I fancy the approach myself... but, much like the grand ideas of OSI in networking, Tim Burners Lee's idea of the Web was never what was going to be used and a lot of the original terms/concepts just plain don't make sense for the Web that exists. It was, originally, a true document system, nothing more. It wasn't until many changes from the original idea that it started to be what people wanted & actually use today. Nor was it even part of Tim Burners-Lee's original concept that 99% of Web traffic would be something other than users viewing & editing HyperText documents.

The User Agent would get a HyperText document from a Web server, render it, and show it to the user. You could parse HTML, display images, and follow links - with the assumption the user would want to edit them like any other document. There was no encryption. There was no scripting for dynamic content. Styling was done via semantic markup. There wasn't even any POST when the User Agent term was initially applied. Everything was supposed to be like one big distributed document store instead of centralized hosting on few servers for many client devices.

All that is to say: Regardless what I think sounds nice myself, if the web was just the original vision & definitions instead of everything it changed and grew to become we wouldn't even be using it to have this conversation.

Interestingly, people seem glad to have telephony companies filter bots even though the original idea of the telephone did not call out robocallers as invalid clients. Probably because, unlike the modern Web, the average person gets a chance to see what it means to allow anything inbound even though the original concept never considered robocallers.