I still think Anthropic should've bought Clawdbot/OpenClaw. Feels like a missed business opportunity to expand your market share by capitalizing on the hype.
"now at OpenAI" were my original words - they did the equivalent of an acqui-hire and "protected" OpenClaw in a foundation.
In the context of the seemingly aggressive machinations of Anthropic your hair-splitting without clarifying beyond "OpenAI didn't buy OpenClaw" seems itself misleading and rather counter to helping conversations progress.
I vibe coded the shit out of a Chrome extension that does that while waiting on CI/CD. Go read the content.js to make sure I'm not hacking your shit, download the repo to your computer, enable developer mode in chrome, "load unpacked", point it at the directory with those files, and enjoy your tool tips.
The tmux usage referenced at the end of the article was fascinating to watch. I’d never considered using tmux as a way of getting more insight into how an agent is working through a problem. Or to watch it debug something
A protocol defines a set of primitives and their interactions, along with prescriptions where necessary, but is otherwise pretty loose generally. Especially in a Postels law world. APIs tend to be highly prescriptive with a highly specific interface and a lot of black box.
Think of the difference between an IETF RFC on say SMTP vs an email API - the RFC describes how clients and servers for mail routing interact through an almost dialog, while a typical email API has highly structured interfaces. Another wat to cut it is an API can be tested as it has inputs and return values dependent on those, while a protocol you can generally only assert compliance with the specification of the protocol.
People often assert protocols have something to so with RPC of some sort but that’s not true. Many language support protocols, which can be very similar to interfaces, but don’t have anything to do with OOP, etc. In language protocols it’s slightly different than network/IPC protocols but the intent is similar.
I'm not the most knowledgeable, but a protocol talks to another process through a specific format.
I personally think its more powerful than writing a new process to replace and existing.
My favorite example is an X11 windows manager implementing in about 18 lines of python.
Obviously there's dependencies to talk to the X server, but the power of a protocol comes from any program written in any la gage communicate with existing code.
I created my own X11 window manager [1] at the start of this year in around 800 lines of C.
I had been using dwm (4000 lines of C) for many years and wished to write my own for a long time, but what made me take the leap was really steveWM [2] and TinyWM [3] which are both super small.
They're both poorly defined, but what I think GP meant is they want something that you use something like a socket instead of FFI to interface with. You need an extra data description layer for a 'protocol' in this context because you can't rely on something like the C data model and calling convention as a given.
The folks at SaySomethingIn, that originally started with Welsh and other Celtic languages, have recently expanded to Japanese. I haven’t tried it myself, but I’ve found some decent success with one of their other courses. It’s all about spaced repetition and focuses exclusively on listening and speaking.
I think it’s not just the camera LED, but the indicators that appear on screen, like the amber, green, or blue dots that appear in the menu bar when the microphone, camera, or screen recording are accessed by apps.
Feels like generated AI art like this is modern clipart
reply