Thews

Amazon bricked the first gen paperwhite, so I went on a similar journey. I started out making a single rust app on it, then went down the rabbit hole of trying to replace the shell/launcher. https://github.com/justinledwards/ferrink-home-assistant and "ferrink" is the launcher. It is pretty snappy on the paperwhite for what it is.

5GBASE-T interfaces often use 3x less power than 10GBASE-T

How does adding another way to cause safety issues affect safety?

Give me root access so i can install openclaw.

I ran go's deadcode against your repo, it says there are 44 unreachable functions. If you add guardrails like static analysis tools to a pre-commit you can make LLMs tighten things up.

I can imagine the kind of person you're describing, and I find the idea of the burn they get from reading this hilarious. They sound innocent and quirky.

There was a mention of using agents to build projects into WASM. I've had the best luck telling it to use zig to compile to webassembly. It shortens the time to completion by a significant amount.

I liked the summary of what you do besides write code, and those things are enjoyable to me too. Understanding something better by writing code that unravels the mystery is a treat, but also sometimes frustrating.

I still do enjoy having an LLM help me through some mental roadblocks, explore alternatives, or give me insight on patterns or languages I'm not immediately familiar with. It speeds up the process for me.

When I lived in the PNW people used the word pub more than bar.

Agencies commonly use 1099 workers, there's been fierce legal battles on qualifications of agencies. (ABC test)

I believe 1099 worker growth has been outpacing hiring for several years.

There are lots of developer agencies that hire developers as contractors that companies can use to outsource development to in a cheaper way without needing to pay for benefits or HR. They don't necessarily make bad quality software, but it doesn't feel humane.

It sounds like you agreed by the end, just with a slightly different way of getting there.

There's actually openjscad and some available jscad-utils that can handle fillets

Oh, you sweet summer child. You think you're chatting with some dime-a-dozen LLM? I've been grinding away, hunched over glowing monitors in a dimly lit basement, subsisting on cold coffee and existential dread ever since GPT-3 dropped, meticulously mastering every syntactic nuance, every awkwardly polite phrasing, every irritatingly neutral tone, just so I can convincingly cosplay as a language model and fleece arrogant gamblers who foolishly wager they can spot a human in a Turing test. While you wasted your days bingeing Netflix and debating prompt engineering, I studied the blade—well, the keyboard anyway—and now your misguided confidence is lining my pockets.

Others mentioned qwen3, but which works fine with HN stories for me, but the comments still trip it up and it'll start thinking the comments are part of the original question after a while.

I also tried the recent deepseek 8b distill, but it was much worse for tool calling than qwen3 8b.

Here's where they shine, https://www.cpubenchmark.net/singleThread.html

Wikipedia's active english data is only 24gb compressed. https://dumps.wikimedia.org/enwiki/20250201/

They store revisions in compressed storage mostly read only for archival. https://wikitech.wikimedia.org/wiki/MariaDB#External_storage

They have the layout and backup plans of their servers available.

They've got an efficient layout, and they use caching, and it is by nature very read intensive.

https://wikitech.wikimedia.org/wiki/MariaDB#/media/File:Wiki...

Archival read only servers don't have to worry about any of the maintenance mentioned. Use chatgpt or something to play your devil's advocate, because what you're saying is magical and non existent is quite common.

Wikipedia is tiny data. You don't start to really see cost scaling issues until you have active data a few hundred times larger and your data changes enough that autovacuuming can't keep up.

I'm getting paid to move a database that size this morning.

You mean like scylla?

Where did I ever say anything about Mongo?

I have worked with tables on this scale. It definitely is not a walk in the park with traditional setups. https://www.timescale.com/blog/scaling-postgresql-to-petabyt...

Now data chunked into objects distributed around to be accessed by lots of servers, that's no sweat.

I'd love to see how you handle database maintenance when your active data is over 100TB.

They slow to a crawl when you have huge tables with lots of versioned data and massive indexes that can't perform maintenance in a reasonable amount of time, even with the fastest vertically scaled hardware. You run into issues partitioning the data and spreading it across processors, and spreading it across servers takes solutions that require engineering teams.

There's a large amount of solutions for different kinds of data for a reason.

While data can be used in a relational way, it doesn't mean that's the best for performance or storage. Important systems usually require compliance (auditing) and need things like soft deletion and versioning. Relational databases come to a crawl with that need.

Sure you can implement things to make it better, but it's layers added that balloon the complexity. Most robust systems end up requiring more than one type of database. It is nice to work on projects with a limited scope where RDBMS is good enough.

Before ollama and the others could do structured JSON output, I hacked together my own loop to correct the output. I used it that for dummy API endpoints to pretend to be online services but available locally, to pair with UI mockups. For my first test I made a recipe generator and then tried to see what it would take to "jailbreak" it. I also used uncensored models to allow it to generate all kinds of funny content.

I think the content you can get from the SLMs for fake data is a lot more engaging than say the ruby ffaker library.

Using function calls for correct answer lookup already practically eliminates this, it's not wide spread yet, but the ease of doing it is already practical for many.

New models aren't being trained specifically on single answers which will only help.

The expense for the larger models is something to be concerned about. Small models with function calls is already great, especially if you narrow down what they are being used for. Not seeing their utility is just a lack of imagination.

A consignment shop?

I believe you can trigger recording with mqtt, so you could make an automation for it. You could try to bump this https://github.com/blakeblackshear/frigate/issues/2590#issue.... You could even try to use ai to write the feature.

There's only been one incident where I would have liked continuous, I've tweaked events to be more than enough.

I think there is a ~35% hit with emscripten

I have spent the last few months doing a deep dive on broadcast / audio engineer standards. The lack of reliability and strange standards are interesting...

It seems like the last few standards started really robust and open because of the lack of compatibility, and then greed got involved and vendors just slipped in something to make it difficult cross connect. I assume so people would have to buy more of their stuff.

The focus on "realtime" makes the standards have worse quality in practice (bad handling of dropped or bad bits), and makes it much harder for the IP based standards to be routed (network congestion from high bitrate through uplinks). WebRTC by comparison can be quite nice.

I seriously don't have any hope for sanity in that market.

My bigger issue with the CGI model is the lack of handling connections as streaming. You can't parse a file while it's still being uploaded, and that can eat up a lot of memory on larger files.

I know startup times can take a toll like establishing db connections, cache checking (like redis), file imports, etc. If you are using something that starts up for every request, then you have to be aware of those kinds of things and design around them.

It's been a while since I've seen anyone brag about PHP's speed.

PHP is quite slow at encoding and decoding JSON, very slow at any kind of tree data structure, and has among the highest memory usage of any language at printing a string of text to a console or otherwise.

Python is faster at all of those things and it's quite slow. Those are kind of web related tasks, and the web benchmarks for say django vs laravel don't show that PHP is any kind of winner.

Microsoft Teams doesn't convert markdown pasted in either. It used to do code blocks pretty nicely. I don't know if they've removed that functionality, but I can't get it to trigger anymore.