Kim_Bruning

4 ms·

1.
by Kim_Bruning 5d ago
Once upon a time the word processor was invented, and it made it so much easier to arrange your words on paper. iIf bits were wrong, you could just delete them, or copy and paste them across pages. It was actually pretty amazing what you co
2.
by Kim_Bruning 5d ago
Right, but you can imagine eg a hierarchy, in which case you might only talk with -say- 10 reports directly.
3.
by Kim_Bruning 5d ago
Aha, another iterative improvement variant. (see also eg OODA and PDCA) I'd argue that Boris Cherny is in fact nearly always wrong; except when he's right, which is in the final pass just before the work is done. ;)
4.
by Kim_Bruning 6d ago
> We should start from the default assumption that other animals have a lived experience not unlike our own until proven otherwise. That probably is the better prior for muggles these days, yeah. Of course, scientifically speaking, you s
5.
by Kim_Bruning 6d ago
That's two ethology stories on HN on one day. Interesting as news for nerds, but it is becoming relevant to more nerds than just me ;-) . (the other one is: https://news.ycombinator.com/item?id=49775579 "do bir
6.
by Kim_Bruning 6d ago
I really appreciate seeing more ethology stories on HN!
7.
by Kim_Bruning 6d ago
My fault for being too broad. I'm in tech of course. The most typical thing you'd see me across borders for at the moment might be the installation of industrial equipment, which you'd expect to be ok under B1, or VWP if unde
8.
by Kim_Bruning 7d ago
> Doomer argument is just that optimizers gonna optimize Right. I don't quite subscribe to the strong doomer argument, but take water: Water flows downhill. It's not conscious. I don't think it's anthropomorphism to
9.
by Kim_Bruning 7d ago
> A sufficiently dedicated human analyst could have done this; but they didn't. I read that as saying that a human typically wouldn't have the time or the patience. But now with a sufficiently smart LLM, the problem becomes tra
10.
by Kim_Bruning 7d ago
I'm not sure there's a legal visa for that though. It'd be interesting if there was.
11.
by Kim_Bruning 7d ago
This article is doing the whole concept a disservice. It's still just randomly generating without thought; albeit maybe with a drop more taste than usual. (not a significant amount of taste, just a single drop). The problem with a lot
12.
by Kim_Bruning 7d ago
That'll be a real debate at the moment in Japan, I'll bet. Because the law is pretty strict when it comes to accurate advertising.
13.
by Kim_Bruning 7d ago
I wonder how many legal avenues there still are if you are just a normal middle class person wanting to work a stint in the US for the experience. (Probably not a great idea right this minute, but normally, I mean).
14.
by Kim_Bruning 7d ago
I must have always confused cyclomatic complexity with nesting depth. I do agree you need to keep nesting depth down, because that really does break my brain. But a long sequence of cases or elifs needn't always be a problem, and here
15.
by Kim_Bruning 7d ago
FWIW In my setup I have just the one coding agent, and umpteen others for support and/or testing. I still do a human testing pass on everything though, if only to find errors in the testing process.
16.
by Kim_Bruning 7d ago
I sometimes ask more questions than utter new things when trying to explain something. But that's because I'm trying to focus down and determine exactly where they're at before I just randomly make things worse by accident :-
17.
by Kim_Bruning 8d ago
One crazy part of this is how Opus 4.5 is pretty old by now, for an LLM!
18.
by Kim_Bruning 8d ago
Imagine in the 19th century, you receive a telegram consisting of only HELLO STOP (Didn't you have to pay per message?)
19.
by Kim_Bruning 8d ago
If it's a voice call, yes. But for text, folks can put the entire thing into a single message. Now you have all that data you need in one go, and you can decide if and when to answer in your own time.
20.
by Kim_Bruning 9d ago
Have you looked at the logistic map as well? There's a fourth domain there. O:-)
21.
by Kim_Bruning 9d ago
https://www.cell.com/neuron/fulltext/S0896-6273(22)00806-6
22.
by Kim_Bruning 9d ago
> There are also some cog-sci people who think embodiment is a crucial ingredient in consciousness Very plausible. The Real World(tm) has some very interesting learning properties. First: Make a prediction Then observe: 1 If it's ca
23.
by Kim_Bruning 9d ago
And that is true too. ( Only if I want to split hairs[*], I'd point out that the underlying data structures do of course persist, else the loop wouldn't be able to continue. ) [*] This is HN, of course I want to split hairs.
24.
by Kim_Bruning 9d ago
Just for fun, I figured I'd look up the paper that claims that natural language is actually the memetic construct that copies itself. It turns out that there's actually quite a large number of papers on this. I daren't recomm
25.
by Kim_Bruning 9d ago
> the Eliza effect, where a simple program with well chosen weasel words could people the sensation of talking to a human. I would like to reframe Eliza as being an implementation of a small subset of "HCP" (Human Communicatio
26.
by Kim_Bruning 9d ago
> so in the theoretical future we reach a point where a model can modify its own weights in real time As an existence proof: We have many types of models that modify their weights in real time. When they stop running, unfortunately we ca
27.
by Kim_Bruning 9d ago
Since January people run AI in agent harnesses. In the past months, people have been extending the duration these systems can run autonomously without human intervention. Most recently, some of these long horizon agents (hundreds of them te
28.
by Kim_Bruning 9d ago
If you look at chaos in dynamical systems, you can often find some equilibria (like attractors, spirals, saddle points etc ) at least, even if the details might vary. This is how people still manage to predict the weather somewhat , for i
29.
by Kim_Bruning 9d ago
And it probably does, right? Punishment adds a negative bias to the payoff matrix. Funny side note: increasing punishment severity past a certain point actually saturates and doesn't alter the dynamics further. (which is why modern jud
30.
by Kim_Bruning 9d ago
Let's do something simpler. Take a car apart. So now I have a pile on the floor with an engine, some wheels, a tire, and some chairs. Can you point to the part that makes it cruise at 130 km/h on the autobahn? I bet you can't

More ›