Brand-new Cessna planes were selling for roughly 2x the US median annual income in the 1960s. Today that number is at around 5.5x.

I've always wanted to fly my wife & kids in a plane, but I just can't afford a high-maintenance luxury asset that costs as much as a house, and fuel costs are prohibitive as well.

so yeah, if someone could mass-manufacture an FAA MOSAIC-certified light-sport aircraft with >2 seats, that could bring the cost of fractional ownership and once-every-two-months weekend use down to my level. And if small local air strips could stop closing too, that'd be great.

---

All to say, aviation used to be semi-reachable as a middle class hobby for family people. Now, the only way to get a family in the air for people like me is to spend years building a kit plane from scratch or fixing up a cheap broken plane.

ur wife and kid will want in flight wifi

Right now the comments that upset me the most are LLM TMI-style comments that break encapsulation by talking about the behavior of specific current callers of a function right above the function definition.

I recently reacted angrily in a PR review comment after encountering one for the umpteenth time... that caught me off guard. I didn't know I was capable of that.

This is what has been frustrating me most lately. Even though I have a rule in my global CLAUDE.md that says:

> Only write comments to explain the why when it is not obvious from the code (rationale, gotchas, constraints). Do not comment on the what — well-named code already says it. Do not comment on how a framework works.

It still keeps adding these bad comments. When I then ask it to review the comments based on my preferences it then deletes most of them or improves them.

Today I asked Claude why it disrespects my preference and it said that the surrounding code was like that and it followed that style. It suggested I add this line to my global CLAUDE.md file:

> The comment rule above beats the style of the surrounding code: neighboring files with what-style comments are not license to write more of them, and comments carried along when porting or copying code must be re-judged against the rule, not kept for consistency.

Let's see if that improves things.

In Claude Code there are also "output styles" that are more deeply embedded - into a system prompt - and agent is also periodically reminded of them during the session: https://code.claude.com/docs/en/output-styles

Maybe these would work better for such cases.

My hypothesis is that Claude is inclined to write so many comments as a way of doing additional thinking

Yeah, it's short form memory or something. Like writing a todo list.

This drives me nuts as well. I hate also hate when LLMs use plan-document references in comments/doc strings too. “Landed in stage 1…”

I have a lot of CLAUDE.md rules to restrict this stuff, but realize the “encapsulation” language is something I’m missing.

There's a cluster of issues like comments like "use a set instead of ..." after changing something which will just confuse people in the future. Or comments referring to irrelevant details of the planning/implementation process.

It's as though the machine can't separate the chat and planning docs from the code itself and so they meld into each other. As though it can't fully grasp that the code will outlive the current session by years.

Anyway I find a checklist approach works well to sort this out. I don't consider looking at machine generated code until after a checklist covering all this sort of stuff has been applied. My checklist approach currently has about 50 items which I have the machine apply by splitting it up across about 20 subagents. Pretty silly but it seriously improves the first-pass quality vs only a few subagents. I find this checklist can effectively eliminate words like "genuine" and "landed" from the code too. Eliminates vague and made up terminology. Makes it less nauseating

Yeah, agreed. These have started popping up a lot more recently, where I get a 5 sentence paragraph explaining how function overloading works in c++.

Claude Code and Opus 4.8 love to describe changes in comments (perhaps because that’s what’s on its “mind” at the time), like “this used to do A but that did a bad thing so now it does B”. I’ve almost convinced it that changes go in the commit message, not the comments.

I have tried prompting it out and providing strong guidelines in my AGENTS.md against it, but I still get _way_ too many useless "explain the code" style comments no matter how much I try. I usually have to do something like "Look at all commits in the past X days and remove (DO NOT TRIM) all comments that are not truly exceptional"

Normally when I can't get claude to follow a prompt I try a lint hook, but it's tough to lint something that subjective.

Well, on the one hand, Hyrum's Law: "With a sufficient number of users of an API, it does not matter what you promise in the contract: all observable behaviors of your system will be depended on by somebody."

so that can be useful information in some situations.

On the other hand, what a horrible out of date mess of comment that can turn out to be a little bit later. Taken as gospel by the next entity (human or llm) to massage that function.

Ugh, another LLM-voice article.

I'm getting to the point that I almost have a physical feeling of revulsion/nausea when I read something outside an open Claude Code session that is written in an AI voice.

I'm so sick of that voice -- word choice, phrasing, style... it's hard to define and frustrating because it's both pervasive, influential on human writers including myself (I found myself using the phrase "doing a lot of work" in a Slack message about a word someone chose to use :grimace:), and presumably compounding on future LLMs.

Same here. I started describing it as the “uncanny valley” of text. It’s like a gut reaction that something is off with the text even if you can’t pinpoint it immediately.

I concur. It's interesting that this seems so difficult to get rid of; despite actively trying to overlook the taste of LLM writing, it still grates, and this does not happen to me in the same way even with communication from half-illiterate human idiots.

It's kinda fascinating how far LLM coding/problem solving has come without making much progress on the annoyingness of their writing. Any theories?

Not that it's an ideal solution, but you have all the tools you need to do something about it. You can put all your preferred writing style guidelines into an LLM and have it ingest and rewrite any text you dislike.

That's like saying a robot will be a better friend if I use a more skin-like material for its face.

The LLM-voice revulsion isn't just about particular idiosyncrasies. Another commenter put it well - the "uncanny valley of text".

Only if the problem is ideological, ie. you don't like the "idea" of llm's. But if in fact your actual dislike is the style of writing, then you can simply change the defaults. It's pretty easy to make a llm conform to your preferred style, avoid the use of em-dashes, or reduce verbosity, etc.

Sounds like that word was pretty load-bearing

It's probably doing a lot of heavy lifting.

I need to push back on this comment.

If only there was a smoke test for this kind of thing.

The TZDB could handle it though

I had not seen that. In the nineties I worked on an alerts system where you could sign up for like some sport or weather data at a certain time of day. We stored the alert times as minutes before midnight and then ran the time to trigger calculations often enough that unless you were some freak that wants weather alerts at 2 am it basically worked to send one alert each day at the appropriate time; we had a special non-OS copy of the tzdb as the users were global. One quarter I forgot to update it and everyone in Mexico City got their alerts one hour off till someone complained and I updated it. We also had data feed alerts, like score changed or stock hit x% over previous high, where the problem is some data is manually entered and can be off by a factor or two of ten from time to time. Had to be filtered. I had a lot of fun.

Some kinds of brains must just not be good at this kind of game.

I used to win spelling bees in school, but I played the entire archive on this game and my highest score was 6/18.

Maybe it has to do with the balance between audio and visual learning. Spelling bees are spoken words in, letters out -- not letters in, words out. When trying to solve these, I kept trying to sound out different orderings of letters, maybe that's not how good players do it?

I think I get anchored too easily based on what my brain sees in the scrambled letters.

A re-scramble button would do wonders for me like in that Spelling Bee game on NYTimes.

It appears macOS native containers runs a separate Linux VM per container.

OrbStack's claim to fame is that all containers run in a single Linux VM, with lots of optimizations on both sides of the VM boundary (including use of a sparse image file for disk storage, which saves a lot of space on the macOS side).

If you run more than 4-5 containers on macOS, the performance and resource usage savings of OrbStack really starts to add up quickly.

> He is a PHP developer from the early '00s who got lucky with Facebook

Nice summary of a good hacker's viewpoint. All his money, power, and other non-technical signals don't really give him the technical authority to weigh on what really makes a technical difference.

That being said, depending on how good information-sharing is at the company, he might have a valuable perspective. But given that morale is low, it's doubtful that engineers feel free to tell the straight unvarnished truth without fear. And in typical large-company fashion, managers above them likely spin things positively as much as possible.

It's concerning to me how the author generated AI images and gave them captions as if they are real depictions.

If I'm going to read an article about something real in society, I'm only interested in images that are true depictions of reality. They may not so perfectly capture what the author is envisioning, but the gap between "perfect for this article" and "real" is what makes a real picture interesting.

1Password checks all these boxes and hasn't yet had a data breach.

Their biggest security hole is probably somewhere in the operational pipeline between 1P browser client developers and the static file servers hosting them.

Unfortunately it's one of the most bug-ridden and unreliable pieces of software I've ever used. I encounter issues with it on a daily basis, but the burden of switching and a lack of superior options keeps me locked in.

I stopped paying them when they killed local valuts, and secondarily when then moved away from native apps. I drifted along on the old 7.x client for awhile with local values.

I've more or less switched to apple keychain/passwords at this point. I need a solution for linux, and have been thinking about some kind of simple 1-way sync issue that dumps stuff from keychain into some other tool for use on linux.

Curious if you have any gripes or concerns about using the Apple keychain/passwords setup. Aside from Apple devices, do you mostly also stick with Safari? Was it hard to transition things like TOTP or passkeys?

i mostly stick to firefox I do some management of moving some passwords back and forth (i'm not yet using the firefox extension for apple passwords because i just learned about it).. but because i use firefox on my phone as well.. nbd.

In terms of TOTP I just use googleauth and oathtool.

I've been pretty happy with Bit Warden on Linux, though I just use the Firefox extension and am not sure if there's a native app

It was a fantastic, fast, reliable piece of software until they sold out for VC funding and went the Electron rebuild route.

I was a paying customer for 15 years and migrated away with the last price increases due to "AI-powered functionalities" and the new features making the product worse on top of already being salty over when they stopped the one-time licenses in favor of subscription.

Used it for years and never encountered a single bug, and I'm quite a power user with hundreds of items stored in it, shared vaults, and access multiple times per day. It's one of the few softwares I happily pay for. Maybe it differs from platform to platform, otherwise I can't explain your comment.

Any example bugs that you've encountered in the last week?

Their flow for regaining access after somehow "disauthorised" laptop, she there's an installed but unused for months plugin is one of the most infuriating.

It won't ask me for my secret key, which I have an can provide immediately, no, it won't allow me to authenticate myself with the phone, because our enterprise vault logs off quickly, I must however do a some absurdly obscure dance because FY, that's why.

Ugh, the enterprise authentication is dumb

1P is open source now?

Optionality. The power available to any one person to do anything in society is directly proportional to their genuine connections to others, even if brief.

Meetups in person with real connections happening (even if brief) create so many new possibilities for those in attendance.

The disappearance of in-person meetups (especially among developers) in the US after Covid is a tragedy - so much could be happening that is not.

The Manifesto for Agile Software Development only happened because a group of developers met up in person. They disagreed about nearly everything, but what they agreed about changed the industry.

So he is the new Epstein of the AI age, just without the obvious problem parts. He has his own antichrist-grade problems which are in a way worse.