> This is the downside of geoengineering, or chasing any engineering trend: it's possible to build amazing things, but we always forget that maintenance in perp

> This is the downside of geoengineering, or chasing any engineering trend: it's possible to build amazing things, but we always forget that maintenance in perpetuity is much much harder. This rings especially true in relation to much of our work in tech, as I think you were pointing to. It’s easy to run an AB test and look at the win, forgetting that for every new thing, you now have to support it. For the easy to maintain stuff, you might even forget how it works, so the team of ten who built it turns into a single part time maintainer.