Pretty much the whole point of the article is that when you say "http vs https" you are no longer just saying TLS vs no TLS. It now means something more.
But does it? I mean, you can start a WebSockets session with an HTTP request too. No one says a WS server is an HTTP server and that HTTP is a confusing thing now.
If you assume a reasonably recent browser and server, then it does. Certainly one might not grant that assumption, but I don't think it's outright wrong to make it, either. Depends on your approach.
Why assume anything? Why not simply call things what they are? The issue is that browsers don't signal that you're using http2 and they decided nor to use a new schema either to help confuse us all. His point is literally based on nothing supporting http2 without TLS. What if something did? Would http also be a meaningless word? Why not call things what they are, not by what the browser is hiding.
HTTP2 can be started via an HTTP over TLS request, that doesn't mean that it's HTTPS as defined in https://tools.ietf.org/html/rfc2818 and https://tools.ietf.org/html/rfc7230