smccabe0

Wireguard. I cannot tell you how valuable it is to have my own WG instance that lets me create a VLAN for all my devices. A simple webapp from my home PC available on my phone to browse notes I've taken in Obsidian gets more use than cloud services of a similar fashion now a days.

And since we're adults, two things can be true at the same time. I'm not a fan of the way sanctions are being used, and I'm similarly not a fan of the premise of the ICC.

Thank you for actually addressing the philosophical elephant in the room. There's way too much vitriol substituting rational thought when it comes to courts in general, doubly so for international courts. They are compacts and agreements, no matter what scale you're looking at, and there's a reason depictions of "Justice" as a lady hold scales and a sword. Neither one works without the other. The term "outlaw" used to mean outside the protection of the law, in both directions. What do these courts protect America from?

That being said, sanctioning them feels a bit odd. Might be some latent jingoism but I don't see a threat from the ICC being substantial to US citizens.

`--as-of` might be the most important feature in your codebase. Temporality of relationships is way way more important than I thought when I first started down the path of modeling relational data in agentic systems. Sometimes knowing why or why not at a glance can be the difference between spending another $50 on tokens retreading old ground. This is very cool, and I will probably discard my handrolled version in favor of your much more robust solution.

Have you considered allowing annotation of why something changed as part of a tx?

There's some empirical backing to this if you consider what LLMs doing as part of the same regime: LLMs take a token stream and inflate it to the N-dimensional space in their embedding. We take a string of words and apply it to our model of the world, our understanding, and memories. I've had a lot of success in understanding the math through that lens.

I think the section on AI from Zero to QED (a proofs in Lean/lang guide) gives a sober path forward from the perspective of market-makers and trading:

"Imagine market infrastructure where agents must prove, before executing, that their actions satisfy regulatory constraints, risk limits, fairness properties, and eventually machine-checkable proofs of Pareto efficiency of market mechanisms. This is a big, hairy, ambitious goal. Not “we reviewed the code” but “the system verified the proof.” The agent that cannot demonstrate compliance cannot act."

https://sdiehl.github.io/zero-to-qed/20_artificial_intellige...

It has and it hasn't. There's modders in the scene who I refuse to use mods from because I know they also have developed cheats.

All of the same supply chain issues you have with packages in a programming ecosystem exist in VS's mod system.

So now that another admin is in power, it's useful to have accurate numbers. But during the last several administrations, job posting numbers were used to backstop failing economic activity, especially post covid.

I'm not mad it's happening, I'm mad it's taken this long to do.