theshrike79

It does affect things though. Like reading books.

I _HATE_ reading books that spend pages and pages explaining how things look to a very detailed degree even when it's not relevant to the plot in any way - it's just set dressing.

I can't see that. It's just word noise to me.

Now I noticed my kid has the same thing, they like other books and actively dislike others. It's the same as me, without me affecting their judgement in any way. Any book that spends too much explaining the environment is boring, books that just get on with the plot are a bit hit.

That's been my goal too, know a bit of everything - enough to know it exists and maybe Know A Guy I can ask if I need to know something more specific.

Now in many cases the "Guy" is an LLM. I know enough to know when the LLM is bullshitting me and when I need to consult an actual human.

And Dropbox is just rsync and cron =)

Packaging and marketing matters.

And based on just on the list above that'd be at least 11 clicks (a lot more since you'd need to find the app in the software manager first before you can click install"

What the current regime has brought to light is that WAY too much of the American "checks and balances" are based on "trust me bro" and "a person with morals just wouldn't do that". Zero penalties for doing the wrong thing.

People can just go in front of a congressional hearing and ... lie. Just plain faced bold provable lies. Nobody can call them a liar because of The Code, but there's also no penalty for them for lying, zero.

And Gish Gallop.

Basically lie so much and so fast that the fact checkers can't keep up. It also helps when their voter base literally couldn't care less about "truth".

Paying to store my code, not paying for them to police who or what wrote it or assisted in writing it.

Privacy is the niche they know Google can never follow.

Apple sells hardware, Google is an ad company.

A Mac mini running an LLM is quiet

A PC with similar capabilities is going to sound like a jet taking off.

If I can't hack it with local access and an AI, a random hacker can't do it either.

I do lament the loss of control, but the increase in security will be objectively good for humanity as a whole.

Which AI companies still have your good will?

It's not just putting up colo space for local models.

It's more about the harness and API proxy you use. They need to be smart enough to know when a (self)hosted model is enough and when to forward to a SOTA model.

The SOTA model _can_ do everything, it's just expensive as fuck. But so is shoving a difficult task to a sub-par model that takes (relative) ages and comes back with the wrong result.

That's when you do the Enterprise deal.

I have first hand knowledge of household name companies with billion euro IP who use OpenAI and Athropic products quite liberally. They have WAY more lawyers than I do and their sole job is to keep the IP safe. They wouldn't sign a deal with even a slightest whiff of the IP being used to train anything.

And if it happens, the penalty for breach of contract would have so many zeroes it'd be enough to buy a country.

I tried this once too and instead bought a big-ass SRS bit and went through a concrete wall instead.

Rock solid and reliable networking =)

It swapped at the Opus/Fable 5 and Sol developments.

Opus 5 uses such convoluted language that it takes me multiple seconds to parse it. I specifically have to instruct it to use some kind of sane language (Caveman the AE STD English hack or something) to make it understandable.

Before that it used to be GPT that used 500 words when 5 would do. The ChatGPT version still does, but Codex is a lot better now.

Ruling against someone who's immune from repercussions does nothing. Especially when they've already done the thing.

Like demolish a whole wing of the White House without any review or control. Starting a wa... sorry. Special Military Operation unilaterally etc.

All illegal but whatchagonnado when they already did it?

Not a user, but from what I gather it does stuff like synchronize themes of ALL applications across the OS.

So when you change the "omarchy" theme to something, every single application installed uses the same theme. They even paid money to get proper theming support added to software so they could do this.

A bit beyond "dotfiles" IMO.

Yes, so does Economist, The Verge and many many other sites.

But I'm interested in maybe 3 articles a month. If I were to pay for them it's like paying for Netflix and watching a movie a month. Or subscribing to a dead-tree newspaper and only reading the comics.

I do get that micropayments are hard, but it's not a technical issue - it never has been. It's all about credit card costs and legalese.

And it also adds costs in every step. Someone needs to move thousands of unwanted books from high end stores to lower and lower end stores. Someone needs to store them in the proper environment etc.

I do get the _idea_ of preserving books, but... people don't care. I just threw out well over a thousand books from my grandparents house this spring.

There were ~6-10 "valuable" books there. Two because I personally knew someone who wanted old war-time books and a few 100+ year old bibles. And maybe two dozen books worth saving, mostly because they were from big-name authors or had stuff that nobody would print anymore (I have detailed instructions how to make laughing gas and how to build an underground chemical lab - hobby books in the 50s were ... interesting :D )

I literally couldn't give away the rest. And I tried. It was all just "interesting, but..." - no way to justify using the shelf space for books that, realistically, nobody will actually ever read again.

Copyright should be globally determined so that if I cant pay fair market value for a piece of content, it's deemed out of copyright protection and I can use whatever means to get it digitally.

Like if a game isn't available for sale anywhere in my region, I can get it without breaking any laws. A book is out of print and I can't pay money for it digitally -> free game.

(Why "fair market value?": So that skeezy publishers don't have an online shop with one physical copy of every book they own for $1Trillion just to fulfill the law)

So trashed, but with a bunch of extra steps then?

It _worked_ after the 11 day migration. Getting rid of unsafe blocks (Which were AFAIK partly a Zig inflicted thing) was just polish.

This has been my choice for a while too, especially when I found out that you can embed a full-ass webapp in a single executable.

Backend + web API + the actual web pages all contained in a single file. If you have sane defaults, deploying is literally just copying a single executable over.

Now compare the amount of renewables China is putting up?

They add coal power because it's super fast to build, but it's not a permanent solution.

AGENTS.md is specifically for agents. Every time you launch an agent in a repo, it's the agent's first time existing in the universe. This file lets it orient with the project.

Humans don't need it. Nor do they need to be reminded to launch subagents for simple tasks for example - which is another command that belongs in AGENTS.md.

It's a process. I just set up a filter on Fastmail that showed me all mail that was delivered to my @gmail.com address within the last two weeks.

Then I went through that list every now and then, unsubscribing or changing email addresses as they came up. At the moment it's mostly just mail from sites I've used Google OAuth for some reason and can't be bothered to change (or just plain can't - one site actually uses email as a primary key and are completely 100% unable to change it.)

I degoogled myself from Gmail/Calendar ... five years ago now? Zero issues with Fastmail that might've affected my life in any way.

I can kinda-sorta remember a few outages, but they were night time in my TZ so I found out about them after the fact.

Also I _love_ how they're not doing the Proton route and trying to become an everything-service. They do mail, calendar and contact sync, that's about it. No VPNs or cloud storage or AI offerings.

This is why I started adding a PROJECT.md file to all my projects and a hook for claude to read it.

It contains (Among other things) stuff like "this is a single user personal project, I'm the only user, this will never be open to the public internet" etc.

It kinda-sorta tones down the proclivity to worry about backwards compatibility and slight edge cases where if someone has edited some template and the new code doesn't support it.

One small snag, when playing vs CPU the angle and velocity are set to whatever the CPU used. So I need to memorise my previous settings if I want to fine-tune my aim.

This is what I've observed too in actual real life.

Massive refactoring operations that would've never been scheduled, because they would've needed a senior programmer allocated to it for at least a week can now be done to ~80% completion with a good prompt and an LLM in an afternoon.

then the senior dev only has to spend maybe a day or two in the final 20%, mostly confirming that nothing was missed.