I don’t think he’s playing 4D chess; I think he truly believes all the “AI is going to eliminate all the jobs” crap. I think his “Claude Constitution” is wishful thinking and his attempts to exert control over what his customers lawfully do with the product he sells them have made his company untrustworthy; certainly so by the US Dept of War.

I think lately his advisors have made him tone down the doomerism noting that it might tank his IPO, and I am uncertain whether his recent pushes towards more regulation are regulatory capture attempts or ideology or both.

The man is smart but IMO shouldn’t be running the company- he should be a CTO and let a business person make the decisions.

As for the government, bureaucracies gonna do what they always do. If you scare them they regulate you. ITAR is a real thing and the government throws it at technology all the time, from the minds that brought you 40-bit SSL in the 90s.

For better or worse, they “understand” and have seen a lot of message queuing code and read lots of message queue support discussions.

Pretty sure this statement is 100% wrong. You’re describing a gambling site. Is that how governments are treating them?

I think that the person misused Google internal information and deserves termination or other discipline, but I’m struggling to otherwise see the harm in what they did. Is insider trading a crime on prediction markets? Doesn’t it contribute to the accuracy of the pricing of prediction contracts, and therefore is good for the prediction market?

I know in some markets crime pays more than legitimate work, but it never ceases to amaze me how much thought, effort, planning, and engineering goes into providing infrastructure IT services for cybercriminals. The people involved definitely have the skills to be profitable at legitimate work; it just puzzles me that they choose to support criminals.

For any nontrivial task I spend 2-8 hours in specification (I spent 3-4 hours on a stateless rust CLI tool design this weekend) and detailed task breakdown in implementation planning.

I use TDD to start with red tests that turn green when acceptance criteria are met.

I write agents to use to check work and they are my enforcers of constraints, as well as fresh eyes. I use these agents for spec review, plan review and code review.

I am actually pretty proud of the projects I create with generative AI. I just apply a lot of discipline so I don’t end up with slop.

I agree with the poster that if you are doing anything that is not a defined happy path for Unifi, it is a freaking nightmare and will likely involve rebuilding, resetting and readopting several times.

Any hierarchical taxonomy classifies on one dimension at each taxonomic level. Invariably someone wants to classify on one criteria when someone else wants to classify on another. Taxonomies that humans use aren’t multi-dimensional. So if there is a disagreement, someone wins and someone(s) has to lose.

No one is wrong; they just have different priorities or preferences or goals.

So now as an architect I never argue (and seldom discuss) taxonomies. I make two points and then bow out:

1. Whatever your taxonomy is, you need a rubric for each level. You need a procedure or set of questions that unambiguously map any $THING you encounter into exactly one bucket. Validate that competent people with no specific domain knowledge can properly classify things with your rubric; it must be repeatable by amateurs, not just experts (software is dumb).

2. Existence trumps theory. If there exists a taxonomy and rubric for what you’re classifying, you need to provide a $DARN_GOOD_REASON why this wheel needs reinventing. Personal preference and your 1% edge case probably don’t justify all the work to reinvent everything.

Then, I go back to the implementers and tell them to design in a tagging system, which is a DIY taxonomy, and except in ridiculous use cases, I can make indexes make it fast enough to let everyone overlay their own classification system.

I once worked for $COMPANY and we had a network scanning application. Always generated a lot of tickets from angry people wanting to scream about bots.

So we put a web page page on each worker that would inherit some details from whatever job was running, and say “I’m a $TYPE_OF_SCANNER FROM $COMPANY doing $THING_THAT_BENEFITS_YOU.

This behavior is covered by our terms of service page at $LINK.

If you believe that we should not be doing this, please contact $SUPPORT and provide this code:

$SCAN_JOB_IDENTIFIER”

Call volume and unhappy customers went way down.

You need to start using SI points that are defined using wavelengths of ground state emissions of a decaying Americium atom.

I wanted to ensure that my writing was captivating and that the audience didn't just read the summary and skip the rest.

I used a chat agent to build a research report on the academic literature about what makes writing boring or not-boring ("not-boring" is NOT the same thing as "interesting"). I synthesized that into a taxonomy, and built tooling for mechanistic analysis and prompts for LLM analysis, synthesis and recommendation.

What this is: research-grounded analysis of a document with concrete evidence-based suggestions of how to improve it.

What this is NOT: it's not a simplistic prompt "Help me make my document better" or even "make every section/paragraph/sentence earn its place". (I am developing a separate skill for that in the same repo, but with a much more diligent approach).

I hope this helps you improve your own writing and effectiveness in business communication.

Comments URL: https://news.ycombinator.com/item?id=48039841

Points: 1

# Comments: 0

It’s a straightforward technical problem to wrap an API or MCP or something around the “create an account” function.

But what will a court do when the agent creates a million accounts, mines bitcoin for a month, and then cannot or will not pay?