RFD 0

Form and Function of RFDs

Requests for Discussion - or RFDs in short - are central to our operation. They indicate that nothing we do or know is set in stone. RFDs are drafted based on all the information that was available at the time of writing. However, just like for our software, new insights arise and new opinions are voiced. RFDs reflect that.

When you want to approximate the truth, you should question yourself

~ Floor Rusman (translated)

We ask for a critical stance to any RFD you read, and encourage additions, removals and updates in conjunction with the team.

Credit goes to Oxide for their inspirational pursuit of openness, transparency and trustworthiness. Our format heavily draws from their RFD page.

Changes

All our RFD source files are tracked here using git. Displaying the diffs of each RFD discussions are now a feature. While reading an RFD, you can navigate through its history in the top-left menu.

Function

RFDs are meant to serve as a basis of a meaningful discussion within the team, but also outside of it. As with programming, writing down RFDs in a meaningful way requires care and attention. By making it standard practice we aim to keep getting better at it, while building a knowledge base worth visiting at the same time.

This knowledge base should not be a mere technical one. We use it as a platform to describe company policy, insights and strategy. By publishing this out in the open, we want to give others the means to hold us accountable for our words. We encourage other companies to do so as well.

Form

There are no strict RFD style guides at this time, but RFDs should be clearly scoped. When it is hard to define the scope of an RFD, it should most likely be split up into multiple separate RFDs.

Just like with software, the quality of an RFD is judged by the reader, never the writer.

History

We keep track of changes to all RFDs via git and show the differences publicly bringing full transparency into the life cycle of an RFD. It's important for us to look back on how decisions were made in the past to continuously improve our methods.