I would have loved to go to this.

Yeah I wouldn't bet on either of those happening soon. I think the only ubiquitous one is an OAUTH-based login.

Yeah I imagine that won't be pretty. But I'm down in OC now.

In Dropserver I'm working towards a multi-auth approach. For now you can sign in to an app if you are a user on that Dropserver instance. But if not, the owner of the app can connect it to Tailscale and share the node with you. And then *that* is how you sign in to the app.

Auth for small self-hosted things is a real pickle. There are so many ways to authenticate yourself. Pick one and you leave some people out. Pick many and auth code grows bigger than your app. Use a 3rd party service and now you're reliant on a 3rd party.

So much for my bike ride this weekend

Nice write up of using a Mac to self host stuff. I think having to use Linux is a barrier for most people who would like to self host, but most self hosting things only run on Linux (or docker).

I’ll plug my own project dropserver.org which now has Mac native binaries, and Tailscale integration.🎉

What xoncerns you most here? They are reviewing all the code as always, so do you mean the environmental and ethical issues of AI generally? Or do you think human code review as they describe it is insufficient?

Own It Forever

$1,100
one-time
Regular $1,500

😵‍💫

With what? An infinitely large SCUBA tank?

Also concerning: how will I ever recover my lost piece of bread?

Same. I always skip over Medium links, unless it's really the only resource there is.

I don't know why people still post there.

Write code to make being human easier. Rip code out if it makes being human harder. Write code to make caring for each other easier. Rip it out if it makes caring for each other harder.

Empathy driven development has literally never failed me. It’s done me better than everything else combined

Awesome. Wanted to go to the LA show but no luck.

I think you might want to stop by the florist if you want to survive this.

Truly personal AI will require credible exit for agent memory. The alternative is gacha game girlfriend AI. Credible exit helps reduce the principle-agent problem by letting you pick up and leave.

And I wish there was more skepticism exhibited towards giant corporations who exist to make profit for shareholders. I don't care if people like pods, but I *do* care if options which let consumers figure out ideal dosing for themselves are artificially worsened as it sure seems like they are.