thom

9 ms·

91.
by thom 8mo ago
Juggling is one of hobbies with the highest ratios of being able to impress random people versus the actual effort you have to put in, and I generally find I never forget 3-ball stuff I learned as a kid. It's also as good as a long wal
92.
by thom 8mo ago
You need the model to interpret documentation as policy you care about (in which case it will pay attention) rather than as something it can look up if it doesn’t know something (which it will never admit). It helps to really internalise th
93.
by thom 9mo ago
Wang’s Carpets usually comes up alongside Solaris as another example of deliberately alien aliens.
94.
by thom 9mo ago
This is a minor thing, but as an introvert, I really try and push myself to model social behaviour to my kids. Saying good morning to people in the street, chatting to other dog owners, being nice to waiters, travelling by bus, there are lo
95.
by thom 9mo ago
Also increasingly well integrated into Godot.
96.
by thom 9mo ago
I installed Bazzite on a slightly esoteric machine, a 16" dual-screen Asus laptop. It's not really my cup of tea as a distribution, either philosophically or practically, but it has some specific patches for Asus hardware and as a
97.
by thom 9mo ago
The browser versions (and the mobile versions) are nowhere near parity with the desktop Windows version, even in quite basic matters of styling. To be honest this annoyed me enough in the end that I just moved to a PandaDoc/CSS/PD
98.
by thom 9mo ago
Anecdata/placebo/whatever: I use BioGaia's Gastrus tablets and they increase my quality of life noticeably, and I can tell when I've been off them for a while. I got refrigerated deliveries of probiotic yoghurt drinks fo
99.
by thom 9mo ago
I use ChatGPT to find products and opinions about those products, same way I'd use Google. I'd never just click on the top sponsored result in Google and buy it, and I'd never just click on a result in ChatGPT and buy it. And
100.
by thom 9mo ago
I got my youngest a ZipString Aracna for Christmas (entirely driven by a tiny cameo in the film Weapons). It is rare that a toy for kids feels magical, but it's been a real hit over the holidays.
101.
by thom 9mo ago
Even in the case where you have a good shortlist of items, the ability to then ask follow up questions in a conversational format is very useful for me. Anyway, just explaining why one might use ChatGPT for this rather than the Google searc
102.
by thom 9mo ago
DeepSeek certainly managed that on the training side but in terms of inference, the actual product was unusably slow and unreliable at launch and for several months after. I have not bothered revisiting it.
103.
by thom 9mo ago
The regular google search AI doesn’t do thinky thinky mode. For most buying decisions these days I ask ChatGPT to go off and search and think for a while given certain constraints, while taking particular note of Reddit and YouTube comments
104.
by thom 10mo ago
Is there any path for Microsoft and NVIDIA to work together and resurrect some sort of transparent SLI layer for consumer workloads? It’d take the pressure off the high end of the market a little and also help old cards hold value for longe
105.
by thom 10mo ago
This, for me, is a perfect description of the entirety of GraphQL tbh.
106.
by thom 10mo ago
Would D&D not work for you?
107.
by thom 10mo ago
All of this should be familiar to the average programmer. The Creator (I call him Colin), rushed for time and bereft of more elegant ideas, has widely deployed lazy loading as a solution to the finite processing power of the universe. Almos
108.
by thom 10mo ago
A corollary to this, for me at least, has always been: don’t try to fake your way into an opportunity. Be yourself, let people pass on the real you if they want, because you don’t want to land a job where you can’t thrive as yourself. Obvio
109.
by thom 11mo ago
I think this expectation that data science code is a thing you write basically top to bottom to get some answers out, put them in a graph and move on with your life is not a useful lens through which to evaluate two programming languages. R
110.
by thom 11mo ago
It’s not that you can’t model data that way (or indeed with structs of arrays), it’s just that the user experience starts to suck. You might want a dataset bigger than RAM, or that you can transparently back by the filesystem, RAM or VRAM.
111.
by thom 1y ago
Give me this, but with a very efficient, opinionated path to put models into production. Give me accessible PM and customer friendly documentation about features and model choices at every stage. Make it reusable and easy to modify. Make it
112.
by thom 1y ago
I think this captures the technical track dual to CEO 'founder mode'. As cringeworthy as many found that term, it captures the plain truth that there are certain types of changes that only people at the top are (or at least feel)
113.
by thom 1y ago
If, after 100 years of cars being available, everybody still rode horses and always got where they're going on time, and car people constantly blogged about it, this would be a great analogy.
114.
by thom 1y ago
My mum spent much of our youth, upon return from yearly holidays to her small seaside home town, tumbling rocks. We were kinda cluttered people growing up, so among other things (an extensive sci-fi collection, hundreds of kitchen gadgets,
115.
by thom 1y ago
People just forget stuff. All this was said about Tiger. Snow Leopard was an entire paid OS upgrade the only selling point of which was that it made Leopard less crap. Your battery used to expand, your GPU used to overheat, you used to star
116.
by thom 1y ago
Yeah, played both of these on the ST, absolutely loved blowing up weird alien squid things and collecting loot.
117.
by thom 1y ago
Is there a more compelling set of examples somewhere?
118.
by thom 1y ago
The Far Side is like the Simpsons for me, many of the jokes have passed into everyday expressions. If I or my wife are on the loo when the other comes in, we’ll always quote one of my favourites: https://static0.srcdn.com/wo
119.
by thom 1y ago
It depends what you're doing. I can get valuable work done with the subset of Torch supported on MPS and I'm grateful for the speed and RAM of modern Mac systems. JAX support is worse but hopefully both continue to develop.
120.
by thom 1y ago
No, this is the exact opposite of my experience. I have little interest in the AI's code, it can be very illustrative but I find it ugly, unmaintainable (for either humans or itself after enough iterations) and regularly wrong. But it&

More ›