197 karma · joined May 9, 2007
Ah, and then it looks like the losing nodes get marked unhealthy:
> In essence, every write operation goes through a voting protocol, and any replicas on the losing side of the vote are marked as unhealthy—unavailable for reads or writes—until they can be repaired. Repairs are automatic and quick. Because a majority agreed either to accept or to roll back the update, there are still at least two replicas available to continue accepting both reads and writes while the unhealthy replica is repaired.
But not sure how unhealthy nodes are discovered. Maybe the coordinator marks them, and if the coordinator crashes it allows stale reads and detects invalid writes?
tldr; `ollama launch claude`
glm-4.7-flash is a nice local model for this sort of thing if you have a machine that can run it
I just saw it a couple of weeks ago, it's great. I linked to him trying to remember one of those lines, but the whole interview is worth listening to
A small preview: after leaving high school at 14 to travel as a magician, a little more than a decade later he managed to get into grad school at Harvard, largely due to a letter of recommendation about his magic abilities (apparently that's a thing).
I also really enjoyed that he was shy revealing his background in magic to other Stanford faculty until he discovered that Paul Lévy also studied perfect shuffling [2].
[1]: http://www-history.mcs.st-andrews.ac.uk/Biographies/Diaconis...
[2]: https://www.chronicle.com/article/The-Magical-Mind-of-Persi/...
[1] http://graphics.stanford.edu/courses/cs268-16-fall/Notes/tor...
[2] http://graphics.stanford.edu/courses/cs268-16-fall/notes.htm...