Sci-Hub link to pdf http://sci-hub.tw/https://www.nature.com/articles/s41567-019...

Wow, this makes way more sense than in the press release. Scihub is a treasure.

I'm not from academia but is it unusual to have 21 authors on a paper?

Listed authors of reports are a bit like credits in film, and similarly, in fields/production where you have a lot of special-effects/complicated scientific apparatuses/huge custom built computer clusters for doing the analysis, you get a lot of supporting personell that you might want to credit.

Oarticle physics papers sometimes list a huge number of authors, record I think is 5154 for a paper from an LHC collaboration titled "Combined Measurement of the Higgs Boson Mass in pp Collisions at √s=7 and 8 TeV with the ATLAS and CMS Experiments"

Nature wrote about it https://www.nature.com/news/physics-paper-sets-record-with-m...

My observations: I rarely see it in computer science related papers. I often see it in biology / medicine papers (but I don't see lots of those papers)

If you wonder what 21 people do for one paper, they have mentioned it in the paper pdf (look for the scihub link here in the comments) on the last page: Only two people actually wrote the paper with input from a lot of others, supervised by yet another one.

Lab experiments involve huge amount of support people who have to do really specific work for each experiment.