Charlie Loyd @vruba@everything.happens.horse

I’m at the point in this ML essay where I’m imagining Ludovico Vicentino degli Arrighi as a YouTube influencer. I may be a bit out of practice writing.

I would really like to live in a world where some small but significant proportion of academics get to consider it part of their job to collaborate with news outlets on accessible and self-contained explainers of stuff in the news.

Mildly concerned that the LLM explainer I’ve seen get the most traction is by a someone whose professional and personal behavior seem very questionable. (And who is also very litigious.)

For what it’s worth, some places I do subscribe to are:

LAT: https://www.latimes.com/

HCN: https://www.hcn.org/

Oaklandside: https://oaklandside.org/

Each of them has done things I disagreed with, sometimes deeply, but not with the sort of broad scope and ugly conviction that the NYT has shown.

Tired of feeling bad about what I’m paying for with my crossword subscription.

I was having unbidden pansharpening thoughts again, so I fixed typos and a pseudocode bug in my old 10,000 word gist on the topic: https://gist.github.com/celoyd/5bb5417b24801e0446ad5977cc3581e4

Proud to say that it’s the best non-academic introduction to the topic on the entire internet. Also, I admit, the worst, because it’s the only one. I don’t know why there aren’t more people interested in this.

Got that feeling I got when I saw the video of the first trucks full of covid vaccine heading for the highway. https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(23)00026-9/fulltext

Sitting here angrily remembering the time I guessed that the distance between two randomly chosen points in a cube might be 1/√2, looked it up, and learned that it’s (1/105) × (4 + 17√2 - 6√3 + 21×ln(1 + √2) + 42×ln(2 + √3) - 7π) ≈ 0.661 and this was only worked out in 1978.

LLMs don’t actually know English grammar, they just abstract patterns from many examples of utterances in English. You and I, of course, humans, with a conscious mind that we can see and understand via the simple expedient of naïve empiricism, we simply read that book that fully enumerates the rules of English grammar (you remember? the book?) when we were infants. QED.

There are a lot of people who think they are making really impressive critiques of LLMs when they are doing the equivalent of saying “Dogs can’t eat chocolate because truly experiencing chocolate requires a spark of the divine, and as as we all know, dogs do not have souls. So don’t feed dogs chocolate.”

You have arrived at the correct conclusion that maximalist claims about LLMs are very wrong. But you did it with reasoning worse than the LLMs can do.

If I were a journalist covering the balloons, I would be talking to Ben. He seems to be finding them in open data, which provides some actual facts in a story that’s very short on them. https://mastodon.social/@benelsen/109881189772230247

The kids on TikTok are obsessed with this new trend called the Don’t Uncritically Platform Dangerous Transphobia in a Paper of Record and Deflect Valid Criticism Like a Coward challenge and you’ll never believe the dance that goes with it!

Last I played with consumer GNSS hardware much, the SiRFstarIII was considered very cool.

I just ran a new NEO-N9M overnight under a residential roof (no survey mode or anything) and got this wonderfully clean pattern.

All units are meters. I don’t know what pAcc actually is and neither does the official documentation.

This is like $50 for a USB-capable board with shipping and tax from a reputable seller. I’m delighted.

There’s what I would describe, in technical terms, as a “lazy fantasy map” vibe to the distribution of NGS survey markers.