This reminds me of something I did in one of the previous companies where I worked.

Like anyone else, when I joined the company, I had various questions: how to access certain systems, how to handle permissions, how to debug specific services, etc.

I compiled all these questions and answers as notes in a Git repository that my teammates could access. I wrote the notes using QOwnNotes, utilizing its Git integration. So, when someone had a question for which I already had the answer, I could simply share my notes, or create/update a node and share it.

The names of the notes were straightforward and easy to follow, such as:

- aws.md

- azure.md

- kubernetes.md

- staging.md

- production.md

- useful-commands.md (jq, sed, base64, etc)

My teammates used this resource frequently. As I was preparing to leave the company, I suggested them to fork my notes repository. I later heard that they continued to use it for many months afterward.

Very good article.

After read about the poor scenario where the Linux accessibility tools is today (https://fireborn.mataroa.blog/blog/i-want-to-love-linux-it-d...), I was wondering: if maybe the developers start to use these accessibility tools to improve their speed reading (and productivity as well), this could also helps to prioritize the accessibility features and bug fixes in Gnome, KDE, Qt, etc.

Sugarcane, I believe.

Yes sugar cane is much cheaper and efficient. But you need very warm climate to grow it. Brazil has that and uses sugar cane for the most part. The USA does not so it uses what grows well in its interior , which is corn.

Vertical tabs is really great.

But there were already extensions for that, which is a better approach than Firefox developing a feature only a small subset of people will use1.

For years every Firefox discussion I saw would have some person saying "why are they wasting their time on X? I won't use it til they have vertical tabs" with scores of upvotes.

This perfectly encapsulates the Mozilla position. Every possible move makes people complain.

Hard disagree. The extensions were terrible, at least for what I want, and probably can't integrate with the UI in the same way.

I totally support that. I know that this can create bad habits, but not everyone wants to become a great piano player; some of us just want to have some fun playing.

Same reason. I can add to this list Readme.com and Notion.

I have the exactly same issue using it with Aider.

I know that this can sound counterintuitive, but the best strategy to keep the phone away from me is to be on my desktop computer.

Different from when I'm on my smartphone, I do not feel any anxiety to check social networks using my computer. So I can focus more on learning some stuff, coding, organizing my personal data, checking my appointments, checking the tech news, or even playing some games (to have some fun).

This is what I've been doing lately too. I've taken all apps off my phone that I spent significant amounts of time on. My phone is now something I use for a few minutes, and then stop using, with the exception of navigational apps and music. My goal was to return to pre-smartphone days where having a laptop or a desktop computer puts just enough friction in so that I don't habitually whip out my device and start scrolling or surfing when I feel bored.

Another part of this for me is not going to sites that have infinite scroll. This means that even on my laptop, I will not go to sites where I cannot finish consuming whatever content was there.

For sites like HN where there is a rotating front page, I have an RSS feed of the front page that refreshes infrequently so I can sample what was there without always needing to return to the front page to look for new content. Currently I have this set to 1 hour. This has been a decent mix between missing interesting content and having a feed that shows me way more articles than I can consume. My RSS reader is self-hosted at home, which means when I leave the house I am not tempted to use my phone to read that RSS content.

I have the same issue using it with Aider.

The model is good to solve problems, but is very difficult to control the unnecessary changes that the model does in the rest of the code. Also it adds a lot of unnecessary comments, even when I explicitly say to not add.

For now Deepseek R1 and V3 it's working better to me, producing more predictable results and capturing better my intentions (not tried Claude yet).

Well, I found these candidates long before ChatGPT