One use case for this repo is to test ORTB implementations; it's much easier to do that if the JSON can be read directly rather than having to parse it out of markdown. It also plays nicer with editors, which will more readily highlight issues like https://github.com/openrtb/examples/pull/2/files , https://github.com/openrtb/examples/pull/10/files .