Notes

I think something I’m going to have to start asking in any future jobs is, “how locked down are your company laptops?”

In my first role, I was quite lucky to be given essentially free reign of the programmes I could install and the software I could use to day-to-day which meant, over the course of two years, I really managed to create a system which worked for me and genuinely felt more effective because of it.

I naively assumed this would be the case in most places and was surprised to find that, in my next company, it couldn’t be more different. The only apps I was allowed to install came from the company’s own, internal registry of “verified” software. Everything needs to be vetted by a security team to ensure it’s safe to use alongside company data, and it has to have a business justification for why it’s needed over any of the existing, approved programs.

Suddenly I can’t use Obsidian to take work notes, I can’t use Todoist to organise my tasks and stay on track. We’re not even allowed to install Spotify on these machines.

I get why this policy is in place, but that doesn’t stop me from finding it a bit disruptive and wishing I could just do everything how I want it. Now I have two entirely different systems for notes and tasks when comparing my work and personal life, and they’re entirely disconnected from each other. Great stuff! 👍

Considering the notion of starting my own business and working for myself, even if it’s just so I can use the software I want to use.

Just watched Maggie Appleton demo how she uses Tana to organise all her stuff and felt a weird mix of awe and FOMO? I don’t collect anywhere near enough data in my day-to-day life to need much of a system to organise it but I kinda wish I did? I love the idea of a Second Brain, all connected together to help me work better and organise my thoughts and interests. On the other hand, maybe it’s just got a nice UI that I like and I’m trying to trick myself into feeling productive by setting up some kind of new system.

I’ve put all my website content into its own private repo and am loading it as a Git submodule now, thanks to help from some helpful guides!

The plan is to start using Obsidian to create and edit content, then push it to GitHub with the Obsidian Git plugin and hopefully everything just updates. I’m not quite sure whether changes to the submodule will trigger the “parent” repo’s GitHub Action to redeploy the site, which is partly why I’m writing this note; I want to check what happens and whether I need to manually redeploy something. If so, I need to find a way to have this trigger ✨ automagically ✨.

Edit: The repo seems tied to a specific version of the submodule, and didn’t automatically pull the latest when I made changes. I think the easiest thing would be to setup some kind of cron job in GitHub Actions that runs the git submodule update command every 30 minutes or so (it could be longer; I don’t publish very frequently) and redeploys if there are any changes. Hopefully that doesn’t eat through my free tier usage limits.

I’ve been trying to get prefetching working in an app I’m building at the moment, but Firefox has been consistently ignoring my wishes and only fetching on click.

I looked in my about:config and, lo and behold, prefetching behaviour had been disabled. I didn’t remember doing that? I turned it back on but it still didn’t work. I checked about:config again, and it was disabled again. What??

To keep it short, uBlock Origin and Decentraleyes—I’m still not sure if I need both—each had a setting in them to disable prefetching behaviour. Apparently it’s harmful, privacy-wise, because it allows the browser to connect to resources that would’ve otherwise been blocked.

What now, then? I probably want to keep prefetching disabled because I think I value my privacy that much, which means there’s no point me implementing it. I’m going to look elsewhere for performance gains.

I’m building myself a bookmarks/read-it-later website to replace Instapaper and one requirements is saving links to it via a browser extension. (I also need to write an iOS app but that’s another kettle of fish).

I didn’t think it would be too bad: it’s just JavaScript right? The problem lies in that there doesn’t seem to be any form of intellisense for the WebExtensions API, so I have to spend half my time flicking back and forth between VSCode and the MDN docs to figure out what methods are available. Additionally, the development loop of manually loading your unpacked extension and having to monitor the browser console (Cmd-Shift-J) for logs seems quite slow?

Maybe I just haven’t been at it long enough but I really hope I find a nicer way to do this.