HN user
The 85 % figure is from Environ. Sci. Technol. 2011, 45, 21, 9175–9179 https://pubs.acs.org/doi/abs/10.1021/es201811s
The 200-500 kiloton/year is from https://pubs.acs.org/doi/10.1021/acs.est.3c05955
For a deeper dive you can peruse the works that cite these articles, there's lots of research into microplastic population dynamics. Unfortunately most reach similar conclusions.
Microfiber cloths are, on the other hand, a major source of microplastic pollution. They are basically just ready-to-snap-off microplastic particles. They shed a substantial % each washing cycle. Something like 80 % of all microplastic pollution sampled in ocean water is from microfibers.
It wasn't a good fit for me. The strip of windows extending past the border of my screen, sometimes showing half a window, triggered a weird anxiety, it kept drawing my attention. I used it for about two months and then ditched it for a more traditional tiling compositor (hyprland) where windows don't overlap the screen border.
Niri is, however, very pretty from a technical standpoint. Modern Rust codebase, good code structure, very easy to understand and start hacking.
The problem with the current EU-wide rating system is that it assumes the interior is ventilated as required by other regulations. Furniture materials are assigned an emission class (E0, E1, E2) based on the steady-state VOC concentration in the test chamber (EN 717-1 and others).
Virtually no home manages to hit the minimums for ventilation (25 m³/h per person, IIRC), especially during winter. So even though an E1-class particle board is labeled as safe with regards to VOC emissions, in practice the room will reach far higher pollution levels than that category allows.
It also doesn't control for the amount of material present, so you can have a small properly ventilated room covered floor to ceiling with E1 particle board furniture, but achieve E2 or worse VOC levels in the air.
Backstory: former employer spent $$ to integrate a VOC sensor into their residential HVAC/HRV solution. The idea was to increase ventilation when odors are present and reduce ventilation when the air was clean again. The engineer prototyped and tested the device in their pre-war home (brick, stone, old wood). All good. They had the first batch manufactured and sent out to customers in brand new homes full of modern materials (engineered wood, vinyl flooring etc). The sensors were permanently saturated (reading maximum VOC value) and the project ended up being canceled, because newly built homes were their entire customer base.
You wouldn't be losing FP niceness with Zig, and the pattern matching and enum situation is also similar to Rust. Even better, in a few areas, for example arbitrary-width integers and enum tagging in unions/structs. Writing parsers and low level device drivers is actually quite comfortable in Zig.