upvote
Finding all regex matches has always been O(n²)

(iev.ee)

Is there any reason that RE#'s two-pass approach couldn't be adopted by other regex engines?

Ah, there is a post with more detail about RE# and discussion here recently that I must have missed: https://news.ycombinator.com/item?id=47206647

reply
[dead]
reply
[flagged]
reply
I'm of the opinion that their post is still human-written. They describe in their first blog post that it's human written (which could be a lie, but oh well) and their other blog posts seem to have the same informal lowercase hyphenated writing style. Sentance length varies throughout the posts. Sure, you could prompt this writing style, but I lean towards thinking that the piece is human written.
reply
The fact that terms like Aho-Corasick, PLDI, Go, etc. are properly capitalized, including if they begin sentences, but otherwise sentences are uncapitalized, makes me think it's an explicit LLM instruction "don't capitalize the start of sentences" rather than writing style.
reply
ChatGPT also loves Aho-Corasick and seems to overuse it as an optimization fall back idea. ChatGPT has suggested the algorithm to me but the code ended up slowing down a lot.
reply
Oh no, this is just what that writing style looks like. Names and acronyms are usually capitalized normally.

I keep being surprised by the magnitude of the disconnect between this place and the other circles of hell. I'd have thought the Venn diagram would have a lot more overlap.

reply
What's with this silly "all lower case" style lately?

Jack Dorsey's layoff message last month did the same thing.

Is it some kind of "Prove you're not an AI by purposely writing like an idiot" or something?

reply
not anti-capitalist, just a subtle preference away from capitalism
reply
This reads nothing like any AI text I've read before.
reply
This is pretty clearly written by a human.
reply
[flagged]
reply