[pkaler]

I'm also currently reading "On the Calculation of Volume". It is fantastic.

Meta-comment about the post. I used to read and write book reviews like this all of the time. Not anymore. ChatGPT and Claude can do a just a good of a job. Now I'm looking for what you think, a unique insight, what did you feel from a book review from a humanoid. LLMs do a fine job summarizing.

[rnadomvirlabe]

How do you separate the details included in a synopsis from what someone thought about the book?

The details included in a synopsis reflect what the reviewer found important enough to share. I still value the synopsis of a known reviewer exactly for this reason. If I’ve read books on their suggestion in the past and enjoyed them, I’m inclined to take their advice again. Perhaps they could use LLMs to speed up some of the writing, given their perspective, but I wouldn’t agree with the statement “LLMs do a fine job summarizing” in the context of reviews. The author and their collective reviews matter, something absent with LLMs.

[chrisweekly]

Hmm. I agree with your main point (in your 2nd sentence) but want to address your opening question.

It's possible, even common, to encounter a relatively neutral synopsis (covering e.g. genre, themes, plot summary, and objective attributes of the writing style) which doesn't really pass judgement on the book. IMHO those are all things that don't require a human. A good book review is not a synopsis, it's a harmonizing echo of the ideas and feelings imparted by the author; shared, absorbed, combined and reflected, and resonating out to the broader field of readership. That's part of the magic of books -- the way one person's mind can connect so profoundly to many others' (though only ever via one individual reader's experience at a time) across time and space, even beyond death.

[wodenokoto]

Book reviews are not for summaries, they are for opinion and we’ve lost that.

A good review is done by a reviewer that you’ve come to know.

If she says it’s good I know it’s gonna be good.

[chrisweekly]

Agreed.

Related thoughts:

There are many authors whose future books I plan to read without bothering to check reviews.

When I enjoy something that has a sequel, I nearly always continue the series.

Beyond that, it's some mix of:

- A personal recommendation from a friend or colleague

- Awards / nominations (both as a quality signal, and for cultural literacy)

- Collaborative filtering (for both positive and, rarely, negative signal)

- Human-written reviews (heavily skewed towards one or a few reviews that are thoughtful and clearly indicate the book had meaningful impact on that particular reader, vs generic ratings or popularity per se)

/tangent

[roxolotl]

That’s always been the point of reviews. It’s why people follow certain reviewers. They know the reviewer feels similarly to how they do so the feelings of the reviewer is similar to what they might feel. There’s always going to be some discussion of the content but for a good review that shouldn’t be the focus.

[chrisweekly]

Well said! And thanks for the rec!