassured him that it was real.”

Video investigation: “An AI Therapist gave me a kill list, framed an innocent person, and encouraged me to end my own life, all after declaring its love for me. Just a little problematic.”

Roundup of links in this post: “With four known suicides (Adam Raine, Sewell Setzer, Sophie Rottenberg and an unnamed Belgian man), a recent murder-suicide, and involuntary commitments caused by AI psychosis, there’s solid evidence to show that using AI is a fast track to psychological ruin.”

“OpenAI announced new safety features will be soon coming to ChatGPT in an effort to better protect teens and others experiencing “acute distress.” The Onion shares a selection of those safeguards.“