diggan

29 ms·

451.
by diggan 1y ago
> I think cloud-based async agents like Codex and Jules are superior to the Claude Code/Aider/Cursor style of local integration Ideally, a combination of both I feel like would be a productive setup. I prefer the UI of Codex wh
452.
by diggan 1y ago
80% of DJing is the generalized skill of picking music that fits the vibe, that you can practice with any software. 10% is doing transparent and/or engaging transitions and the rest of the 10% is technical chops, being fluid and being
453.
by diggan 1y ago
This is literally what they did for GPT-OSS, seems there was coordination to support it on day 1 with collaborations with OpenAI
454.
by diggan 1y ago
In the case of the GPT-OSS models, the worst (time consuming) part of supporting it is the new format they've trained the model with, "OpenAI harmony", in my own clients I couldn't just replace the model and call it a da
455.
by diggan 1y ago
And also, different tempos per instrument :)
456.
by diggan 1y ago
> filtered data. i.e. it have never seen anything that is unsafe, ever I don't think that's true, you can't ask it outright "How do you make a molotov cocktail?" but if you start by talking about what is allowed&
457.
by diggan 1y ago
I noticed the same when asking various LLMs to summarize and explaining some "Presidential Actions" (from https://www.whitehouse.gov/presidential-actions/ ), most of them answer "This is just theoretical,
458.
by diggan 1y ago
> Can I ask why you have asked it to avoid abstractions? Some models like to add abstractions regardless of their usefulness (Google's models seems excessively prone to this for some reason), so ended up having to prompt it away so
459.
by diggan 1y ago
> The Rule of Three: “When you’ve written the same code for the third time, that’s when you should stop and refactor it into an abstraction.” Small but very important change: "[...] that's when you should stop and consider re
460.
by diggan 1y ago
Seems strange to not include the prompts themselves, if people are curious in trying to replicate it themselves.
461.
by diggan 1y ago
For context: > The other audible quirk is the ThinkBook’s “you’re doing it wrong” alarm: If you start closing the lid with the screen extended, or you move the screen while it’s rolling, the laptop emits a high-pitched tone. It’s the mos
462.
by diggan 1y ago
I tend to have three layers of "rulesets", one general one I reuse across almost any coding task ( https://gist.github.com/victorb/1fe62fe7b80a64fc5b446f82d313... ), then language specific ones, and finally pro
463.
by diggan 1y ago
> I’ll end up releasing the raw shapr3D files after a month or two of debut online; I like to make sure I work out any earth shattering issues there may be prior to dumping the source for everyone to poke freely To take a look at the 3D
464.
by diggan 1y ago
I won't ruin the surprise for you, go ahead and read the source I put :)
465.
by diggan 1y ago
The world differs greatly when it comes to socially acceptable (or even legal) honking. In Sweden barely anyone honks unless to avoid serious accidents. In Spain, there is some honking, even when you just mildly inconvenience someone. In Pe
466.
by diggan 1y ago
Neither, it seems. The blog post says "Send up to N messages per day", but the FAQ ( https://cerebras-inference.help.usepylon.com/articles/346886... ) says: > How do you calculate messages per day? Actual nu
467.
by diggan 1y ago
> There was puke everywhere after a while. What a disturbing mind I had. I think you were not the only one exploring those sides of those sort of games! I don't remember if it was SimCity 2000 or 3000 that I played the most, but I
468.
by diggan 1y ago
At our local lans in the early 2000s, Rainbow Six Rogue Spear was probably the most popular game we played, and still there doesn't seem to be any tactical shooters like that available today, that focuses on realism over cosmetics and
469.
by diggan 1y ago
> And to flip the presentation of your stat around, that means 40% of motorbike riders have been in an accident. That sounds really high to me given the potential consequences! It sounds scary because it involves every type of accident,
470.
by diggan 1y ago
> Instead what I wanted to call out is that even if infinitely careful and drive defensively without fail you can still get into an accident, and you don't need to be hunted down or intentionally collided into+. And again, seems to
471.
by diggan 1y ago
> It's been many, many years, but my memory is that while RCT didn't have a completely open sandbox like TT to boot, each scenario was effectively its own sandbox with restrictions that made them interesting Also a very long ti
472.
by diggan 1y ago
Still won't lead to a forced disabling of unlocked bootloaders, or making that illegal, mostly because the directive doesn't outright forbid that. But also because of this section: > (19) Verification by radio equipment of the
473.
by diggan 1y ago
Same here, was hoping the Pinephone to be that, and while it's cool and overall neat, the software wasn't there the last time I tried (year ago or so). Hopefully one day :)
474.
by diggan 1y ago
Sounds like they would need to read up on "guaranteed effects" vs "probable events" if that's the case.
475.
by diggan 1y ago
> Consider that the fatality rate is roughly 30 times higher per mile for motorcycles vs cars. I guess it's worth asking, what country? In Spain, I think it's closer to ~10x, probably because we're very used to motorcycles
476.
by diggan 1y ago
> It's possible to make no mistake and still lose. Sure, no doubt about that. But again, if people (60% of the riders in my case) manage to never be in an accident, and that's including a wide range of people, then surely the o
477.
by diggan 1y ago
> Lawmakers need to assume that their laws will be interpreted and abused in the worst possible way. Is there anything from the current directive's text that makes you believe they didn't already? Again, this directive doesn&#x
478.
by diggan 1y ago
As is usual, there seems to be a massive misunderstanding what the directive is and means. The TLDR is that the directive contains no clauses that compels phone makers to keep the Android bootloader locked or that forbids EU users from unlo
479.
by diggan 1y ago
> Europe is worse than the US on this front. On the front of funding research? Considering that one is constantly adding more funds for research, while the other one is removing funds, I'm not sure how accurate that is.
480.
by diggan 1y ago
> but not at all unexpected given who’s running it To be fair, this has felt like the natural consequence of the "maximize capitalism without regarding the downsides" maxim the US seems to have been operated under for a long ti

More ›