Going through the treesitter main branch switch nightmare after updating neovim to 0.12. Fu....

Agreed. I was floored by it when I managed to run it on my rooted Kindle.

The Age of Faith, by Will Durant A sweeping history of the Middle Ages, from the fall of the Roman Empire to the death of Dante Alighieri

Finished reading The Age of Faith.

Some notes here: www.siddharthagolu.com/posts/readin...

Added Wander to my website at www.siddharthagolu.com/wander/

Such a neat way to browse the small web

I used to love reading engineering blog of Facebook. Now I just can't bring myself to care about their endless optimisation of the doomed algorithm.

Their latest blog post is called "Enhancing social discovery on Facebook Reels".

Man I couldn't care less.

GitHub - nohajc/anylinuxfs: macOS: mount any linux-supported filesystem read/write using NFS and a microVM macOS: mount any linux-supported filesystem read/write using NFS and a microVM - nohajc/anylinuxfs

Anylinuxfs looks like a sane alternative, if you're primarily using the drive on Linux and need to use it occasionally on Mac.

github.com/nohajc/anyli...

In search of the best filesystem for cross-platform compatibility (or at least b/w MacOS and Linux), and turns out there's no "best" solution, even in 2026.

exFAT is what gets usually recommended, but it trades off on performance on both Linux and Mac and recovery/repairs are uncertain.

Personalized captchas FTW

This looks great!

GitHub - notune/android_transcribe_app: Offline voice input keyboard & live subtitle tool for Android Offline voice input keyboard & live subtitle tool for Android - notune/android_transcribe_app

Trying out Android Transcribe app: github.com/notune/andro...

Uses a quantized Parakeet V3, but runs surprisingly well on my ancient 6 years old phone.

This is perfect

Writing as fitness exercise This is the first-part on writing as necessary friction.Jean-Baptiste Oudry (from Getty Open Content)It makes sense that many writers, when faced with the existential dread of generative AI, return to...

I've seen lots of people getting disheartened by the impact of AI tools on writing, including myself, and this comes as a good reminder to shift your focus on the first principles - the act of writing itself.

vivqu.com/blog/2026/02...

Love this.

I'd also be interested to know the stats behind "Pay what you like", if you're comfortable sharing. Clicks -> conversions etc.

Micropayments is such an unsolved problem and it'd be valuable to get a data point like yours to put things in perspective.

And thank you for your writing!

Case in point: for the longest time, Gajim on MacOS was broken in multiple different ways.

Just opened the wiki again today, and voila, they have nightly binaries available for Mac now: gajim.org/downloads/sn...

Discord going ape shit with their policies was the best thing to happen to XMPP.

I'm noticing so many fresh voices in the community chat, trying to adopt and hack around the protocol for different use-cases.

Bluesky settings for turning off auto play videos

Twitter settings for turning off auto play videos

I got a taste of how addictive it could be if I just kept the autoplay videos setting turned on.

I used to use Twitter sparingly to get the day's highlights for AI world. And today I realized how dramatically my feed had shifted from text-only to videos-only.

No wonder TikTok/Reels are addictive.

A reddit comment on the thread "why do people consider Wes Anderson as pretentious" which says: "The problem with Wes Anderson isn't that he has a very specific style (you wouldn't go to a Monet retrospective and complain about how many times he painted water lilies), it's that we don't have a hundred Wes Andersons all pursuing their own styles. As it is, it's like crawling through a desert and finding a fountain that only dispenses your least favorite soda. The problem isn't the soda, the problem is the lack of choice and supply."

Well put. I wish there were more Wes Andersons out there.

Damn, congratulations!

Ah, so this is the origin story of this wallpaper. I used to run a script which would change the background depending on the time of the day; remember loving it so much.

A screenshot of a work-in-progress web UI to generate speech using mlx-audio server

Building a web UI which connects to a locally running mlx-audio server and allows you to download the generated speech.

Tested this for a few days, and sad to say, another one of those overhyped launches. It would skip some words consistently and would break the flow for long sentences.

I shifted to using Qwen3-TTS instead and the quality of the voices are much better.

The same realization as for everyone else - I noticed my abilities declining with more usage. While I do get to execute much faster, this tradeoff is currently not worth it for me.

I've decided to slow down and start treating it as a great supplement to my thinking, not a replacement.

A comment on HN (Part I) in response to Mitchell Hashimoto's post on his AI adoption journey. It says: "LLMs are not for me. My position is that the advantage we humans have over the rest of the natural world, is our minds. Our ability to think, create and express ideas is what separates us from the rest of the animal kingdom. Once we give that over to "thinking" machines, we weaken ourselves, both individually and as a species. That said, I've given it a go. I used zed, which I think is a pretty great tool. I bought a pro subscription and used the built in agent with Claude Sonnet 4.x and Opus. I'm a Rails developer in my day job, and, like MitchellH and many others, found out fairly quickly that tasks for the LLM need to be quite specific and discrete. The agent is great a renames and minor refactors, but my preferred use of the agent was to get it to write RSpec tests once I'd written something like a controller or service object. And generally, the LLM agent does a pretty great job of this. But here's the rub: I found that I was losing the ability to write rspec. I went to do it manually and found myself trying to remember API calls and approaches required to write some specs. The feeling of skill leaving me was quite sobering and marked my abandonment of LLMs and Zed, and my return to neovim, agent- free."

A comment on HN (Part II) in response to Mitchell Hashimoto's post on his AI adoption journey. It says: "The thing is, this is a common experience generally. If you don't use it, you lose it. It applies to all things: fitness, language (natural or otherwise), skills of all kinds. Why should it not apply to thinking itself. Now you may write me and my experience off as that of a lesser mind, and that you won't have such a problem. You've been doing it so long that it's "hard-wired in" by now. Perhaps. It's in our nature to take the path of least resistance, to seek ease and convenience at every turn. We've certainly given away our privacy and anonymity so that we can pay for things with our phones and send email for "free". LLMs are the ultimate convenience. A peer or slave mind that we can use to do our thinking and our work for us. Some believe that the LLM represents a local maxima, that the approach can't get much better. I dunno, but as Al improves, we will hand over more and more thinking and work to it. To do otherwise would be to go against our very nature and every other choice we've made so far. But it's not for me. I'm no MitchellH, and I'm probably better off performing the mundane activities of my work, as well as the creative ones, so as to preserve my hard-won knowledge and skills. YMMV I'll leave off with the quote that resonates the most with me as I contemplate AI:- "I say your civilization, because as soon as we started thinking for you, it really became our civilization, which is, of course, what this is all about." -- Agent Smith "The Matrix""

I'm on the fence about using Agents, and while Mitchell's post was quite grounded and illuminating, I found myself agreeing more with this POV.

TLDR: unless you're Mitchell Hashimoto and can work on hard technical problems by yourself, keep the ability to think to yourself.

Give Nightly a try.

Superior than the stable version in features/performance and I've never had any problems with stability.

I use Firefox Nightly on my phone and it works rather seamlessly.

I use Zen on the desktop, signed in with my Mozilla account. When I browse FreshRSS on my phone, if I come across an interesting link/blog post, I share the tab with the browser instance and then read on desktop later.

Lesson learnt - always use python 3.11 - nothing else, nothing more

How is it that even after such a monumental change with `uv`, the python ecosystem is still giving me headaches when trying to run machine learning models?!

I've done all sorts of sorcery of using `venv`, `uv tool ...` and yet still facing errors.

Matt Levine's new post on xAI and SpaceX merger

Matt Levine is always a pleasure to read

Blog post: ace-step.github.io/ace-step-v1....

ComfyUI workflow: blog.comfy.org/p/ace-step-1...