iwwr
iwwr on Hacker News
Posted by
iwwr on Tuesday September 15, 2026 @06:51AM
from the show-hn dept.
https://github.com/e6qu/sclaude
I've been working on and off on this project to have a reasonably easy way to run a docker sandbox on my machine. Recently with claude "auto mode" not working as "automatically" as before I resurrected this. Hopefully it is of use to you too.Usage: sclaude / scodex just like you'd call claude or codex.To update: sclaude update (updates the script and tooling)It runs locally in a...
Thanks for the clarification!
AFAIK continuously welded tracks (like those used in high speed rail) are also slightly tensioned, so a break in a single point could make it look like a whole section of track is missing, as tension is released.
What is it using for interconnect?
Each step in that chain is harder to do and more computationally expensive.
Could you use a unionfs like overlay2 for more advanced pid hiding?
What about auto-formatting the json and sorting all keys, to create some kind of canonical form? Then we can use standard textual patch.
Genuinely interested. What problems did you have dealing with the standard reconciliation mechanism provided by ArgoCD and by k8s itself. I understand the advantage of the operator approach, but it might be hard to show the state in ArgoCD and somewhat breaks the idea of gitops.
Can we benefit your project in a more limited but agentless way? Limiting the types and CRDs we allow in k8s makes operations better, especially with the aggressive upgrade cycle that k8s already imposes.
Why do you need an in-cluster operator/pod?
This is a problem for me as a language learner on mobile. I set a keyboard with (for example) Russian, and suddenly the world looks Russian, even though I am far from comfortable in it yet. I wish there were a way for the browser to ignore some system languages. This is worse for apps because I also get served with localised versions, where I need to uninstall, remove system language and re-install.
Is this useful for large json files, on the order of GiB?
Deep down in the sun photons are absorbed almost as soon as they are created. In a sense the photons at the surface of the sun are the result of nuclear reactions thousands of years back. However neutrinos escape pretty fast and so can be used to gauge the status of nuclear reactions now.
Or maybe switch to an operating system that isn't a security dumpster fire?
In some restricted environments (e.g. containers in CI, build servers or AWS Lambda) you may have limited disk space or no write access at all, so it is useful to have a tool that can stream to memory while decompressing on the fly, and discarding what you don't need.
> Who wants to send an email through the smart speaker?
People of limited mobility, without working limbs, holding babies, cooking etc.
The ancient Roman concrete was very different from modern: it was more thick, could not ne poured etc. But more importantly, it had no rebar, which is the main thing that corrodes over time. Ancient buildings could use molten lead to join together pieces of stone, but steel could not be made in the volume needed for reinforced structures.
Thanks!
So basically the fragments are escaping through the aerogel.
But why Aerogel? Is it easily compressible?
> The goal is to define a datatype by cases, where one can add new cases to the datatype and new functions over the datatype, without recompiling existing code, and while retaining static type safety (e.g., no casts).
Isn't this made moot by the prevalence of static linking or fat binaries in go-land?
It's far easier for a project with many contributors. Plus, PRs and branches are cheap.
Nuclear appears to be the only realistic contender. But the mission architecture of Artemis goes around the problem by having only short stays on the lunar surface.
Even with nuclear, it can be far safer: small reactors need less containment. There is no risk of fallout without an atmosphere, and no ground water to contaminate.
Still, relying too much on it and failing (melting down) will put in danger the whole human space program. For low-power unmanned applications the RTG remains unbeatable.
Could be that content creation tools are hard to do and the need for moderation. Second Life had a problem of easily accessible penises, while Minecraft servers have an issue with swasticas.
Some call it "chaos", others might call it "badly written software".
Maybe start scaffolding first. Split up the docs into standard sections: who this is for, what it intends to describe, level of detail (marketing, executive summary, runbook). Then fill in the sections at the level needed.
That one had some fundamental problems due to the choice of graphite moderator. Another selling point: the ability to refuel while the plant is running also did not pan out. In the end these reactors had somewhere around 50% capacity factor, compared to more mainstream designs that used water (light or heavy) at 90%. They are also being decommissioned early, again, due to graphite cracking over time. That said, it's not unfeasible to imagine these problems being ironed out in the end, but rollout coincided with the Thatcher Era, which deprioritized industrial developmend and fundamental R&D. It was not a good deal to buy one of these 1st gen UK reactors, compared to a GE/Westinghouse LWR/HWR.
Besides just cost, the engineering constraints can be lower by making designs less complex and easier to do, also, faster. Free delta-v or smaller size constraints, as well as more predictible pricing. Or, 2 for the price of one, or, split up instruments on multiple spacecraft. Reusable spacecraft are a little like that :)