The Sword and Laser discussion

For those interested, a new book recommendation AI project's just cropped up:

- https://www.librarian-ai.com/ (related HN discussion: https://news.ycombinator.com/item?id=...).

The HN discussion also mentions a few other options in this area which I'd not come across before:

- https://www.gnooks.com/
- https://www.librarything.com/

Hope that's of interest to some... I'm off to read some Peter Clines to see how good the recommendations are, as he came up on both gnooks and librarian-ai's recommendations for me...


---
EDIT: other book discovery links:

- https://shepherd.com/ (human based recommendations)

I really enjoyed all the Clines books I've read. '14' is probably a good place to start

Librarian AI recommended books I’ve already read while gnooks recommended authors I already know. 😆 Recs are fine otherwise.

AndrewP wrote: "I really enjoyed all the Clines books I've read. '14' is probably a good place to start"

A really good book.

We read the 2nd book in that series as an S&L book pick in Feb, 2021 (The Fold) It was one of the guest picks (Rod Simmons) when Veronica was on maternity leave.

Both books are in the same universe, but only loosely connected, with a few of the same characters appearing in both.

I thought "14" was slightly better than "The Fold". Both were really good.

Hmmm... I haven't read Three Body Problem (I watched the whole chinese tv show on youtube though) or Red Rising yet.

Tamahome wrote: "Hmmm... I haven't read Three Body Problem (I watched the whole chinese tv show on youtube though) or Red Rising yet.

"

Not sure how they got Red Rising out of the books you input. It’s more Tunnel in the Sky/Lord of the Flies, except a lesser version of those.

I wish they listed some matching tags...

AndrewP wrote: "I really enjoyed all the Clines books I've read. '14' is probably a good place to start"

Ah thanks; will put that to the top of my list :)

Trike wrote: "Librarian AI recommended books I’ve already read while gnooks recommended authors I already know. 😆 Recs are fine otherwise."

Yeah, it's a shame you can't "swipe right/left" to say you've already read the book & whether or not you enjoyed it... Definitely most of the recommendations I got were books I'd also read / generally pretty mainstream so not discovering any niche gems. I got more interesting results when I mixed genres; as outside of SF/Fantasy I'm less well read, so putting "Wild Swans" (historical/biographical) and "Pillars of the Earth" (historic fiction) alongside "We Are Legion" gave some books I know and like (e.g. "The Martian"... though that comes up for most SF inputs), and some I'd not heard of (e.g. "The Nightingale") so hope to try out... I'm not sensing common themes; more just "Readers who liked this may also like" vibes, but will be interested to see how accurate it is.

Would love it if it could scan your GoodReads, and from that infer "you're into world building and character development; some humour essential; pet deaths a no-go", then recommend books that fit those criteria, filtering out those you've already read... but I guess that's less AI / more algorithmic/heuristic.

John wrote: "Would love it if it could scan your GoodReads, and from that infer "you're into world building and character development; some humour essential; pet deaths a no-go", then recommend books that fit those criteria, filtering out those you've already read... but I guess that's less AI / more algorithmic/heuristic."

That would be interesting if it were sophisticated enough. When my cousin and a neighbor had babies a few years ago, the algorithm kept suggesting baby stuff to us based on the things we bought for them. It really needs to sort gifts from self-purchases, or maybe just take other things into consideration. If I’m having two kids at 54, you better recommend me good nannies, a ticket out of town, or a bottle of speed.

Trike wrote: "John wrote: "Would love it if it could scan your GoodReads, and from that infer "you're into world building and character development; some humour essential; pet deaths a no-go", then recommend boo..."

I wonder what age parents start becoming grandparents and that factors into their algorithm at all?