horsawlarway

5 ms·

31.
by horsawlarway 2mo ago
Yeah, I think engineering orgs mostly jumped the shark. Metrics like pr count and commits have always been terrible gauges for success compared to business performance. But they're easy to measure, and even easier to game now with AI.
32.
by horsawlarway 2mo ago
I think this is one of the more important points in this thread. The problem is that too many pieces require experiences that the majority of the classroom just doesn't have. It's not that students don't venture far enough
33.
by horsawlarway 2mo ago
And none of these are meaningfully different on high end laptops from other providers. Apple had a window with M1 devices where I'd probably agree they were just better. Apple also has a history of making devices that are very expensi
34.
by horsawlarway 2mo ago
Apple makes nice hardware, but it's definitely not superior for everything, and the gap between other manufacturers isn't really there anymore on the upper end. So sure - the Macbook is better than a cheap ~$500 laptop. But betwe
35.
by horsawlarway 2mo ago
I've read most of this. I find it be incredibly self-contradictory. I don't even think this aligns much with the actions of the company. > Second, there is a good chance free information really does undermine authoritarianism,
36.
by horsawlarway 2mo ago
Yeah... Anthropic gives off big "I'm sorry, Dave. I'm afraid I can't do that" vibes.
37.
by horsawlarway 2mo ago
Approaching this from the perspective of a potential customer and not a designer - I find that I actually like the smaller/open model output quite a bit more. Kimi, GLM, and Deepseek all absolutely run away with the "Can I quickly
38.
by horsawlarway 2mo ago
Most of these problems are actually solved with containers, though. An opinionated release of k8s (k3s, microk8s, etc) does wonders for making much of it just disappear. You pay a slightly higher up front cost in complexity, in return for l
39.
by horsawlarway 2mo ago
Nah he's entirely right. Serve http, let the user do https. If you want to put caddy in front of yours? Great it works. For me, running a much larger setup? Great, it also works. For users who never expose it beyond an IP address on la
40.
by horsawlarway 2mo ago
The above post fails to distinguish between unmaintained and native. Native is great. I'm with you entirely. Failure to maintain the space is not. That causes expensive and annoying problems.
41.
by horsawlarway 2mo ago
To face this directly: I think an honest answer is probably "yes". Our buying habits empowered the type of company that wants to put AI into everything. Whether they continue to do so or not will also depend on our buying habits g
42.
by horsawlarway 2mo ago
I mean, I get it - I somewhat agree with your take, but this is somewhat of a "damned if you do, damned if you don't" case. I think you undermine your argument when you don't understand the complaint. Because I'd a
43.
by horsawlarway 2mo ago
I think it was one of those cases where "the road to hell was paved with good intentions". I'm old enough and was very pro-android in the early days when it first released. I absolutely remember the frustrations around havin
44.
by horsawlarway 2mo ago
I'll take this bait. First up - I completely agree that parts of the terminal experience are decidedly lackluster, you can make them "fine" but the general experience doesn't align with consistent OS principles users are
45.
by horsawlarway 2mo ago
The internet is already a class based internet. Why do you think SO many services want you to login, or force you to their mobile apps? They are curating your experience to be as extractive as possible, and if you're not worth much m
46.
by horsawlarway 2mo ago
Oh give it a rest. You want to talk about killing a system... look at all the perverse incentives that pop up when companies are allowed to sell user attention instead of decent services and products. If the modern social media internet die
47.
by horsawlarway 3mo ago
Worth buying? Maybe. Actually bought and distributed? No. Again, hardware just doesn't roll out that fast, and you're talking about trying to distribute expensive hardware to a population with a 47k median yearly wage. I've
48.
by horsawlarway 3mo ago
Even if we agree with this take (and I do think it's a likely take that you're right on the long term), it doesn't change that it seems likely we're in a bubble, and it probably will pop. We see a similar paradigm with l
49.
by horsawlarway 3mo ago
If you read the full post, I'm not actually sure I agree with the title. Personally - if I were judging... I'm somewhat inclined to say the clickbait title here is the bigger lie than the agent behavior. To recap: 1. It didn'
50.
by horsawlarway 3mo ago
Since you asked... Tools. I buy more tools than I should - Wood working, electronic, and more general purpose. It's not always cost effective, but I generally pick a new project (ex - making my dining room table, installing solar pane
51.
by horsawlarway 3mo ago
I'm not the above poster, but my experiences generally line up with theirs. The first 20% of context is usually very solid. Agent stays on task, is responsive and focused on the original goal. The last 20% of context is usually pretty
52.
by horsawlarway 3mo ago
I find this is mostly just a "mindset" difference. OSS software aligns mentally with "tool". It only exists because someone had a problem to solve, and they wanted a tool to solve it. Generally, people should expect to
53.
by horsawlarway 3mo ago
I don't understand this mindset. I get that people who are going to upgrade to the latest flagship phone every 18-24 months will probably do better with this, but realistically... those people are fucking insanely financially irrespon
54.
by horsawlarway 3mo ago
This would be my read. Hell - my children just this year would have counted for 10+ doses (my daughter 6 all by herself for a series of severe ear infections).
55.
by horsawlarway 3mo ago
> That pales in comparison to how many people unaligned AI will hurt. Under what argument? In which scenarios? Basically - bullshit. I'm calling bullshit on this argument. It's easy to hurt people already. The "difficul
56.
by horsawlarway 3mo ago
It takes an astoundingly small amount of effort to buy an automatic weapon in the US and go hurt people. Or to buy materials to make an explosive device and hurt people. Frankly, even with AI those are both comically easier than the idea th
57.
by horsawlarway 3mo ago
I'm vaguely sympathetic to this argument. But only vaguely. Google uses its monopoly position in advertising to basically ensure that you allow them to scrape your site (or if not you personally, the majority of revenue driving sites).
58.
by horsawlarway 3mo ago
I don't think the goal of this data is necessarily model improvement. I think it's marketing, advertising, and product refinement. Ex: all the things Google wants your search data for. It's somewhat silly to think the value o
59.
by horsawlarway 3mo ago
Two things. 1. Deps are an organizational hassle. They need review, get flagged as security and legal risks, and (mostly incorrectly imo) have been somewhat tarnished by the idea that deps are all just "leftpad" - simple things th
60.
by horsawlarway 3mo ago
Personally - I think this cuts both ways. Things sold as "unlimited" are often much more expensive for the average user than proper rate based billing would be (see - mobile data, internet data in general, those soft drinks [incl

More ›