dkersten

10 ms·

151.
by dkersten 1y ago
This is similar to why I prefer LLM's to behave less human-like and more robotic and machine-like, because they're not humans or human-like, they are robotic and machine-like. The chatbot is not my friend and it can't be my
152.
by dkersten 1y ago
I hate the new website. I know the product well enough so know what content is on their site and yet I struggle to navigate the new design. It also really doesn’t make me want to. It also makes me scared for the product itself, it doesn’t f
153.
by dkersten 1y ago
Given how much some companies spend on their cloud services bills without batting an eye, I definitely believe this. They care about making more money, not so much about spending less, even though both are ways to increase profits.
154.
by dkersten 1y ago
Titanfall 2 had excellent movement! The only games I’ve played that surpass it are the tribes games, which just have a whole new level of fluid movement for 3d games.
155.
by dkersten 1y ago
I agree. I used to use vscode, then switched to Zed and used it for over a year (without AI). In February of this year, I started using Cursor to try out the AI features and I realised I really hated vscode now. Once Zed shipped agent mode,
156.
by dkersten 1y ago
Off topic, but if you really want to use an alternative to qwerty, try colemak — it keeps many of the most common shortcuts the same (ie QAZXCV all remain in the place) which makes it a little easier to transition. (It also has a few other
157.
by dkersten 1y ago
My pet peace is when I point out a problem it responds with acknowledgement and then explaining why it’s wrong. Like… I already know why it’s wrong, since I’m the one that pointed it out!
158.
by dkersten 1y ago
What I really want from Zed is multi window support. Currently, I can’t pop out the agent panel or any other panels to use them on another monitor. Local-first is nice, but I do use the AI tools, so I’m unlikely to use this fork in the near
159.
by dkersten 1y ago
Yes. However people are unwilling to take this approach unless things get really really bad. Even then, the powerful tend to have such strong control that people are afraid to act out of fear of reprisal. We’ve also been gaslit into believ
160.
by dkersten 1y ago
Here is how GPT self-described LLM reasoning when I asked about it: - LLMs don’t “reason” in the symbolic, step‑by‑step sense that humans or logic engines do. They don’t manipulate abstract symbols with guaranteed consistency. - W
161.
by dkersten 1y ago
Because the people with the money aren’t going to just give it to everyone else. We already see the richest people hoard their money and still be unsatisfied with how much they have. We already see productivity gains not transfer any benefi
162.
by dkersten 1y ago
Humans can break things down and work through them step by step. The LLMs one-shot pattern match. Even the reasoning models have been shown to do just that. Anthropic even showed that the reasoning models tended to work backwards: one shott
163.
by dkersten 1y ago
You can trivially demonstrate that its just a very complex and fancy pattern matcher: "if prompt looks something like this , then response looks something like that ". You can demonstrate this by eg asking it mathematical questi
164.
by dkersten 1y ago
Max/MSP actually kinda encodes this into the language too: when all else is equal (eg if you connect the same data output to the input of two print nodes), nodes on the left happen before nodes on the right (normal input/output co
165.
by dkersten 1y ago
I agree. I feel like the more strictly defined the problem space can be and the more guarded rails that can be added and checked, the better the LLM will be. The data derived from such constructs also doubles as providing additional context
166.
by dkersten 1y ago
Agree! I’m not an AI engineer or researcher, but it always struck me as odd that we would serialise the 100B or whatever parameters of latent space down to maximum 1M tokens and back for every step.
167.
by dkersten 1y ago
I was responding to this bit: > It assumes cookies. It assumes a server.
168.
by dkersten 1y ago
Correct. But you can also use cookies for that, without violating GDPR or the ePrivacy directive.
169.
by dkersten 1y ago
> which is where GDPR fails. It assumes cookies. It does not assume anything. GDPR is technology agnostic. GDPR only talks about consent for data being processed, where 'processing' is defined as: ‘processing’ means any o
170.
by dkersten 1y ago
The law is very clear, if you actually read it. It doesn't care what technology you use: cookies, localstorage, machine fingerprints, something else. It doesn't care. It cares about collecting, storing, tracking, and sharing user
171.
by dkersten 1y ago
I tried a bunch of todo apps, task trackers etc and also tried a txt file. None of it really worked for me. I tried bullet journals, I couldn’t stick to it. What did end up working for me is a clipboard with a sheet of paper that I replace
172.
by dkersten 1y ago
Only if it’s economically feasible. If it takes a city sized data center and five countries worth of energy, then… probably not going to happen. There are too many unknowns to make any assertions about what will or won’t happen.
173.
by dkersten 1y ago
My point is that “AGI has human intelligence” isn’t by itself enough of the equation to know whether there will be exponential or even greater-than-human speed of increase. There’s far more that factors in, including how quickly it can proc
174.
by dkersten 1y ago
If AGI has human capability, why would we think it could research any faster than a human? Sure, you can scale it, but if an LLM takes, say, $1 million a year to run an AGI instance, but it costs only $500k for one human researcher, then it
175.
by dkersten 1y ago
No .vimrc importing, but remapping capability is quite comprehensive.
176.
by dkersten 1y ago
I don’t say I don’t know how to use one. I was a longtime vim and then neovim user. But I don’t want to use one. I still use vim for quick terminal edits, git commit messages, and doing stuff over ssh. But for my day to day heavy text editi
177.
by dkersten 1y ago
Not everyone wants to use a terminal editor and of those people, not everyone wants to use vscode.
178.
by dkersten 1y ago
When did you last try Zed? If it was recently then that hasn’t quite matched my experience, for me vim keybindings work quite well. But I did think the same when I first used Zed over a year ago.
179.
by dkersten 1y ago
I went slightly differently: I used Zed for about a year and a half exclusively, without using any AI features, and then switched to cursor to try AI features out. When Zed released its agent mode, I switched back to Zed. I absolutely agree
180.
by dkersten 1y ago
Jsut today I was doing some vibe coding ish experiments where I had a todo list and getting the AI tools to work through the list. Claude decided to do an item that was already checked off, which was something like “write database queries f

More ›