Today I learnt about "bypass charging", a phone feature where it can run directly off the mains without using the battery at all. No battery charging, less heat, less battery wear. Nifty. www.androidauthority.com/phones-with-...

I'm considering using WebP more: for example, my websites would load faster and browsers have supported it since 2020: caniuse.com/webp That said, GitHub just added support in August 2025, so I suspect I'd still find use cases that don't work yet: github.blog/changelog/20...

Test code is total: we require it to always terminate or it's a failure! It also typically has 100% line and branch coverage. I feel way less nervous about refactoring tests, you can always just run them.

I certainly see the appeal of an LLM system with full context and tool use (OpenClaw), but the lethal trifecta puts me off deploying it. I do like the idea of an agent with a heartbeat though. A bunch of nice cron-style workflows can be built on top.

Date of birth pickers are a surprisingly nuanced UI problem. What do you pick as the initial value? How do you allow users to easily move by large amounts (years) as well as small amounts (the exact day)? Defaulting to today and even allowing future dates is funny though.

I've written a website that archives all my posts/tweets/toots/skeets across different microblogging platforms! It's fun to be able to see similar posts that I wrote at completely different times. It also lets me edit links that have bitrotted. miniblog.wilfred.me.uk

When I moved back to the UK, I ordered SIMs for both myself and my wife. We ended up with consecutive numbers! It's really convenient sometimes.

An AI benchmark website that tries to run comparable benchmarks regularly to discover when LLM performance is degrading: aistupidlevel.info

Futhark's design is deliberately trying to stay simple to keep long term design sustainable. The author also has an interesting discussion of relative versus absolute paths in import syntax, and how relative paths can make local analysis easier. futhark-lang.org/blog/2018-06...

The vast majority of libraries have a single maintainer. Even if you weight by downloads, roughly half of downloaded libraries have a single maintainer too. opensourcesecurity.io/2025/08-oss-...

Show, don't tell: effective advice for writing for both LLMs and people! www.haskellforall.com/2026/01/prom...

I've been working on a search tool for a website with 9K items and a dumb implementation works just fine. I'd normally gravitate to a library or tool like elasticsearch (ES). Adding custom filters etc is trivial on handrolled code. I wonder what scale requires actual search infra like ES.

I see people advocating for writing a FAQ, but I really struggle to write one from scratch. The best FAQs are written as a response to, well, frequent questions. How do you populate the initial FAQ?

Wonderful post from the lead LLVM maintainer reflecting on how the project is run, the design issues in LLVM today, and opportunities to make it better: www.npopov.com/2026/01/11/L...

My default assumption is that external libraries are better than what I (or an LLM) would write in a v1. The extra effort to publish a project generally signifies that the author has spent a good amount of time on the problem. I end up prompting LLMs to prefer external code.

LLMs are great at style transfer, and this works surprisingly well for syntactic transformations too. I rewrote a small website's EJS templates to 1500 LOC of TSX with a oneshot in Claude. It took ~12 minutes, but it just worked. (TSX seems nicer than EJS for static analysis.)

It seems really odd to me that git uses 'cached' terminology for staging. For example, `git add somefile.rs; git diff --cached`. Caching implies state that you don't mind losing. There's very little state I want to lose in version control!

Are there any nice cross-platform ways to trigger a notification from a terminal? I've been using pushover.net but you need to use a browser and it's noticeably slower than the native options. (Pushover is great for mobile notifications though.)

Terminad is a really interesting approach to rendering markdown in the CLI. It deliberately doesn't render links, so you can always see both the link text and the URL. In a browser you have hover to see URLs, but not necessarily in a terminal. github.com/Canop/termimad

This feature in @developer.chrome.com's DevTools is so helpful. Every time I get stuck on a calc() or other CSS functions, I solve the bug much quicker with it. Thanks to everyone who worked on this! 👌

Should code completion offer private methods/fields? If the user chooses a private method, you can offer a quickfix to change its visibility. It's clutter though. I frequently find myself wanting fields that I haven't exposed yet, and frustrated that the IDE hides them.

Blogged: Devlog 2: The Importance of Sandboxing Exploring PL opportunities when you have a first-class sandbox. www.garden-lang.org/blog/sandbox...

Is there a good canonical name for autofixes, where a static analysis tool can apply a change to resolve an issue? Clang calls them fix-its, LSP calls them quickfixes.

I've realised that AGENTS.md / CLAUDE.md solve two distinct problems: * It tells the LLM the conventions of the project, avoiding follow-up prompts. * It precomputes a summary of the project, saving a bunch of searches at the start of a session.

Igalia's post on self-hosting CI runners is a masterclass in build engineering. It shows dealing with weird API constraints, but still finding good designs. I also learnt that some CI systems use OCR on Apple's installers to automate! www.azabani.com/2025/12/18/s...

I've been experimenting with different pagination UIs. It's so common to have arrows, but I've realised they're redundant here. When you have the adjacent values as well as the final value, you don't need > and >> arrows too. Thoughts?

In many respects I feel an LLM benefits from a monorepo. I've vibe coded a bunch of helper CLIs recently and end up repeating the same preferences for all of them. I want all my CLIs to use clippy, parse arguments with clap, etc. They're distinct tools in distinct repos though.

I think you could build an interesting IDE with a tiny embedded LLM in addition to the usual tooling. Features like 'extract method' would be much nicer if an LLM could provide a name. Choosing a good name is virtually impossible from just a typed AST.

Delighted to see that TOML has released a new version! TOML is overall a great standard but I understand they had limited people with the power to cut a new release. github.com/toml-lang/to...

You can often estimate the age of a website based on how well it displays on mobile. All my sites end up with media queries in the CSS for narrow screens. It's so hard to design a single UI that scales from phone to desktop otherwise.

I find it odd that people recommend Docker for sandboxing agentic coding tools. Isn't it easier to just create a separate user account on the machine? It's an established security boundary, and viewing output is easy (just make the user's home directory world readable).

I'm never sure what to name my remotes in git. I tend to use 'mine' so I can add other forks later, but sometimes I use 'gh' or the traditional 'origin'. What do others use?

Caddy is really growing on me. It's nice having a web proxy that includes letsencrypt functionality without any additional configuration. I've tended to use nginx with a separate letsencrypt setup for projects before, and it's much more awkward.

Co-Authored-By: An old Stack Overflow answer, blindly accepting the compiler's suggestions, and a linter.

One subtle behaviour of Claude that wasn't obvious to me: whilst each conversation is transient, permissions persist across conversations. So if you've given permission to run e.g. 'cargo test' or even 'cargo run', you need to be sure that all future invocations are safe too.

I'm never sure how to convert words with hyphens into SCREAMING_SNAKE_CASE. For example, should built-in function be BUILT_IN_FUNCTION or BUILTIN_FUNCTION?

I've added an eval button to the website of my programming language! It's a been a lot of work to lock down appropriately. It still needs some UI polish but it's so nice to be able to try snippets immediately. www.garden-lang.org

I'm trying to decide what program I should show on the home page of my PL. Hello World is too simple to show much syntax. Fibonacci is OK but the reader may not know what print(fib(10)) should show. Maybe print(greet("World")) is better? It gives you a function definition at least.

Claude asked me a question today: was I looking for an Emacs plugin (because I was talking about elisp) or a Rust program (because I have configured Rust preferences)? I'm really impressed, it's rare to see LLMs ask follow-up questions. (I wanted Emacs in this case.)

Delighted to see that Typescript 7 is moving to conventional LSP for its IDE services! devblogs.microsoft.com/typescript/p...

I've been using "Expected Int, but got String" for my type error messages, but I've been wondering if I could do better. "Expected Int here, but this value has type String" or "This expression requires Int, but the value is String". Do you have a favourite?

IPv6 adoption is approaching half! www.google.com/intl/en/ipv6... shows 46% as of today.

Difftastic 0.67 is released! * Added support for protobuf, improved bash and YAML. * Updated 16 parsers to the latest upstream version (thanks Antonin Delpeuch!) * A ton of display polish for word highlighting, alignment, and binary files. github.com/Wilfred/diff...

If you created a large, successful OSS project, would you want to be BDFL or move on after a period? I see tradeoffs on both sides, although it's not a decision I've needed to make myself.

A cute systemd feature to get the system clock roughly correct: if the system time is before the compilation time of the systemd binary, it moves the clock forward to that time! lwn.net/Articles/714...

I've not seen this before: a captcha to distinguish adults from children! The in-app purchases were behind this screen. The primary user is a 3 year old who can only count to 20 😊

I've been learning some Common Lisp by writing some simple JSON munging programs. I'm sure it's not best practice — I should probably deserialise to CLOS automatically — but it's a nice way to get comfortable with the basics. It feels weird deliberately ignoring helpers though.

"After I linked difftastic, one of my friends immediately used difftastic to find a stealthy bug, five stars!" www.scannedinavian.com/tools-built-...

Shower thought: using a tool like cargo-semver, could you build a package registry where the uploader never chooses the version number? E.g. your last release was 5.2025-09-13 and you've just changed a type, so today's release is 6.2025-11-05.

One advantage I've come to appreciate about Dash/Zeal docsets: it's really nice having focused search. The text search is constrained to the languages I care about enough to download the docset, substantially increasing the relevance. In Google I'd need to specify the language.