Most of the authors are actually missing. Full processing would have yielded multi-trillion row dataset. I didn't rally have that kind of compute with me.

I have even tried running the cross-join on BigQuery... after one hour, only about 3% was done.. so, had to cancel it.