The people reviewing were often experts in the very very niche area of mathematics the paper was written in. So a) it took them less time and b) reading and understanding a proof in their niche is valuable to them.
There was another way. Desk reject, major revision. 900 pages is not reviewable. Its practically gibberish. This is being released for fear of being scooped