pbowyer

Seems fair to me. The VC-backed SQLite services have been shouting about their services as if they came up with the idea, and this only balances that out.

He's right that histories start too late, and I'd like to see more posts like this that trace the history back further than the last 5 years. Like for databases, a lot can go back 20+ years and it's good to read how the idea came about (independently from many people), how it developed over time and why.

Also "Blitz Street" with Tony Robinson, where they build a street of typical 1940s houses and slowly increase the WW2 bomb size to see the damage the would do.

I'm just about to build my own external temperature and humidity sensor, using a Sensiron SHT45 sensor with integrated PTFE filter.

I've had Tuya and Aqara sensors outside, undercover and out of direct rain for the past 3 years, and every one lasts about a year before the humidity sensor gets stuck at the maximum reading. Even bringing it indoors doesn't get it working again. I'm hoping the built-in heater on the SHT45 will help when we're sat in sea mist or sea fog and keep the sensor working for longer.

What would be the correct way to do benchmarking/profiling here? I ask as I've started to run some for Vinyl cache to see what performance issues I can find, and I've quickly learned just how hard it is to get good, repeatable benchmarks that isolate the right thing.

I think we're both looking for the same improvements and I haven't found them in a writing tool yet: https://news.ycombinator.com/item?id=49331036

I like seeing what people create, and you've created something with some interesting features, particularly around the path map and keeping all drafts. I'm forever duplicating a paragraph and rewording it 6 or 7 times, then deleting the rest - and the next day think "no try #2 was better than the one I kept".

Ignore the haters, they're a HN rite of passage. Good luck with your writing!

All of this, but underlying it I've come to the conclusion that Markdown isn't a good format for a writing environment beyond just simple prose. There's too much left up to the flavour of it and it's got weaknesses.

I've been experimenting with using Markdown (in Typora) in place of Word, but it (and others I've looked at) have zero support for image positioning or sizing, or a way to have different (pre-defined) styles on different tables.

Which is a shame because I like keeping my documents in plain text, in version control and efficiently read by my LLMs. LyX (https://www.lyx.org/) for LaTeX is the closest to what I'm thinking of that I know of, but maybe something exists for Asciidoc or Typst.

Friendly reminder that GPT-5.6 in Codex cannot spawn Luna subagents, only Terra [1]

To fix this you currently need to make your own copy of the bundled model catalog [2] and opt Luna into MultiAgent V2.

1. https://github.com/openai/codex/issues/32031

2. https://github.com/openai/codex/issues/32031#issuecomment-51...

Can you say more? This sounds like it could be more useful than trying to build a knowledge base (say qmd) and then get the models to search it for the right context each time.

> I feel like this benchmark reiterates my disbelief that anyone uses the latest Anthropic models for any productive work. They seem to be the best at burning tokens and spawning unnecessary subagents even for well-defined and tightly scoped tasks.

I keep Claude around for some specific tasks:

- Linked up to Figma MCP to implement front-end stuff

- Data analysis, in the "Connect AI to a data source and ask questions" way. I've tried both Opus 4.8 high and GPT 5.5 high for this and Opus is stronger because it gets the intent in the question better

I used to keep it around for planning too, but the 4.8 plans have had more holes than swiss cheese.

> I'm only using Opus at all because GPT-5.5 in the subscriptions only has a small (400k, but 258k effective) context window.

Do you find that makes a difference in your work? I've been using 5.5 high/xhigh to optimize and benchmark a C codebase, and just reading the initial code virtually fills the first context window. A session will auto-compact 5-15 times, but it seems to do okay in spite of that because the task is mainly focused on the latest window each time.

I think for programming the strength of GPT over Opus is winning here over the context window.

The one positive of this is it seems to have livened up the open source project (Vinyl Cache). Features that haven't been added for years are now being worked on, whether they step on the toes of the commercial offering or not.

Jira is the one product I feel needs to be AI native.

AI native in the sense that it papers over the pain points.

New JIRA admin? AI will set it up to do what you want (after all, Atlassian has a great training set as they can see which Cloud installs work well)

Need to set up a workflow? Bam, AI to do that.

Need to onboard a user or manage permissions? Again, have a chatbot to do it (as a time-to-time Jira standin Admin, changing permissions always needs doing in 2+ places and devolves into a "Can you see this yet?" round of questions)

Previously covered at https://news.ycombinator.com/item?id=48053623

What's led to the higher prices than Germany? Usually substantially lower earnings would mean lower prices, even if not substantially lower (look at the UK, higher prices than much (all?) of Europe, average earnings slightly less).

> Extremely debatable. They still have never fully implemented health checks and auto healing.

Agree.

Plus there's the monitoring of the host that is always overlooked in articles. I've ended up chucking Monit on there to monitor disk usage et al, and also used it for monitoring compose too and restarting containers.

And then there's Healthchecks.io, and external uptime monitoring... the list goes on. Properly monitoring systems, even single server systems, is not simple.

Only at the "hmm that seems an interesting idea" level.

Thanks for the links, going to have a read and see if I can apply any to my work.

If you can type code fast and accurately, it sounds a great process to use. You're using LLMs for the bit where they bring great value, and yourself as a higher quality coding agent :)

Developer of 20+ years here, can't give you an accurate multiplier but I am faster.

Because spotting holes in specs has never been one of my strengths. And working without technical colleagues much of the time, it's a boon to be able to "rubber-duck" my ideas with something that is at least more intelligent than plastic.

Grabbing multipliers from thin air, the coding bit may only be 2x faster with a poorer-quality outcome, but working out what's needed is a good 5x faster.

And yes, I'm using the same adversarial AI MO as @wood_spirit, combined with Matt Pocock's excellent /grill-me and /grill-with-docs skills [1] and Plannotator [2] to review the plans.

1. https://github.com/mattpocock/skills

2. https://github.com/backnotprop/plannotator

> Hermes is one of these OpenClaw clones

So that's what it is. Reading its README I thought it was another harness like Pi [1], but with built-in memory so it remembers what it learns, and gets more capable the longer it runs.

Like Letta [2], Dirac [3][4] and the other "more experimental harnesses that look interesting but I haven't had time to try out".

1. https://pi.dev/

2. https://www.letta.com/

3. https://dirac.run/

4. https://news.ycombinator.com/item?id=47920787

What's the privacy/data security like? I can't find that on that page.

Edit: found it.

> We may use your Content to operate, maintain, improve, and develop the Services, to comply with legal obligations, to enforce our policies, and to ensure security. You may opt out of allowing your Content to be used for model improvement and research purposes by contacting us at [email protected]. We will honor your choice in accordance with applicable law.

Section 3 of https://www.kimi.com/user/agreement/modelUse?version=v2

> Stripe APIs being simple and easy is a meme from the 2010s. It isn't anymore.

I'm working with Stripe subscriptions at the moment for a charity taking donations via their website. The subtle differences between subscriptions done through Stripe checkout and subscriptions set up yourself using Stripe elements are by turn infuriating and frustrating.

The documentation is geared towards people using checkout. Stripe's own AI help could find us a bit of information which going through the documentation didn't give us, and it even struggled to find the reference in the docs for it.

One product, two different ways to use it, and slightly diverging feature sets between the two. Argh!

The Peter Principle: https://en.wikipedia.org/wiki/Peter_principle

Let us know if it does, because we all want it to work :)

Dealing with Google is a nightmare. I'm one of the volunteer sysadmins for https://forum.buildhub.org.uk/, a DIY and self-build forum. For 10 years it ranked very well on Google, particularly in the UK, and then on 28 December 2025 it disappeared from Google's index.

Nothing has helped, the Google forums are tumbleweed and there's no one to reach out to for what could be an algorithm change or something gone wrong. I'm a paying Workspace customer and it's made me think I need a backup plan in case I'm ever suspended. Reports like this don't encourage.

This nono? https://github.com/always-further/nono

> Just remember to disable CC’s auto-updater if that’s what you’re using.

Why?

Can anyone enlighten me how having a coding harness when for most customers you say "we won't train on your code" helps you do RL? What's the data that they rely on? Is it the prompts and their responses?

If you were creating a new programming language in 2026, which DateTime/Temporal library would you copy and why?