I want to know About ActivityWatch's Future Development – Any Exciting Updates Coming?

Hey everyone,

I have been using ActivityWatch on & off for a while now & I have gotta say, it’s a super useful tool when juggling multiple projects or just trying to keep track of how much time vanishes into “research”.

I want to know if there are any features or big updates planned in future?? Maybe improvements to syncing, dashboard customization or even integrations with other productivity tools? I poked around the GitHub repo and saw some activity but I want to hear directly from devs or regular contributors if there is a roadmap or something you all are working toward.

I have been balancing this tool with some study time while prepping for my CISSP Course & it made me think how cool it would be to see trends based on subject or project tags. Anyone else doing something similar? I want to hear your thought.

I want to hear your thoughts.

Thank you.:slight_smile:

2 Likes

There are a few things moving forward right now, @brayo is making development easier by moving towards https://github.com/ActivityWatch/aw-tauri instead of the current https://github.com/ActivityWatch/aw-qt as main entrypoint for the application.

This should help us streamline and modernize the ActivityWatch stack run by most users. Finally switching to aw-server-rust as the default server (better performance). Should also enable features like notifications to work on macOS, once the Python modules are rewritten in Rust and embedded in aw-tauri. Will also help us finally support Wayland out of the box on Linux by shipping with https://github.com/2e3s/awatcher/ for Linux systems.

These might not be the fanciest plans from a user perspective, but it helps us maintain ActivityWatch in the long run and provides a better experience for new users.

I’m personally spending most of my time developing https://github.com/gptme/gptme, but there’s quite a bit of overlap as I’m using gptme to analyze my ActivityWatch data. Trying to build a sort of AI analyzer/coach/buddy by using https://github.com/gptme/gptme-agent-template together with my quantified self code https://github.com/ErikBjare/quantifiedme to autonomously spot trends and perform analyses.

2 Likes

Fancy, what about just using a regular chat thread in an LLM?