Hacker Newsnew | past | comments | ask | show | jobs | submit | bradgessler's commentslogin

I want macOS on my iPad Pro so I can plug my Moonlander keyboard into it and have a “laptop” that doesn’t destroy my wrists.

You can have that today. If you get a USB-C breakout for the dock, it'll treat the Moonlander, or any keyboard, like a normal keyboard. You can not destroy your wrists right now, as I sometimes choose to do.

I’m pretty sure they are saying they want both that keyboard plugged in and MacOS rather than the limited iPadOS; not that they think MacOS is required to be able to plug a keyboard in.

Exactly. iPadOS can’t run a Unix shell and docker, but macOS can.

I don’t even want touch to work on the iPad running macOS. Just let me run my own input hardware against it.


MCP has been a weird ride. I built https://terminalwire.com before MCP was a thing to make it way easier for people to add a CLI/TUI to their web apps/SaaS.

Then MCP comes out and AI explodes, sucking all the air out of the room for non-AI tools.

Now it seems like AI can work with CLIs better than MCP, so I’m tempted to slap AI integration all over the project to better convey the idea.

It’s crazy how quickly MCP has run it’s course and watching an entire ecosystem rediscover things from first principals.


I just setup private video for https://beautifulruby.com and can confirm the trickiest part is setting up and hosting HLS.

I extracted a RubyGem at https://github.com/beautifulruby/hls that I use to point at a folder full of videos, then my scripts converts them into HLS and uploads them to a private Tigris S3 bucket. I then have to rewrite playlists from the server with pre-signed S3 URLs.

It’s not that it’s difficult per se, but it does require a meticulous attention to detail to put all the pieces together.


A few things:

https://terminalwire.com streams a TUI/CLI from web apps without having to build an API and

I recently finished 45 videos at https://beautifulruby.com/phlex, so now I’m talking more about them and

Updating https://sitepress.cc to work with Rails 8.1 and thinking about another set of videos for it at https://beautifulruby.com/sitepress.


Quite the plot twist there

I’m surprised 1Password isn’t trying to get into this game.

Imagine a browser that lets people define sandboxes and scope credential that an LLM may access to complete certain tasks?

I’d love to have a browser download all of my utility bills for taxes. I’d setup a browser, limit it to a few URLs, give it access to a few credentials, and then let it rip.


I recently finished shooting a video course for Phlex and I found that naming the Turbo Stream section at https://beautifulruby.com/phlex#unit-5 was the most challenging because of Turbo Streams and Turbo Broadcasts.

My recommendation: Turbo Drive & Pagemorphs are what most people should use for most problems. Drive reduces flickering between page loads and Pagemorphs list to channels on the server to reload the page if something changed.

I think Turbo would benefit from positioning itself more closely to Rails and simplifying the terminology. If you look at the https://hotwired.dev website, you won’t find Rails and you’ll find a bunch of words like Stimulus, Drive, Streams, Broadcast, etc. The docs aren’t entirely coherent either.


What do you mean by “Pagemorphs”? A quick Google search suggests you’re the only person using this term so it’s hard to know what you’re recommending. I think it must mean e.g. https://turbo.hotwired.dev/handbook/page_refreshes?


Been using Turbo (and Turbolinks before it) for 10+ years, mostly outside Rails. It's awesome in that context.

Can't really see how making it more Rails-centric would help - more likely it'd just cause a fork for everyone using Hotwire without Rails/Ruby.


Any plans for purchase parity for South East Asia?


Shoot me an email, brad@beautifulruby.com


Exactly that.


I keep sending links like this to AWS when their sales reps reach out, then I ask for a discount, and they never get back.


I love that this was completely overshadowed by "Zed works in Windows!"

This team ships so much that if they sold an LTS product, it means they'd support the release for 24 hours.


I’m surprised none of the frontier model companies have thrown this test in as an Easter egg.


Because then they would have to admit that they try to game benchmarks


simonw has other prompts, that are undisclosed. So cheating on this prompt will be catched.


What? you and I cant see his "undisclosed" tests... but you better be sure that whatever model he is testing is specifically looking for these tests coming in over the api, or you know, absolutely everything for the cops


You are welcome to test it yourself with whatever svg you want.

I am quite confident that they are not cheating for his benchmark, it produces about the same quality for other objects. Your cynicism is unwarranted.


OpenAI / Bing admit it's in its knowledge base.

are you aware of the pelican on a bicycle test?

Yes — the "Pelican on a Bicycle" test is a quirky benchmark created by Simon Willison to evaluate how well different AI models can generate SVG images from prompts.


Knowing that does not make it easier to draw one though.


It doesn't make it harder.


What is special about the prompt


All of hacker news(and simons blog) is undoubtedly in the training data for LLMs. If they specifically tried to cheat at this benchmark it would be obvious and they would be called out


> If they specifically tried to cheat at this benchmark it would be obvious and they would be called out

I doubt it. Most would just go “Wow, it really looks like a pelican on a bicycle this time! It must be a good LLM!”

Most people trust benchmarks if they seem to be a reasonable test of something they assume may be relevant to them. While a pelican on a bicycle may not be something they would necessarily want, they want an LLM that could produce a pelican on a bicycle.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: