If you're wanting to hyper-optimise systems and only hyper-optimise systems for particular purposes - then, like with the original introduction of engineering - the correct approach was always going to be "wait six months to a year for the metas to emerge, then follow one of those from a guidebook" - and that would be the case even with more info from Frontier as to what the qualitative effects of things are, because the "real" metas would only come out with players twisting systems beyond Frontier's original design intent.
I don't see how you could plausibly "brick" a system, though. Everything you build does something, every asset has some sort of use and benefit. Even a Colony Orbis parked around a non-landable planet where nothing can influence its economy type is providing an excellent location to sell a lot of high-value goods as well as general increases to various system properties, a convenient shipyard, a nice orbital large pad place to pick up missions, etc.
In the meantime, for those of us who do enjoy experimenting to see what happens, this is the only chance we'll get. Colonisation will only be new once.