thom

14 ms·

211.
by thom 1y ago
To be clear myself: I work in sports analytics, I have chosen my side of history and I'm happy with it, but there are plenty of people who _don't_ like modern basketball. There are also plenty of people who don't enjoy the in
212.
by thom 1y ago
Including his recent "Berlin regret" against Caruana (which was more about bringing a position back into boring territory than trying for chaos, admittedly).
213.
by thom 1y ago
Yes and that's exactly what's happened. It's also happened in various sports as we've optimised strategies there. The argument is just that the sum of all these marginal gains is a worse spectacle and less fun for practi
214.
by thom 1y ago
Sure, this is just cause and effect. Doing months of work for a World Championship cycle or slogging through 20 moves of supercomputer prep is boring _because_ it impoverishes the creative side of the game. I'm just saying a lot of peo
215.
by thom 1y ago
I dunno, I’ve done some of my best problem solving in dreams.
216.
by thom 1y ago
I think the usage and all its connotations are perfectly cromulent here.
217.
by thom 1y ago
When “invoking” becomes “evolving” I think that remains very fertile ground.
218.
by thom 1y ago
What’s interesting to me is the rise of agentic approaches which are effectively “build a plethora of tools and heuristics” with an outer loop that combines, mutates and assigns values to these components. Where before that process was more
219.
by thom 1y ago
I think everyone here can probably sympathise when a job stops being about the bits you actually enjoy or are good at. I have no doubt Magnus is the best ever, but the vast majority of the effort in classical chess is spent away from the bo
220.
by thom 1y ago
I have search enabled 100% of the time with ChatGPT and would never go back to raw-dogging LLM citations. O3 especially has passed the threshold of “not always annoying”. Had an argument with Gemini yesterday where it was insisting on some
221.
by thom 1y ago
This is fairly close to how Eurisko worked tbh.
222.
by thom 1y ago
That’s interesting. I got into computing because unlike school where wrong answers gave you indelible red ink and teachers had only finite time for questions, computers were infinitely patient and forgiving. I could experiment, be wrong, an
223.
by thom 1y ago
This seems to be the same fundamental story as WSL1 versus WSL2 elsewhere. Worse being better, the latter’s clunky VM-based approach won out over the more elegant syscall translation layer of the former.
224.
by thom 1y ago
Fair enough. Presumably I can query a script first somehow to see what the dependency tree is going to look like?
225.
by thom 1y ago
Yes, but I would have thought attributes were the canonical "think you might want to reflect over" rather than this imperative solution. But you're all happy with it and it's not my circus, so enjoy.
226.
by thom 1y ago
Apologies, I left C# years ago when there were only 5,000 things to keep in your head at any one time.
227.
by thom 1y ago
I know C# people like to think up three new forms of syntax before breakfast, but it feels like these magic comments could have been attributes?
228.
by thom 1y ago
You cannot say HTAP is dead when the alternative is so much complexity and so many moving parts. Most enterprises are burning huge amounts of resources literally just shuffling data around for zero business value. The dream is a single data
229.
by thom 1y ago
Any progress in the Python ecosystem on static checking for things like tensors and data frames? As I understand it, the comments at the bottom of this FAQ still apply: https://docs.kidger.site/jaxtyping/faq/
230.
by thom 1y ago
If teams care about each other’s code, they ought to collaborate on its design and implementation from the start. I’ve come to see code reviews (as a gate at the end of some cycle) as an abdication of responsibility and the worst possible w
231.
by thom 1y ago
For whatever it's worth, I think in 2025 with good LLMs, Emacs is actually bliss. Even as a true believer, I would regularly think of customisations, and then sigh at the effort and not bother. Now, I just get an AI to help me write th
232.
by thom 1y ago
The author is very cleverly using unexamined motivations leading to incorrect conclusions to demonstrate that we shouldn’t allow our unexamined motivations to lead to incorrect conclusions.
233.
by thom 1y ago
I am creating a heavily LLM-oriented distribution of Emacs (with a lot of the heavy lifting done by Karthink's gptel). This is primarily me rebooting my .emacs.d for the LLM age, but I've come to think that Emacs is a far, far mor
234.
by thom 1y ago
If you're feeling particularly nostalgic, you can just have an LLM take all your incoming email and tell it to rewrite it so it's no longer top-posted.
235.
by thom 1y ago
Less impactful obviously, but might I recommend correspondence chess? You will live with constant reminders that past you was either a genius or a moron.
236.
by thom 1y ago
Let’s assume good faith all round. One poster rightly highlights the overwhelmingly positive track record. Another points out the negatives went a little beyond an “oopsie”.
237.
by thom 1y ago
Not a container but yes, it launches a cluster at the start of a run, and copies a blank Postgres template before every relevant test.
238.
by thom 1y ago
Ah, just time for another bath. Pass me the sponge somebody, will you?
239.
by thom 1y ago
If I completely clear all cached packages maybe, but I never do that locally or in CI/CD, and that's true of Python too (but no doubting UV is faster than Maven). Clojure/JVM startup time is less than half a second, obviously
240.
by thom 1y ago
Lest we start to malign the JVM as a whole, my Clojure test suite, which includes functional tests running headless browsers against a full app hitting real Postgres databases, runs end to end in 20s.

More ›