Thanks Sean for this really good clarification. I think you are hitting on a key problem with the personal data ecosystem in general. We are all dependent on competent performance of neighboring layers, and would benefit from standardization on these layers, but “competent” turns out to depend heavily on the use case, and use cases vary quite a lot in the details. This works against standardization. So there are a lot of people “rolling their own” in all areas. The lesson from previous rounds of tech evolution is that it’s unrealistic to expect not to have to break down and recompose your system several times to accommodate the evolving architecture. It won’t be pretty, but that’s what we’re in for. In other words, you probably WILL have to create your own database to “speedily” move on with your project, and then replace it down the road if a standard emerges. This is not the answer you probably wished for, but at least having a clear view can allow you to allocate “just enough” time to it, knowing it isn’t a permanent solution. At the detail level, designing for modularity will also help you in the future when the standards change. (This is easier said than done, I know.)