dceddia

The plastic parts almost definitely, but this seems like it has a couple springs and hinges that make it more of a pain.

dceddia

先月議論

I’m just gonna take this chance to drop a rec for Joe Hudson’s work for anyine interested in learning more about emotions and how to work with them in yourself and others. It’s been super impactful for me. (YouTube/podcast)

On this point in particular (that we’re broadcasting these signals), in one of the AoA courses I took there’s an exercise where you and a partner take turns guessing at the emotion the other person is feeling. You’re just staring into each others’ eyes on zoom, and one person is going “angry. sad. joyful. fear. Angry again.”

When I was the guesser I had no idea if I was even in the ballpark.

But when the other person was guessing my emotions, they were probably 70% accurate with what I was feeling. We discussed afterward and they had the same experience (I had apparently gotten most of them right). It was wild.

dceddia

2 か月前議論

Handy is great. It exposes a bunch of open models beyond Whipser too, and though I haven’t tried too many of them, I’ll throw in a rec for the Parakeet model which feels pretty much on par with Whisper for accuracy and is way way faster.

dceddia

2 か月前議論

As a solo dev, I’ve found myself spinning up little servers for various things and then just letting them run for months between needing to make changes.

At first (and for admittedly way too long), I used this as a way to try out fun new frameworks - Node+Express for one thing, Phoenix for another, SvelteKit for a third.

I noticed it was a huge pain to dive into these things once every 6 months. I’d forgotten how it worked, and for some of them at least, I could look up docs and examples.

My Node+Express thing was the worst because it was all homegrown. There’s very little convention in that world, and you have to make your own. No docs were coming to save me, and this was in the Before Times, like 3 years ago pre-LLM.

Anyway I ported everything to Rails and it’s wonderful. I know how it works, there’s almost 30 years of examples online and they even mostly still work, and LLMs are great at it too.

Lots of power in a good framework, in a situation that’s a good fit for it!

dceddia

3 か月前議論

I resonate with a bunch of this. The idea that what I make is somehow not as good as the “real” version of whatever it is (where “real” is hard to define, but roughly just, always better than what I made).

You mention the emotional neglect and the connection to childhood, and I get the sense you’re interested in figuring it out. It made me think you might like Joe Hudson’s work on YouTube.

dceddia

3 か月前議論

I was connected to a VPN and saw the same. Went away once I disconnected.

(Must say I'm not a fan of how, increasingly, taking any steps to preserve privacy is seen as deviant, or justified because bots)

dceddia

4 か月前議論

> I'd like to experience my brain not thinking about anything at all for once

You can! While you’re alive and everything! I’ve had this experience of few times of it being very quiet, and overall my brain is a lot quieter since then (and I see reports from people who’ve gone further/deeper; I get the sense the path is never-ending).

There’s a ton of resources and people out there that basically point to the same thing in different ways. Meditation is one of the ways but it’s not the only one. Some keyword soup if you want to go searching: jhana states, Jhourney, Art of Accomplishment, Joe Hudson, Zen, Buddhism, awakening, Michael Singer, Loch Kelly

There are varying levels of “woo” in this, and if you’re on the woo-averse side, Joe Hudson’s stuff is a good way in.

dceddia

5 か月前議論

Yeah, forgot when I wrote this comment that the thing about keychain was to pass that auth token into a Docker container, which I gave up on (Tauri desktop app needs to compile Rust and link against other stuff, different architecture inside the container blah blah)

More or less what it says in the README:

    fence -t code -- claude --dangerously-skip-permissions

Or wrap it in a function as an alias

    # cat prompt.md | ralph
    function ralph() {
      fence -t code -- \
        claude --verbose --dangerously-skip-permissions --output-format stream-json -p "$@" \
        | jq -r 'select(.type == "assistant") | .message.content[]? | select(.type? == "text") | .text'
    }

dceddia

5 か月前議論

The Claude Code client adds system prompts and makes a bunch of calls to analytics/telemetry endpoints so it's certainly feasible for them to tell, if they inspect the content of the requests and do any correlation between those services.

And apparently it's violating the terms of service. Is it fair and above board for them to ban people? idk, it feels pretty blatantly like control for the sake of control, or control for the sake of lock-in, or those analytics/telemetry contain something awfully juicy, because they're already getting the entire prompt. It's their service to run as they wish, but it's not a pro-customer move and I think it's priming people to jump ship if another model takes the lead.

dceddia

5 か月前議論

I went down this rabbit hole a bit recently trying to use claude inside fence[0] and it seems that on macOS, claude stores this token inside Keychain. I'm not sure there's a way to expose that to a container... my guess would be no, especially since it seems the container is Linux, and also because keeping the Keychain out of reach of containers seems like it would be paramount. But someone might know better!

0: https://github.com/Use-Tusk/fence

dceddia

5 か月前議論

This look nice! I was curious about being allowed to use a Claude Pro/Max subscription vs an API key, since there's been so much buzz about that lately, so I went looking for a solid answer.

Thankfully the official Agent SDK Quickstart guide says that you can: https://platform.claude.com/docs/en/agent-sdk/quickstart

In particular, this bit:

"After installing Claude Code onto your machine, run claude in your terminal and follow the prompts to authenticate. The SDK will use this authentication automatically."

dceddia

6 か月前議論

Interesting about the level of detail. I’ve noticed that myself but I haven’t done much to address it yet.

I can imagine some ideas (ask it for more detail, ask it to make a smaller plan and add detail to that) but I’m curious if you have any experience improving those plans.

dceddia

6 か月前議論

I concur, also with no benchmarks to share, but I had the experience of rewriting a video editor timeline to use WebGL instead of the 2D canvas I started with and it got much faster. Like being able to draw 10k+ rectangles at 60fps became easy, where with 2D canvas it was stumbling.

dceddia

6 か月前議論

With their tagline being “video for developers”, isn’t this their whole thing? It seems like another service would be a better fit if having a management UI is a requirement.

dceddia

6 か月前議論

So I’m probably in a similar spot - I mostly prompt-and-check, unless it’s a throwaway script or something, and even then I give it a quick glance.

One thing that stands out in your steps and that I’ve noticed myself- yeah, by prompt 10, it starts to suck. If it ever hits “compaction” then that’s beyond the point of return.

I still find myself slipping into this trap sometimes because I’m just in the flow of getting good results (until it nosedives), but the better strategy is to do a small unit of work per session. It keeps the context small and that keeps the model smarter.

“Ralph” is one way to do this. (decent intro here: https://www.aihero.dev/getting-started-with-ralph)

Another way is “Write out what we did to PROGRESS.md” - then start new session - then “Read @PROGRESS.md and do X”

Just playing around with ways to split up the work into smaller tasks basically, and crucially, not doing all of those small tasks in one long chat.

dceddia

6 か月前議論

That looks so ridiculous that it has me wondering how hard of a technical change it would’ve been to change that drag target, and if they just punted on it.

dceddia

6 か月前議論

Is it possible for Ghostty to figure out how much memory its child processes (or tabs) are using? If so maybe it would help to surface this number on or near the tab itself, similar to how Chrome started doing this if you hover over a tab. It seems like many of these stem from people misinterpreting the memory number in Activity Monitor, and maybe having memory numbers on the tabs would help avoid that.

dceddia

7 か月前議論

In many cases today “gif” is a misnomer anyway and mp4 is a better choice. Not always, not everywhere supports actual video.

But one case I see often: If you’re making a website with an animated gif that’s actually a .gif file, try it as an mp4 - smaller, smoother, proper colors, can still autoplay fine.

dceddia

7 か月前議論

I had kinda suspected this just based on my own experience of paper vs screen, but hadn’t run across any research.

After seeing your comment I went looking! I found this interesting: https://phys.org/news/2024-02-screens-paper-effective-absorb...

dceddia

7 か月前議論

> That's how it's always been only now you can buy the cert through Azure.

Where can you get an EV cert for $120/year? Last time I checked, all the places were more expensive and then you also had to deal with a hardware token.

Lest we talk past each other: it's true that it used to be sufficient to buy a non-EV cert for around the same money, where it didn't require a hardware token, and that was good enough... but they changed the rules in 2023.

投稿

Toggle macOS battery icon with plug/unplug

Windows NT on 600MHz machine opens apps instantly. What happened?

コメント