Interesting graph, but is 'video' really a different class of traffic to web? Every other type shown in the graph is literally a different internet protocol (email, telnet, web, etc). Separating 'video' from 'web' seems a little unwarranted.

It's true that you could make a distinction - the web is hyperlinked multimedia and video isn't very web-like according to that definition.

But I don't think you can claim from that graph that 'the web is dead'. The weaker claim that 'traditional hypermedia is giving way to video' would be better supported, but less interesting.

tldr - false dichotomy