Description
Before you have a reaction to this make sure you read the whole thing.
For the past couple of weeks, it has become clear to me that this project should die.
There is a lot of confusion for users, developers and me on how this actually works and what it can and should do..
I know that something like this has the potential to be useful, but currently, due to the structure of it, any changes are like painting over a rusty car.
My issues with it.
- The documentation is bad.
- The code itself is generally bad.
- The documentation is bad.
- The code is in need of optimization.
- The documentation is bad.
- The code is too complicated for anyone to contribute to this.
- The documentation is bad.
Will the current version stop working?
no
The current version of this will still continue to work, the thing that will change are my efforts on feature development.
Bug fixes will still be added but I will not add more to it.
What now?
The only thing that makes sense is for me to stop development of it, and rather start over with something new.
For that I need your help, I have a general idea of what but that is me, I do not know what you (as a user or developer) what/need it to do.
This is the time for you to help decide!.
The first thing I want to change is the name, when this project started it was simple, it updated your custom components and that was it, now it does that and much more.
I have landed on HACS (Home Assistant Community Store) for now, since I believe that cover more of what it actually does.
If you have ideas on what you like it to do, please open an "issue" here https://github.com/custom-components/hacs/issues (I need input from both users and developers to do this)
All ideas will be considered.
The general idea is to take how custom_updater and custom-component-store work today with inspiration from hassio add-ons, and make a new awesome framework.
Until the new project (codename HACS) have an MVP, I will still continue to support this with bug fixes, it is only new features that I will stop adding.