bjourne

22 ms·

631.
by bjourne 1y ago
You come off as whiny and entitled. My advice is to work on your social skills and how you present yourself because that will give you the most bang for your buck.
632.
by bjourne 1y ago
It probably could be very useful for neovim. The lua runtime it uses doesn't support fine-grained process management which is very important for any program with untrusted plugins.
633.
by bjourne 1y ago
Every experienced dev knows what happens when they are subjected to issue trackers with loads of features: some managers require all those features to be used (cause that's what they paid for!). So instead of spending your time debuggi
634.
by bjourne 1y ago
> All these use cases work great, I save a lot of time. So why are you complaining? I use AI all the time to give me suggestions and ideas. But I write the perfect code myself.
635.
by bjourne 1y ago
Totally agree. GitHub issues is probably the best bug tracker/ticketing system you can imagine. Intuitive interface, simple, clean, and fast. Just waiting for a Microsoft redesign to completely fuck it up. :P
636.
by bjourne 1y ago
https://www.paulgraham.com/submarine.html
637.
by bjourne 1y ago
LLMs is a subset of sequence learning which is a subset of ML which is a subset of AI. But even in the LLM subset the field is so deep and vast that you can only learn one small subset. It's a subset of a subset of a subset of a subset
638.
by bjourne 1y ago
That weird neighbour of yours who like to meddle in others affairs? They've done it before...
639.
by bjourne 1y ago
There has been lots and lots and lots of attempts to replace steel with wood in construction. These attempts have gone nowhere. So what is to say that this time it will be different? If wood is so good for tall construction why isn't i
640.
by bjourne 1y ago
Blötsnö, nysnö, pulversnö. I can make up new ones too: lastbilssnö (truck snow). With agglutenative languages counting words doesn't work.
641.
by bjourne 1y ago
Pairs of resumes and job descriptions with binary labels, one of the hired person was a good fit for the job, zero otherwise. Of course to compile such a dataset you would need to retroactively analyze hiring decisions: "Person with re
642.
by bjourne 1y ago
You can dilute soap and shampo (if you have regular men's hair) 3:1 with water and it will still work fine.
643.
by bjourne 1y ago
Nothing nihilistic or conspiratorial about it. Money gives you power over other people. You having more power over others means by definition that they have less power -> Zero sum.
644.
by bjourne 1y ago
Wealth tax is a thing.
645.
by bjourne 1y ago
Gates giving away his money is good, but it would be better for the billionaire class not to exist. Money is power is a zero sum game.
646.
by bjourne 1y ago
Progressive tax scales solves that problem neatly.
647.
by bjourne 1y ago
> Extra challenging this year with tensions running rather high because of Israel's participation. If they can manage to pull this year off without any significant disruption, then the overall security and orchestration really are t
648.
by bjourne 1y ago
I think the parent's point is that if Google with millions of servers can't make performance optimization worthwhile, then it is very unlikely that a smaller company can. If salaries dominate over compute costs, then minimizing th
649.
by bjourne 1y ago
Apple unlike the other Silicon Valley giants has figured out that latency >>> throughput. Minimizing latency is much more important for making a program "feel" fast than maximizing latency. Some of the apps I interact wit
650.
Notes on the Origin of Implicit Regularization in SGD(inference.vc)
1 points by bjourne 1y ago|0 comments
651.
by bjourne 1y ago
Pyright is only for type-checking and it lacks many features you'd expected from a modern LSP (I forgot which). Hence, it was forked and someone created basedpyright to fix it: https://github.com/DetachHead/based
652.
by bjourne 1y ago
Quite the contrary. The idea behind the semantic web was to make content machine-readable by manually annotating it . For instance, this comment would have fields like "author", "date", "language", and maybe &
653.
Exploring Timm vs. Torchvision Resnet18 Difference(blog.problemsolversguild.com)
1 points by bjourne 1y ago|0 comments
654.
by bjourne 1y ago
Can you give an example?
655.
Ask HN: What do you use for spell and grammar checking?
5 points by bjourne 1y ago|5 comments
656.
by bjourne 1y ago
Ime it doesn't work well on old gpus. E.g., on my Quadro P400 desktop I still use xorg because it is much snappier. On my new laptop with Arc it is still slower, but I can live with it since the animations are smooth.
657.
by bjourne 1y ago
In order to save its industry from the Nazis, the USSR moved over 1,500 large factories eastward in under six months in 1941. So the US in 2025 could certainly build at least a few factories in under a month. It would be a complete waste of
658.
by bjourne 1y ago
He's not wrong. It is technically possible. However, transforming the US into a command economy may not be the best thing to do.
659.
by bjourne 2y ago
What's your take on man vs. machine? If AI already beats Master level players it seem certain that it will soon beat the Geoguessr world champion too. Will people still derive pleasure from playing it, like with chess?
660.
by bjourne 2y ago
I'm 100% sure smug Haskellers have something very important to say here. :)

More ›