I think it's more the keming of the domain portion of the HN title, especially combined with HN's rather small font size choice (it's a meager 8pt¹!) there, and that it just happens that the mis-kemed result ends up with "John Mastodon", and is thus not trivially noticeable as "wrong"…
(I read it the same way, too.)
(¹I personally have a browser override for HN's tiny font choice; I thought that 12pt was the universally agreed upon "base text" point size, and "10pt" was "small text", but HN's "normal" is 9pt.)
Agreed. I need a larger font on a lot of sites nowadays, but HN is probably the one that behaves best with simple browser zoom. I have it set to 125 or 150% depending how tired my eyes are..
That would be next-level immersion! You could probably achieve this by rendering the LLM’s response using a handwritten font—maybe even train a model on your own handwriting to make it feel truly personal.
Exactly! There’s something about handwriting that makes it feel more personal—like scribbling notes in the margins of a spellbook. The shift from typing to pen input definitely changes the vibe of interacting with AI.
I generally listen to white noise when I am trying to concentrate, but its Dave Clarke's techno/electo podcast, "White Noise"! https://www.youtube.com/watch?v=36FkrrpzFOA