maxk42

It's Saturday in China.

I love classless CSS for rapid hand prototyping. LLMs are making this increasingly less necessary, but it's still great for creating tiny, performant CSS pages and some of the smaller ones can even be inlined in fewer bytes than it takes to make a 2nd round trip to the server to check if the asset has been updated. This is one of my secret weapons for creating web apps that load and render so fast you can feel the difference!

The status quo is people googling shit, finding either legalese they can't understand (but maybe think that they do) or information that was legitimate ten years ago (and is dangerously inaccurate today). If I ask a question and get an answer that's been inaccurate for years I'm going to have as much or more false confidence as an AI bot that misunderstands the source. The difference is the AI bot gets better with time instead of worse. It may not be perfect but it's a step in the right direction: Let them cook.

Of course the src directory is in VCS - it's arguably the entire point of the VCS. The entire project - including documentation - should be in VCS. Documentation is not source and should not be treated like it. But it is crucial to generating the source and always has been treated as a second-class consideration. Now that we have AI agents that can handle all the fine details of keeping the documentation up to date, I believe. We seem to be in agreement that docs should be under revision control. So the only point we seem to disagree on is where they go. My original claim was that they don't belong under the src directory. I still maintain that position. You could say that the broader point of the article was to put them under VCS and I started out with that position so there is no disagreement there. I just think it's poorly thought-out to place them under the src directory and it's even more shortsighted to place the docs in a subdirectory called 'md'. To me, these seem like the thoughts of someone who is just discovering how to craft code with the assistance of AI: They're good initial instincts, and potentially just a handful of completed projects away from a better solution, which is what people have been doing since before AI: Keep the docs separate. Keep them under revision. Keep them up to date. But documentation is not source, and even if you specify the project in intricate detail you're not going to get the same deterministic result by piping them through the same AI. And that's always been the way documentation has worked: It specifies things in as much detail as we can, but in the end someone has to translate that into source code. And typically the person writing that source code has to make decisions or trade-offs that were never specified in the docs. Now with AI agents we're able to back-port every line of reasoning into the docs and create much better and more up-to-date documentation, and that's exactly how I build with AI. But we're not able to treat the documentation as if it's a deterministic system that will generate the same outputs each time. Therefore it is meaningfully different from source code and should be treated as such. (Although it should be kept very close to the source code in a structure that is intuitive and appropriate for the project.)

In any event: If we're talking past each other or simply not understanding one another then I thank you for a respectful conversation.

I like how you wrote "the argument here is that maybe ad-hoc prompts is not the best way to capture that because you lose the instructions" as if you didn't begin this thread by contradicting my statement "Docs in VCS, yes." and claiming I missed the entire point of the post. It seems you've now agreed with me and yet still somehow wish to argue. I think you've missed the point of my post. You should reflect on that.

No, this is not a source file to be used to generate code. It's a spec to be used to build the code to - same as any other document. Aside from the fact that docs can be in any format - not merely .md - they do not belong in `src/`. Why? Well for one, many of my projects have multiple `src/` directories because they involve multiple binaries. But the documentation used to generate that source is in the project, not the `src/` directory of one of its binaries. I'm building a video streaming site that uses one server for the SFU, one server for the web interface, one server for the websocket connections, as well as STUN and TURN servers. None of the project works unless all these separate binaries are working together and communicating. Which `src/` directory should I place these markdown files in? What about the `.sql` files I use to spec out the database schema before passing it off to the LLM? What about the mermaid files, code snippets, `.txt` files, and data files used as part of the documentation project? Should they all go in `sfu/src/md/`? No. The project is one whole with multiple `src/` directories and they all require their own documentation. Beyond that they need access to each other's documentation so that when the agent is working on one part of the system it can still be aware of how the other pieces operate. Documentation should absolutely be under version control, but placing it in the `src/` directory is plainly the wrong call. And there is more documentation than just markdown. This was a shower thought that wasn't completely fleshed-out (and the article admits it wasn't) and never should've made it to the front page of HN.

Docs in VCS, yes. Docs in /src, no.

I'm building InterviewMe ( https://interviewme.now/ ). It lets employers interview you via AI and is intended as a better way for people to get deep answers about your work history and experience than they could by looking at a simple online resume or LinkedIn. You upload your resume, answer some questions about your background, optionally upload any other documents you have (cover letters, awards, articles, etc) and it ingests all of that and creates a conversational agent people can use to ask questions about your professional experience. Check out my profile to get an idea of how it works: https://interviewme.now/@max and then get your own: It's free!

Westinghouse makes a line of "non-smart" TVs. A few obscure Chinese brands do also.

Sometimes the reasons people give are not the real reasons.

Shit. This was my last near-stock Android choice with great battery life. What's the alternative now? I hate Pixels and don't really want to give Google my info anyway. Is there something decent left?

Openrouter definitely supports vision models. Why would you have to give up vision?

My experience has been quite the opposite. I have done far better than my peers and have less debt. I wonder which country you're in?

Even if they are... road to hell and all that

After recently figuring out how to get CUDA running on Fedora I'm inclined to agree.

Seriously, GPUs are a mess and keeping LLMs from helping us use them properly is practically a crime.

A year ago I was using claude and found it unexpectedly blowing-away already-settled code. I found this frustrating and just wanted to drop a "save point" where I could compare the before and after versions of certain files. So I built a little utility called "bup" that tarballs the whole code tree and gives it a unique name. It ignores prior backups and has options to ignore certain subdirectories, VCS directories, artifact directories, or secrets files. And you can add a single switch (-s) to create a shell script that automatically backs-up the current project. Yes I can use git for this, but this is honestly way easier, especially if I want to compare/edit different versions. No names. No files. No trees. Just "bup .". And it's 100% bash. https://github.com/maxk42/bup

I'm probably the only one who'll ever use it but I like it.

I managed an adult platform with 200 employees and $75 million in revenue and a dedicated Risk Analysis team. I think I know a thing or two about this, but thanks for the input.

The timeline expired 17 years ago. They've been granted numerous extensions. This deadline is neither new nor "suddenly tight". They were granted another extension last year. They're simply informing Liberty there will be no further extensions. If you have to wonder who's wrong: The utility that's been granting extensions for 17 years or the utility that's been missing their deadline for 17 years I think you'll have your answer if you think about it.

Tahoe is not a town. It is a region containing many towns. Most of it will continue to have power (mostly through NV Energy). This is a single utility that has failed to plan for this planned transition for 17 years. Invoking the data center bogeyman for political sympathy is the only thing going on here. Liberty has failed repeatedly to address this problem for nearly two decades and it's beyond time they were cut loose. I have sympathy for the people who have to suffer from their failure but this is not NV Energy's fault nor is it related to data centers at all: NV energy has been rapidly expanding capacity faster than growth in consumption and has 19 active geothermal plants in the area with more on the way as well as wind and solar.

Chargebacks. "Oh, honey, I don't know how that got there. A hacker must've stolen my card! I'll call the bank immediately!" Worked in the adult industry and traditional e-commerce. It's a perennial problem.

And they're Australian, not American.

I guess you missed the part where they turned around and became profitable.

GameStop has zero debt and billions in cash on its books. It is not a stretch for them to be making this offer. It really does make sense here for both sides.

The US also has postal codes that map to individual addresses. People just don't know them.

For my use case I need MSL to support fp64. Until that happens I don't care what hardware changes they make: I'm not going to be filling racks with M5s and they're not producing something I can use to even tinker with AI with in my spare time. Apple has lost the AI war before it even got started IMO.

Did nobody notice that this is a spam blog designed to sell NAD+ supplements?

My concern is Cloudflare will implement this plan in a way that makes it very profitable for large players and absolutely kills any new entrants to the market in the future.

...which explains the link to China. NK natives probably do not typically have access to computers or the open internet, but the children of certain elites are educated in China. There may even be a collaborative effort between the two states.

Hey - not everything is bullshit on "Who's hiring" threads. I once got eight rounds of interviews with a company before being ghosted by someone off of an HN thread.

Sorry, but how would you ever prove a job ad is fake?

"Were you going to hire someone for this role?" "Yes." "Case dismissed."