osigurdson

15 ms·

91.
by osigurdson 5mo ago
I actually like AI mode in Google. My main reason is if I just have a quick question it seems a lot quicker than logging into ChatGPT/Claude as I can just type it in the address bar. Of course DDG / others can do the exact same th
92.
by osigurdson 5mo ago
It requires a nascent analogous architecture to biology (neural nets) and a discovery (next token prediction) to exactly match evolution. I think that is unlikely. Instead it will be a tool that is capable of doing certain things that machi
93.
by osigurdson 5mo ago
The title is a little misleading as it sounds like it might hint at the end state of next token prediction - i.e. will we reach AGI with it and so on. Instead, this is entirely about "means of production" and "class war"
94.
by osigurdson 5mo ago
I thought the interior looked pretty nice - lots of retro physical switches, etc. The exterior doesn't look like a normal Ferrari but maybe that's on purpose. A "normal" Ferrari buyer would probably buy a normal Ferrari.
95.
by osigurdson 5mo ago
I find that deeply understanding code is nearly the same level of effort as writing it. That's what I don't really get about AI productivity claims as there is is really only one way to get the coveted 10-100X boost: stop looking
96.
by osigurdson 5mo ago
Maybe smart to convince everyone else to fire and slow down while you hire and speed up. I don't think it is actually planned, but perhaps smart in hindsite.
97.
by osigurdson 5mo ago
It goes faster, at least for a while, if you don't look at the code.
98.
by osigurdson 5mo ago
I've definitely tried that with claude code. It seems AGI will be needed to solve this one.
99.
by osigurdson 5mo ago
"Data driven" essentially means "scalar driven". There is nothing wrong with it if your chosen scalar is a proxy for anything that matters. Of course, usually no one can explain this mapping.
100.
by osigurdson 5mo ago
This is a ThinkPad USB-C docking station. I think it is basically the same as what you are talking about but if not, the main convenience is that you can plug in one cable which provides power, 2 external displays and a mouse. Super great i
101.
by osigurdson 5mo ago
It seems like a natural result. People have been trying to use dashboards / metrics to roll up / indicate how well teams and individuals have been doing for a long time. Therefore, "part 1" was already in place. Now, som
102.
by osigurdson 5mo ago
I have some issues with docking and audio with Arch. Maybe I am asking too much to have a docking station work perhaps. I still don't feel the need to go back to Windows however.
103.
by osigurdson 5mo ago
I wouldn't say I get worse results with Go than I do with Python.
104.
by osigurdson 5mo ago
It won't be a career if AI gets good enough that you don't have to read / understand the code - otherwise, AI won't have much impact on jobs I don't think.
105.
by osigurdson 5mo ago
It seems the market decided to cut 20% of Cloudflare's market cap today.
106.
by osigurdson 5mo ago
How does low quality translate to shareholder returns?
107.
by osigurdson 5mo ago
I suspect it isn't even really "greed". It is just the slow mold growth of an org chart optimizing comfort for itself instead of value for customers. Generally, startups / founders are the only anti-bodies against this t
108.
by osigurdson 6mo ago
It is fine until cognitive debt reaches a point where you have to essentially have re-write just so you understand it. Very good for speed running through a problem space.
109.
by osigurdson 6mo ago
Agree. There is so much focus on "let's do the same thing we are doing now with fewer people". It is very boring and uninspired. How about "let's do something that we couldn't do before", instead?
110.
by osigurdson 6mo ago
The author calls it out at the end but still spends a long time creating a false equivalency. Offshoring to humans in another country is not the same as having machines do work domestically. Really, the only thing that matters is: "May
111.
by osigurdson 6mo ago
I always thought they should teach calculus first.
112.
by osigurdson 6mo ago
The barrier to entry seems pretty low, technically at least. Maybe someone will create something with a different twist that will catch on. TikTok was able to carve out a niche after all.
113.
by osigurdson 6mo ago
It's about 300MiB and maybe 5% of a core if the cloud provider offers a free managed control plan. If you want / like Kubernetes it isn't a deal breaker.
114.
by osigurdson 6mo ago
Agree, "What are you working on" is getting diluted. However, I've concluded that this one (posted by david927) is the de-facto "real one".
115.
by osigurdson 6mo ago
Lots of debate about SQLite vs Postgres. One thing to note is you can certainly run Postgres (Kubernetes even if you want) and your app for $20 / month.
116.
by osigurdson 6mo ago
Can you be more specific about these? For instance can you map "analyzes our current marketing efforts" to precisely what is being fed into the context window. What prompt is used to "find something useful"? Like many he
117.
by osigurdson 6mo ago
I'm confused about it as well. I've installed OC locally and also on a VM. I don't get it so far. But then again, I'm not willing to give it all of my passwords which is probably why I'm not seeing much value. It is
118.
by osigurdson 6mo ago
Agree. It is like just having one continuous chat session with ChapGPT forever. Of course they do have memory already (at least ChatGPT does). I ended up turning it off however, because it kept bringing irrelevant stuff into the current con
119.
by osigurdson 6mo ago
Just curious, what is the delta with this and Claude Code / Codex CLI for the SWE work?
120.
by osigurdson 6mo ago
It sounds like you know about this / have researched it. Are you saying that any go function, even func add(x,y int) { return x + y}, is going to have such overhead in all situations? Why wouldn't Go just inline this for instance

More ›