LEO satellite constellations would seem to be a bad idea
... because, quite apart from the effect on astronomy and, eventually, on other launches, they are incredibly wasteful of resources.
Think about it: an LEO satellite has a lifetime of around a year, after which it burns up in the atmosphere and must be replaced. The materials used to make the satellite and its electronics, some of which are quite rare while even the copper and gold used for PCBs and contacts are not exactly common, are completely lost since, on re-entry, the entire thing is reduced to tiny, unrecoverable pieces in the upper atmosphere.
By comparison, the unthinking fat cat who, 'because he can', buys a new iPhone every year and landfills the old one is a paragon of materials thrift. Why? Because the location of landfill sites are known. They can, and will be, mined for their more valuable content. This will be recovered and reused: a landfill could easily have a higher concentration of the many of the commercially useful rare elements than the mines they are currently extracted from.
This rant was inspired by today's 'Life Scientific' on Radio 4 - an interview with Sarah Bridle, a data scientist who has found the skills she developed searching for dark matter and dark energy are equally useful for dealing with the effect of food production on global warming. Its available from BBC Sounds and is worthwhile listening to.