switchbak

4 ms·

1.
by switchbak 7d ago
Yes, I remember those too. The costs of manual memory management were real and were not low. But costs on the cloud are real too, especially now. I’ve been living in JVM land for a very long time, but now it’s especially clear how important
2.
by switchbak 8d ago
This is BONKERS. I spend a lot of time encouraging my manager and coworkers to ... write some actual english. To not just point an AI at a PR and call it a day. To actually collaborate before we jump into something - leaning on the knowledg
3.
by switchbak 10d ago
You know - I would expect that could very well happen. Assuming the first hump of the hype cycle doesn't completely burn people off from ever realizing the second.
4.
by switchbak 10d ago
I really have no idea what you're on about. There's TONS of examples of people abusing this already. Will someone make an actual product to help guide a company? For sure. Will it be abused immediately, of course! Is that cynical?
5.
by switchbak 10d ago
100% Our hiring pipeline was 100% full of AI cheaters recently - like 6 in a row that we interviewed (broken process too). Not exactly the same, but close enough and just shows how many people would be willing to waste a bit of time in an A
6.
by switchbak 10d ago
Depends- i care about the quality of power tool I buy. There’s plenty of competition, but most of it I won’t touch because it’s some “Zissbou” brand off of Amazon. The competition on the high end is great! Makita, Milwaukee, and others batt
7.
by switchbak 10d ago
True … I’m the same, but I consider that good engineering. Designing it with the use in mind, factoring in all the stresses adequately but only just. That’s more like how an airplane is made: weight/strength only where it’s truly neces
8.
by switchbak 12d ago
I was at ground zero at Nortel during that time. Quite an interesting (depressing?) place to be at the very start of your career!
9.
by switchbak 12d ago
When every get-rich-quick gonzo has exactly ZERO barrier to spamming the world with their terrible ideas, we're in for a particularly bad time. Ie: this f**in guy: https://www.youtube.com/watch?v=FRGLToHAtgc
10.
by switchbak 12d ago
10,610 commits since June. How much are you spending on AI?
11.
by switchbak 12d ago
Part of the issue is that the worst kind of people are already head over heels excited about this. We're already seeing the Crypto->Web3->LLM get rich quick folks take to this like wildfire. And much like wildfire, they'll r
12.
by switchbak 12d ago
"By late 2025, frontier models had gotten good enough that running a real-life vending machine was no longer a challenge" ... is this satire? Isn't it kind of hard to lose money running a vending machine in a good location?
13.
by switchbak 12d ago
Yeah you'd think this is something Google's stuff would detect and strongly punish, but somehow we ended up in this dystopian crapified parallel universe instead.
14.
by switchbak 12d ago
This is like the high waisted, mom-pants of browser UI design.
15.
by switchbak 12d ago
If you live in a nation, you do have some say in the policies that directly affect your livelihood. This is the whole reason we have nations. One need not (always) invoke racism or xenophobia to explain why folks might take issue with high
16.
by switchbak 12d ago
Or you know, you could take issue with specific concerns of specific implementations. Driven mostly by corporate greed and cynicism, over a great many decades. Also in an industry that you’re in and know something about. You’re still allowe
17.
by switchbak 12d ago
That takes me back. I made an unreleased Android dual-N-back game in the late 2000’s, in Scala no less. I got a little bummed when the big claims of the N back training failed to be replicated though.
18.
by switchbak 13d ago
I had exactly the same thoughts. I often have similar thoughts on other benchmark sites, where supposed performance is way off base from my experience. I’m not sure what the methodology of these are, but they certainly don’t match what I ex
19.
by switchbak 13d ago
I’m shocked that’s still possible, shouldn’t we be able to fix that at the browser level?
20.
by switchbak 13d ago
Yes, I use the messing Chinese models extensively. Like you, I very much appreciate them and sometimes prefer them. But there is a bar of complexity at which they fail, and repeated invocations typically doesn’t make much progress. This is
21.
by switchbak 13d ago
I dunno, I felt less human when I had to take days to do large scale menial refactors. I felt more like a minion or borg member back then. Now I feel like I’m middle management aboard a star destroyer. Lateral move?
22.
by switchbak 14d ago
Yeah, you and me both. Look at it on the long scale of history, change is a constant. Everything is impermanent. It’s not our choice to have this forced upon us, but it’s our choice as to how we adapt. Or not. Personally, I’m choosing to be
23.
by switchbak 14d ago
Sonofa, it’s like those corporations are the KGB or something. It’s getting really hard to keep up on all their shady shit.
24.
by switchbak 14d ago
I didn’t say async/await need language features (though they usually do), I meant that high level async orchestration can be represented with suitable language features - often in ways that are more interesting than async/await. “
25.
by switchbak 14d ago
I’m trying to hear what you’re saying, but using these Marxist cliches makes your statements a little off putting to me.
26.
by switchbak 14d ago
That’s only one metric. It doesn’t matter how cheap it is, if it continually fails to solve a problem. Yes they’re getting better. No, I don’t think this will be a panacea. If only for the fact that this is China (more specifically the CCP)
27.
by switchbak 14d ago
Why do people bother with this llm stuff? Is it to mine the karma system? It’s so obvious and clearly dumb, I don’t understand the point otherwise.
28.
by switchbak 14d ago
These higher level primitives that they mention provide the primitives for exactly that. This can be found in other paradigms besides async/await. I do find that the ergonomics of this are highly dependent on a few features of a langua
29.
by switchbak 17d ago
I presume you're trying to be friendly here, but your first couple sentences come off a bit harsh. I like that he's well into his life and still enjoying writing (or authoring, or whatever we call it now) code, and looks forward t
30.
by switchbak 17d ago
You've added a similar response like 5 times, but I don't see you adding more information. Yes - model hosts can do nasty things to you aside from changing the underlying model. That doesn't mean it's cool to have them c

More ›