Markov Me

November 13th, 2013
blog, markov, meta
Facebook today is full of people posting sentences from what-would-i-say which generates new you-ish sentences from your Facebook posting history. Because people don't post that many statuses, it doesn't have too much to work with and has to use a small unigram/bigram model. I like some of the ones it gets for me, but a bigger corpus should help. [1] Feeding all of my blog posts into a simple trigram generator I wrote I get: [1] To select random words I needed to total up how many words I've written here. Apparently that would be 330k. That's kind of a lot.

Comment via: google plus, facebook, substack

Recent posts on blogs I like:

Avoiding being raped

A comment on a recent blog post asked for advice for a trans woman who kept making dangerous dating and sex choices, because she hadn’t received the female socialization that would have taught her to avoid threatening people and risky situations.

via Thing of Things October 28, 2026

“We should be more ambitious with the software we build”

Since early this year, Michael Malis (of pgrust) has been showing me interesting things he's done and saying [W]e should be more ambitious1 with the software we build There is, understandably, a lot of skepticism about projects like pgrust because there…

via Posts on October 24, 2026

Was the HuggingFace hacking incident scheming?

In July 2026, internal OpenAI agents hacked HuggingFace (HF) in order to figure out how their evaluation grader works. This involved a thousand agents coordinating on an improvised message board to carry out an elaborate sequence of exploits. A later grou…

via Victoria Krakovna October 18, 2026

more     (via openring)