Or that it was caught but didn't surface fully before release?

A helpful governance policy here might be that anything that mutates user content without opt-in consent requires a distinct sign-off or a double sign-off. If the goal is to prevent this from happening in future.

Would you be at all interested in expanding that project to outlast you?

And even if you personally wouldn’t be so inclined, surely you know or have met people who might?

Once you have the self replication, expanding scope may just be additional code…

local-governor/epics/e-epics/e014-clinical-domain-model/runs/run-e014-01-ops-catalog-20260302-173907-244c82

- Attempts

+ Steps

  - Step 1

  - Step 2

  - ...

  - Step 13

job_instance.json

changes_final.patch

run_report.md

improvement_suggestions.md

local-governor is my store for epics, specs, run records, schemas, contracts, etc. No logic, just files. I want all this stuff in a DB, but it's easier to just drop a file path into my spec runner or into a chat window (vscode chat or cli tool), but I'm tinkering with an alt version on a cloud DB that just projects to local files... shrug. I spend about as much time on tooling as actual features :)

I’ll usually work out the big decisions in a chat pane (sometimes a couple panes) until I’ve got a solid foundation: general guidelines, contracts, schemas, and a deterministic spec that’s clear enough to execute without interpretation.

From there, the runtime runs a job. My current code-gen flow looks like this: 1. Sync the current build map + policies into CLAUDE|COPILOT.md 2. Create a fresh feature branch 3. Run an agent in “dangerous mode,” but restricted to that branch (and explicitly no git commands) 4. Run the same agent again—or a different one—another 1–2 times to catch drift, mistakes, or missed edge cases 5. Finish with a run report (a simple model pass over the spec + the patch) and keep all intermediate outputs inspectable

And at the end, I include a final step that says: “Inspect the whole run and suggest improvements to COPILOT.md or the spec runner package.” That recommendation shows up in the report, so the system gets a little better each iteration instead of just producing code.

I keep tweaking the spec format, agent.md instructions and job steps so my velocity improves over time.

--- To answer the original article's question. I keep all the run records including the llm reasoning and output in the run record in a separate store, but it could be in repo also. I just have too many repos and want it all in one place.

They weren’t just saying ‘AI writes the boilerplate for me.’ They were saying: once you’ve written the same glue the 3rd, 4th, 5th time, you can start folding that pattern into your own custom dev tooling.

AI not as a boilerplate writer but as an assistant to build out personal scaffolding toolset quickly and organically. Or maybe you think that should be more systemized and less personal?

I found it again when mail order Netflix first came out in my early 20s. It was like rediscovering my childhood and completing a character arc.

Accessing information on a customer's server or tenant (I have been assured) would require a court order for the customer directly.

But... as an 365 E5 user with an Azure account using the 4o through Foundry... I am much more nervous than I ever have been.

It’s not exactly about rules consistency either as stated by the GP, though that’s part of it. It’s more about strong consistent application of scientific principles even theoretical or untested principles.

This is in contrast to futuristic fantasy with no real focus on the science. But futuristic or space fantasy can be very consistent just like magical systems in fantasy can be very consistent. Hard scifi has to be constrained by plausible consistent science and that science is typically a main character in the story, or even THE main character.

Addiction is highly dependent on the immediacy of the drug’s effect on the brain. For this reason, people who smoke cocaine (directly or as crack) are highly likely to get addicted. It may be the majority. People who snort it are less likely to get addicted. And there are a lot of casual cocaine users (snorting) who do not develop a long term life altering addiction. I believe he (researcher friend) said it was 10-15% who go on to become addicted. Still substantial and dangerous but much less than smoking.

His research was looking at delayed onset of cocaine in rats after they pushed their cocaine lever. At longer delays more and more rats showed little interest.

This is part of the overall addiction picture. Decoupling drug use behaviors (smoking, snorting, and lever pressing) from noticeable drug onset prevents reinforcement of the behavior and makes addiction less likely, often much less.

This would explain why nicotine patches and gum would potentially be much less addictive than cigarettes.

https://www.bipm.org/en/si-base-units/kelvin

Anyway, TIL.

At least that’s my attempt to defend the GP’s statement.

But maybe there was more to it?

By contrast, the meaning of the title of the sequel "a deepness in the sky" is fairly literal once you get to know more about the alien species and their habits.

So typically there's a lot of exposure, facilitation, and processing work FIRST. There's not really a route to forgiveness in my opinion until someone is pretty deeply in touch with all the feelings associated with how they were treated. I take a mindful self-compassion approach to experiencing and reprocessing memories and emotions drawing a lot from Paul Gilbert, a self-compassion researcher and advocate in the UK.

Once a lot of the feelings have been processed, including anger and resentment, we often get to a point where the anger doesn't seem to serve much purpose for the person anymore. It's not covering up something they didn't want to feel (guilt, shame, sadness, fear, etc) and it's not "punishing their parents" in a way that many people hold on to. More often than not, the client just spontaneously forgives at that point, and forgiveness looks like seeing the situation as a bad one that everyone involved was stuck in and no longer having a need for the perpetrator to suffer or take responsibility.

So acceptance and compassion then processing and reprocessing with self-compassion. Sometimes a discussion of blame and forgiveness and what they do and don't mean.

I think it actually relates to the original article. There’s a difference between mere interpretation and what actually happened.

“You should forgive your parents, because one day you’ll be older and see their perspective” collapses “interpretations develop and mature” with “some events are a problem”.

Both can be forgiven and it’s probably a good idea to do so. I think it’s not helpful to generalize in that way.

Also abuse happens a lot. It may not be the majority but it is NOT rare.

I stand by my objection in this case

Yes, as we age we come to see things from a more experienced perspective and the perspectives of the adults we grew up with and change, but going all the way to "that is why it is important to forgive your parents" is a big big step.

It is typically a good idea to try to get to forgiveness, you're right, but there's a lot of very indefensible behavior out there.

But I think it would be crazy to not to differentiate between immediate experience and what we've been told. Not even because immediate experience is always more accurate. Sometimes it is NOT, but it's a different source of information subject to different problems. Often more trustworthy but not always, though the "not always" can be ameliorated a bit by understanding some of the limits of personal experience.

I was really only taking issue with "interpretation is all we have" applying in the original story - that there is a difference between "my interpretation about something I experienced" and "my beliefs about a thing I did not experience".

Yes the author's story changed, but it changed because he found out that he was lied to by the police (or perhaps, if we want to be generous, "unintentionally misled") not because his memory was fallible.

To get back to the point I took issue with, in the story "the facts" mattered an awful lot. It was a lack of access to the facts that caused the problem not "an incorrect interpretation" of what the author experienced. The latter happens all the time, but interpreting our experience differently (e.g., reprocessing a traumatic memory with self-compassion and seeing it as unfortunate and something to learn from) is a different thing than finding out what we were told was a lie. Both change our story, but one is indeed a reinterpretation and the other is a change in belief or knowledge.

I think it's important to separate those two things. I think some want to treat them as the same. I think that can cause problems.