The real lesson here should be that source code shouldn't be stored in a text format, but in a well-defined strict binary format that stores the parse tree directly, which completely eliminates the need for formatters.
I agree with this sentiment. It also provides an opportunity to store other metadata with the code. It lets different people choose their formatting preference when rendering back to human readable code.
The source code is text thing seems like a "legacy" concept.
At the very least I think this is an interesting thing to explore. Maybe it leads nowhere...
