Kogasa

Stokes’ theorem. Almost the same thing as the high school one. It generalizes the fundamental theorem of calculus to arbitrary smooth manifolds. In the case that M is the interval [a, x] and ω is the differential 1-form f(t)dt on M, one has dω = f’(t)dt and ∂M is the oriented tuple {+x, -a}. Integrating f(t)dt over a finite set of oriented points is the same as evaluating at each point and summing, with negatively-oriented points getting a negative sign. Then Stokes’ theorem as written says that f(x) - f(a) = integral from a to x of f’(t) dt.

Just wait til you find the 1-1/2" screws

We aren’t trying to establish that neurons are conscious. The thought experiment presupposes that there is a consciousness, something capable of understanding, in the room. But there is no understanding because of the circumstances of the room. This demonstrates that the appearance of understanding cannot confirm the presence of understanding. The thought experiment can’t be formulated without a prior concept of what it means for a human consciousness to understand something, so I’m not sure it makes sense to say a human mind “is a Chinese room.” Anyway, the fact that a human mind can understand anything is established by completely different lines of thought.

This fails to engage with the thought experiment. The question isn’t if “the room is fluent in Chinese.” It is whether the machine learning model is actually comparable to the person in the room, executing program instructions to turn input into output without ever understanding anything about the input or output.

Reminds me of 2048 making a slightly worse clone of Threes and then releasing it for free.

Before I blocked the instance I had nothing but miserable interactions with Hexbear users, and it had nothing to do with political opinions.

I’m not suggesting that. It just looks like he did in fact choose to have that hair color.

It looks dyed

That’s not necessarily wrong, but not the big explaining factor here I think. The technological challenges behind aligning ML models with factual reality aren’t solved, so it’s not an engineering decision. It’s more that AI is remarkably easy to market as being more capable than it is

Monkey laundering.

Docker is lighter and easier to manage than a VM. I run a collection of services as docker compose services inside a NixOS host VM. It’s easy to start, stop, monitor, update etc. even from a different computer (via ssh or docker contexts). It’s great.

deleted by creator

It’s just as easy to run in a Docker container and I would recommend this anyway.

What platforms would you like your app to run on? Then, which UI framework supporting those platforms would you like to use? Then, look at the framework’s documentation to find a sample starter project that you can run as an app, and modify it from there

I do not know what that means

If only we were still having the conversation.

That’s… not the point either. The point is that “reporting false positives isn’t a bad thing” is only true up to a point. The discussion is then “is this before or after that point.” Which, given the context of the bug, isn’t really a given. But I don’t want to have that discussion with you anymore because you’re annoying.

“What if the boy who cried wolf got lucky and didn’t get eaten in the end”? Seems to have missed the point of the parable a bit.

I didn’t say the CVE was valid. I explained why it was a mistake. I didn’t say “disclosing security bugs” is, in general, a bad thing, I said raising undue alarm about a specific class of bugs is bad. It’s not a matter of “less or more information,” because as I said, a CVE is not a bug report. It is not simply “acknowledgment of information.” If you think my argument has no merit and there is no reason why “more information” could be worse, you’re free to talk to someone who gives a shit.

C# tells you the call site/method name and line number right at the top. It’s only really annoying when you have aggregate exceptions, which sometimes occur because someone async’d wrong