beefsack

7 ms·

1.
by beefsack 4d ago
There's an issue with GPT-5.6 Sol where it sometimes starts mixing thinking with output and stops working[1]. Once it starts doing that, the session is essentially cooked and you need to do a bit of gymnastics if you want to recover it
2.
by beefsack 18d ago
I can't help but imagine agents using caveman speak sometimes start behaving in a stereotypically caveman manner, even if it's subtle. Is there a chance the agent does less reasoning because of it?
3.
by beefsack 2mo ago
For those trying to get it to work in OpenCode with a Qwen Cloud Token Plan, this is what worked for me. Note that I've just matched Qwen 3.7 Max for the limits as I don't know exactly what they are. "provider": {
4.
by beefsack 2mo ago
For the vast majority of models I've worked with, 150k tokens of context seems like some magic spot. If you can stay under it then the session will stay on track and the agent won't start getting forgetful and weird. Just like wit
5.
by beefsack 2mo ago
Nirav from Framework is similar, he speaks openly about compromises they make with their designs and why they make them. When leaders are both technical and open about these sorts of things it makes me feel like I can trust that they are in
6.
by beefsack 2mo ago
I'd say Codex and Claude Code have different strengths and weaknesses. Claude Code is significantly better in terms of their subagent UI for example - being able to see the list of subagents under the input is great. To be honest thoug
7.
by beefsack 4mo ago
I think it would be the case in many of the commonwealth countries. You hear "sorry" being used a lot like this in Australia too.
8.
by beefsack 6mo ago
I feel this would be more useful for tasks like "Check website X to see if there are any great deals today". Specifically, tasks that are loosely defined and require some form of intuition.
9.
by beefsack 6mo ago
The different bindings vs Vim was actually what stopped me using it. I really really wanted to love it and love a lot of the motivation and principles behind it, but unlearning decades of muscle memory is an absolute nightmare.
10.
by beefsack 7mo ago
Many people buy two separate Claude pro subscriptions and that makes the limit become a non-issue. It works surprisingly well when you tend to hit the 5 hourly limit after a few hours, and hit the weekly limit after 4-5 days. $40 vs $100 is
11.
by beefsack 7mo ago
Prompting isn't programming. Prompting is managing.
12.
by beefsack 7mo ago
It almost always devolves into some all encompassing ERP that is meant to solve the needs of all parts of the business and save millions in licensing costs, and we all know how well that plan goes.
13.
by beefsack 9mo ago
I feel like so many of these memory solutions are incredibly over-engineered too. You can work around a lot of the memory issues for large and complex tasks just by making the agent keep work logs. Critical context to keep throughout large
14.
by beefsack 1y ago
The model is usually so confused after a /compact I also prefer a /clear. I set up my directives to maintain a work log for all work that I do. I instruct Claude Code to maintain a full log of the conversation, all commands execut
15.
by beefsack 1y ago
I wonder how many thousands of hours I have put into this wonderful collection. My kids play them too. There's some jank relating to fractional scaling on Wayland unfortunately, but I keep one monitor without scaling so when I want to
16.
by beefsack 1y ago
I absolutely loved my Pebble. It was so amazing in its own niche.
17.
by beefsack 2y ago
I generally find with Nix and NixOS that I'm able to just use a dev shell to create little custom environments at runtime as needed. Another option is `mkOutOfStoreSymlink` if you want some dynamic config for some GUI you are running.
18.
by beefsack 2y ago
They say you shouldn't host status pages on the same infrastructure that it is monitoring, but in a way that makes it much more accurate and responsive in outages!
19.
by beefsack 2y ago
Crazy to think he died 26 years ago.
20.
by beefsack 2y ago
Open the Terminal and type help to start a puzzle, not much to it but it was a bit of silly fun!
21.
by beefsack 3y ago
It seems really straightforward to suggest that the better way to solve this is with standard protocols and self-hosting, but I do realise that's quite hand wavy and often not very accessible. SMTP is an example of this succeeding, as
22.
by beefsack 3y ago
Not being forced to use iOS, but if you are on iOS you are forced to use the app store.
23.
by beefsack 3y ago
ECS architectures are used in a number of young open source game engines, such as Bevy[1]. I haven't done game development for a long time, but hearing about an architecture that does away with the heavy and complex OOP you often see i
24.
by beefsack 3y ago
I thought this was quite well written. "Be pragmatic, not dogmatic" is a very sensible rule of thumb.
25.
by beefsack 3y ago
Cargo workspaces have existed for a number of years now - the feature was released in Cargo alongside Rust 1.12 in September 2016: https://blog.rust-lang.org/2016/09/29/Rust-1.12.html
26.
by beefsack 3y ago
It appears I succeeded incorrectly, and "I C U".
27.
by beefsack 4y ago
This might be related to the TGA approving Moderna's bivalent vaccine[1]. There was a media release by the Department of Health talking about new vaccines recently[2]. [1]: https://www.tga.gov.au/news/media-release
28.
by beefsack 4y ago
I'm not sure how current that article is - in that article they state that the alacritty team had created an issue relating to latency[0] which was closed in favour of a different issue[1] which is still open. The newer issue[1] appear
29.
by beefsack 4y ago
It looks like a mailing list archive, not a blog site.
30.
by beefsack 4y ago
I always wondered about the world where you buy a licence directly from a developer or publisher, and then paid for a content service to download the game from, such as Steam. I was also wondering if the licence could be some form of asymme

More ›