dminik

13 ms·

181.
by dminik 6mo ago
I don't think you can. The comments section of the page is also behind the block for you, no?
182.
by dminik 6mo ago
I'm not opposed to even removing the comment outright. That being said, GitHub does not even offer a time sorted search. Meaning that most of the results are going to be quite old and useless. Second, API keys being shared on GitHub is
183.
by dminik 6mo ago
Try this one. Should remove most readme keys: Edit: self censor based on a request
184.
by dminik 6mo ago
Yeah, the Sheets integration is weird. It's usually ok when it wants to place something down the first time. But then it seems incapable of making any changes to it. Or even acknowledging the data in the sheet. What's up with that
185.
by dminik 6mo ago
You can have multiple models served now with loading/unloading with just the server binary. https://github.com/ggml-org/llama.cpp/blob/master/tools/serv...
186.
by dminik 6mo ago
I don't see how that is relevant? If he really did steal that money, it's not his to give. You can't take someone's money and then not only not give it back, but also give it away.
187.
by dminik 6mo ago
It's a really weird situation, but using public transport WiFi cured me of this thinking. The amount of times that the initial HTML, CSS and JS came through, but then choked on fetching the page content was insane. Staring at a spinner
188.
by dminik 6mo ago
Maybe this is just a skill issue, but even with several attempts I just can't figure out why I would use stacked diffs/PRs. Though maybe that's because of the way I work? I notice a lot of examples just vaguely mention "
189.
by dminik 6mo ago
This would be true if the models were capable of always completing the tasks. But, since their failure rate is fairly high, going in a wrong direction for longer could mean that you take more time than a faster model, where you can spot it
190.
by dminik 6mo ago
I needed to create a Facebook account to verify some Facebook posting functionality worked as we wanted it to. Ok, easy. Tried using my work email. No, not allowed. Hmm, try creating a new email. Nope. Tried to create an account on my perso
191.
by dminik 6mo ago
This is a pretty interesting topic. For GO, switching to Linux (with an AMD card) was a free performance boost. I gained like 30fps. For early CS2, the performance on Linux was terrible. Now, the peak fps is slightly worse, but the frame pa
192.
by dminik 6mo ago
I imagine the new pay per use pricing for the X API has something to do with it. If you're reaching single digit percentage impressions and now you have to pay for that as well ...
193.
by dminik 6mo ago
> We do not want to limit to only detected Vecel project, because we also want to help with greenfield projects "Help build me an AI chat app". Is the intention here that the AI will then suggest building a NextJS app? I can&#x
194.
by dminik 6mo ago
An idea can also reduce value. Or prevent you from producing value in the future. Knowing when an idea is bad or not worth doing is a skill in itself.
195.
by dminik 6mo ago
Some feedback: - Nice idea, though after playing Turing complete, I would like to skip the beginning and move to stuff that makes GPUs different to CPUs. But it's understandable. - I'm not smart enough to intuit NAND from transist
196.
by dminik 6mo ago
It depends on the hardware, backend and options. I've recently tried running some local AIs (Qwen3.5 9B for the numbers here) on an older AMD 8GB VRAM GPU (so vulkan) and found that: llama.cpp is about 10% faster than LM studio with th
197.
by dminik 6mo ago
Sometimes the impression I get from commenters on HN is that they would sell their own grandmother for money. Much less than just not considering morals/ethics, it's seen as a weakness here.
198.
by dminik 6mo ago
Well, I wasn't that worried for the astronauts before, but now that I know they're running windows, I'm not so sure.
199.
by dminik 6mo ago
In what way? Does it matter that there's an Adobe logo on the top if I can just point an LLM at some leaked code and get PictureStore?
200.
by dminik 7mo ago
I can't believe as developers we were worried about AI training on licensed code. It turns out it didn't matter at all. You can just point an LLM at some source code and you're off scot-free.
201.
by dminik 7mo ago
There are a lot of comments here saying something to the effect of "spaceflight is inherently unsafe" or "you can't always guarantee safety." I find this rather concerning. Surely there is a difference between &quot
202.
by dminik 7mo ago
Have you bothered to ask the gambler if they want to risk it? No offense to the astronauts of course, but asking people that have dreamed of this opportunity their whole life doesn't actually tell you all that much about the actual saf
203.
by dminik 7mo ago
The giant advantage regular games have is that I've yet to smash my hand into a wall playing them. I think that the relatively low living space area for most of the world is a huge strain on VR adoption.
204.
by dminik 7mo ago
If you're not sure what something is saying, how can you be sure that the AI had picked the correct interpretation?
205.
by dminik 7mo ago
> Zuck and tech as a whole pivoted to drop safety investments the moment this government came to power. I think the question to ask here is, if both Meta and the current administration don't care about child safety, why is the age v
206.
by dminik 7mo ago
Why are you asking lawmaker questions of people on HN? What kind of answer are you expecting? Just because I don't know how to write a law that can prevent it doesn't mean that I can't recognize an actual issue when I see it.
207.
by dminik 7mo ago
> The proposal to move age verification to the OS level would give Meta less information about the user, because the OS, not Meta apps, would be responsible for gating age content. I find it hard to believe that meta doesn't already
208.
by dminik 7mo ago
Right. I went through and re-did the sheet myself and yeah, it's not great. Incidentally I got to $4.5B myself.
209.
by dminik 7mo ago
I've had to reread the first tweet a bunch, but I don't think Ed is wrong there. As far as I can tell, in February Anthropic projected their 2026+ annual revenue at $14 billion dollars, based on a month long period. If you added t
210.
by dminik 7mo ago
It's much simpler. Reactive UIs and declarative UIs let you built things more more easily. You can almost entirely ignore an entire class of issues (state mismatch for instance). That of course led to developers building much more comp

More ›