The SF Anthology Problem – Solved

classicsofsciencefiction.com

22 points·sohkamyung··8 comments

This is a standard set cover problem. At these sizes it should be possible to find a guaranteed optimal solution rather than heuristics AFAICT.

(Posting here just for completeness…) Yes it can be solved optimally: https://shreevatsa.net/post/set-cover-ip/

I didn't see this 1964 Hugo-winner by Simak on the list. Somebody fixed that.

http://lib.ru/SIMAK/waystat.txt

Hmmm ... I guess that's not a short-story. Now ... what's the title of his short story (same topic) about the gophers with toolboxes who 'bent' a guy's house?

Sorry, the real problem is figuring out which 10 stories are actually worth reading, per Sturgeon's Law.

Well... the anthologies are selected from a much deeper pool of magazines, so it's possible the percentage is higher.

How is this comment relevant to this post?

The whole point of the post is that the author has already told you what 10 stories* are actually worth reading and is now telling you which anthologies to buy to read those 10 stories.

*: Technically it is a list of 275 stories worth reading from across the entire history of science fiction, which is why there is a math challenge to figure out exactly which 22 books you needed to buy to get those 275 stories.

It’s relevant because someone had to make some comments that fall into the “90% of everything” category. :)

Apparently the real problem is getting people to read beyond a headline.

TL;DR: which stories to read was already solved.

Reply on news.ycombinator.com