I just watched https://www.youtube.com/watch?v=Q4dYwEyjZcY https://www.youtube.com/watch?v=Q4dYwEyjZcY this video about the early HTML standardization process,

I just watched https://www.youtube.com/watch?v=Q4dYwEyjZcY https://www.youtube.com/watch?v=Q4dYwEyjZcY this video about the early HTML standardization process, and it seems to explain all the ills of HTML. So indeed, there is a huge optimization opportunity of having a stricter error model. Also, I'm really wondering how much battery could be saved when surfing such pages. Also I'm sure there is a lot of potential going in the pre-parsed document model. But that's a next level kind of engineering I guess.