matltc

Hah I'm doing this with some models of that same era judging by the CPU you cited. I had frankensteined one laptop from like the best parts out of three I had from that era, got xubuntu on it, running fine-ish but the Wi-Fi antenna ran through the on-board display chassis, which was destroyed. Took off the top clamshell, was gonna put ether in it and use as a desktop, but oops! The clip holding ethernet cable fell out during disassembly! No prob, pop it back open, oops! One of the screws holding the mobo is stripped! (Also, wife goes into labor while everything disassembled) No sweat, get something from ace to extract. I think I ended up throwing it back together and not testing?

Forgot about it for a couple years, wanted to put opnsense on it and see if I could replace stock equipment leased from ISP, nope. No display.

Fun to a point but probably should just spring for a used thinclient at this point, but prices are scary and I hate letting all this hw go to waste

> enough ... to intervene over Taiwan

Probably not. My guess is the Pentagon's calculation would be that the cost to intervene would be unacceptable and could not be sustained anyway; many reports say the US would deplete their ammo stocks for such a conflict in like a week, and that was _before_ the blunder in Iran

> Better ... than none?

For sure. But doesn't solve the problem of "how do I survive if the power grid goes down for over a year?" Best solution is to get out of USA prior to that event. I don't think any amount of prep will significantly change your outcome in that scenario if you remain in the US.

What makes you say that? The cost to China to invade will still be incredible with zero US intervention. Wsj had an article last December <https://www.wsj.com/world/china/how-china-chinese-invasion-t...> on what this might look like

So like a month of food and water? If the big electrical grid pop happens <https://www.nytimes.com/2026/08/18/magazine/national-blackou...>, we're looking at years, which no storage solution can accommodate. So while this is sensible for most outages, doesn't do much for this event, which becomes quite possible if (1) China moves on Taiwan and (2) USA intervenes on behalf of Taiwan. I would expect that China would activate their dormant ICS and telco access in such a circumstance to knock the US out of the conflict basically immediately

I suppose this could be true with the rest of CRINK since they are all capable cyber adversaries, but China is on another level in terms of capability and it's their best lever with little cost. I suppose they have to calculate whether the US would bring out the nukes in such a circumstance

This is one thing that has really hit me: I have a question and I always default to Gemini flash or haiku instead of finding quality writing on a subject. So I get the same tinned bullet-point format every time, and my knowledge has become wider, but can only be so deep--only so many things can fit in one's head at any time--so I pay the opportunity cost of finding something that I enjoy sinking my teeth into.

I think about my favorite bands, or games, etc. the ones I cherish most are those hidden gems that I either stumbled upon during some rabbit hole or as a recommendation from a friend. How do I get that now? I can ask Claude, I guess.

If you do try to go down rabbit holes with AI, you get a deluge of information, but it is basically like what Google has become, minus the ads: aggregates the most easily accessible data, blends it up and distills it to some slurry that might kind of be coherent, and which will definitely sound authoritative, but... Is it?

And, of course, then what happens to the hidden gems that have a very low (whatever the equivalent of) PageRank but maybe because they're not highly optimized, ultra-processed and generally agreeable or popular. The channels through which one could readily receive information are all being flooded out/absorbed. Maybe I should try Kagi, or get into irc and build that rss graph.

Two's complement just had a great episode on this topic. None of us have absorbed the costs of the tradeoff we are all making by using these things, much like we didn't for using social media, or the slide into the techno-corporate surveillance state that happened after the patriot act. We are paying those costs and I am certainly feeling them more than ever recently.

I love goooooold

Appalachian dialect

Amazing what one can learn from rtfm

What is up with the README?

Love how Gates and Nadella had nothing to say until this year, after they clearly lost any hope of building a moat around it

Selling your soul for a discount. I almost never download native apps. If they go to the lengths shown in this article to side-channel fingerprinting, imagine what they do when they have native privs.

I got into this stuff super late. Have kids and a job. Would love nothing more than to go get a proper cs degree. Would probably go for computer engineering. Alas, my undergrad GPA was poor, so grad school is out. Can't picture dropping six digits on a degree from like Trump University or wherever, and seems silly if not impossible to go get a second undergrad. Sucks because my alma mater had an amazing cs dept, I just didn't know or care back then. Oh well.

Especially with advent of agents, I have dropped all illusions of ever getting paid for this outside of like freelance work. Will keep working through cs:app3e for the love of the game.

Well said. I get the sense that C is kind of ossified today, and that the whole point is its compatibility benefit. I wonder what even makes it into the next C standard? I think c99 was the last meaningful one. Everything else is offloaded to the compilers, God bless them. -Weverything

Don't know anything about clojure, but isnt this true with any language or library that implements schema validation? I'm thinking zod, for instance. Does clojure make it trivial?

Saw this in npx ccusage@latest claude output. Had only used opus but showed sonnet. Can't remember if the jsonl retains which model is doing what, but meh

Wouldn't be surprised if there are knobs that get turned as a function of the revenue they might expect you to generate.

I was a 4.6 acolyte from April til the fable drop, lost that quick, cancelled and took a break, came back a month later, tried opus 5 and liked it, so unpinned 4.6.

Results were great at first, and they're still not terrible, but I have noticed a regression in accuracy, so to speak, where I am pointing out issues that are quite obvious in review.

I pretty much use sonnet 5 low/medium when I have a plan to solve a simple problem and depending on scope, opus low/medium for more complex/bigger scope implementation, and only go high when it's very complex or I'm spitballing architecture/solutions and iterating plan. Never go xhigh or max.

The verbosity is insane though, opus 5 documents everything and just regurgitates whatever lead it to the design choice in there, which makes it more opaque because it's talking about something that was discussed once in a session that no one else can see (except their backend ofc)

I don't even try to steer it away from that with harness, because it doesn't work and just ends up agonizing over whether it should write some comment. Three paragraphs waffling on that on verbose output

I did however have it write a script that basically is git add -A -p for comments though, haha.

I'm $20/month, have all my telemetry toggles off, don't really over engineer prompt/context, just some basic skills for repeated patterns.

Dropping this all-timer: http://www.art.net/~hopkins/Don/text/rms-vs-doctor.html

Now it's back to reality

Feeling the finality

Of his greatest achievement

He couldn't believe it

But it's true, now he's feelin' blue

So between me and you:

Think twice before you do

Ironically: Claude Code docs

Tends to latch onto whatever was true at the time of its training run unless you point it to the llm.txt.

Yep! Only (potential) downside to --bare is it runs in print mode; can't kick off an interactive session with it. The flags I listed above are a close approximation.

Anthropic had something about API-only in the description a couple months back, but they've been waking all those back for months since the fable access rollback fallout

You can do this but you have to pass a million long options like --disable-system-prompt --allowed-tools "" etc

I have some wrappers but the only one I really use is a variant of that pinned to haiku for quick questions.

Totally. Ralph wiggum was added to every big shop's stdlib (harness?) just a couple weeks after it dropped. That simple iterative workflow basically underpins all of the notable/novel achievements of agentic coding or whatever.

I'm sure many more examples in the "official marketplaces" for mcp/skills/what have you

Prompt engineering was deprecated around the time Opus-/Codex-class models dropped imo. Replaced by skills up until about opus 4.5/6 dropped. Pinned model to Opus 4.6 slug in ENV.ANTHROPIC_MODEL, disabled the adaptive thinking thing, and saw my skills start causing degradation

Only ones I use today are for very specific quirks (eg wiredtiger/mongodb 8+ incompatible with ext4/Linux 6.19+ specifically causing segfaults. Have a 20 line mongo skill that says as much. Pinned docker container to mongo 7, can prob delete it now)

I spent a few days reading up on the docs for these things, hook lifecycles, tried writing a few, but they never work as documented, or the documentation changes so frequently that whatever you built is deprecated by the time you get it humming.

Now if I have some non-trivial unit of work, I basically iterate on spec in plan mode then put it on auto and let it rip. Way better results with Fable. jury out on Opus 5, but no regression like 4.7/8

Usually it's just echo "do this lil thing then pr closing issue 123" --model sonnet --effort low. Works well enough, sonnet 5 low is a workhorse and quite resourceful in a good way when things go sideways; doesn't cheat its way out IME

Can you explain how you piped recent agent actions/turns/outputs? I was looking into this yesterday and tried playing around with streaming json, including hooks, but i (really Claude) would need to write a parser (prob with jq) to filter the noise. Was thinking we tail -f that and select whatever we care about. Kind of an observer to catch such spiraling.

One of those programs that shoulda been in coreutils

Thanks for the references, especially FMD tip; have the Google version disabled due to lack of trust.

Wow, straight from the source! Heartfelt thanks for maintaining this project; it becomes more pertinent by the day as we slide ever deeper into living into this dystopian surveillance state.

Thanks for the info! Might take the dive.

Curious to hear: how much effort did it take to migrate to GrapheneOS? I own a pixel, grabbed it in anticipation of unlocking, but the effort it seems it would take from reading GrapheneOS docs has stopped me from committing to it on my daily driver. Would you please provide a quick time estimate and any snags you hit?

I am with you but I inevitably think it's a skill issue. Some devs put a lot of work into the harness. I tried doing this, felt like running in place to me. the lack of RFC-tier standardization and ever-shifting api is demotivating as well since you might have written a harness that will be obsolete with next model gen/version bump

There probably is a correct way to do this, but your average "vibe coder", or even a competent solo dev just yoloing a project, are not implementing such harnesses. In the former case, they may not be able to; in the latter, you're probably better off in 1:1 human to agent workflow due to indeterminate/bad outputs and the amount of back-and-forth in prompts/specs/reviews required to get high-quality work

Then again, I saw Matz harness for his compiled ruby project was just like perms.allow find and grep. No skills, no agents.md or anything like that. Would love to see those prompts! I'm sure the fact that it was Matz had a lot to do with that working, though