It’s not; if enough clients implement a very basic feature all clients need to implement it or their users will think the client is broken.

If a few big clients implement HTML rendering on kind:1s then a client that doesn’t will show raw HTML and people will think it’s sucks and move on.

Protecting from complexity at this level is important if we don’t want to end up with a handful (or less) apps in nostr. 

 I guess trying to protect against edits, and HTML rendering, and nanny bots, and any other feature that not all clients are necessarily interested in implementing just seems like an impossible expectation to me.  I would argue that experimentation at this level is more important - especially since clients are still broken in many different ways anyway.

If users think that a client sucks because it shows raw HTML instead of rendering it the way some other clients do, that sounds like how progress ultimately proliferates.

As an aside, FWIW, I'm not convinced that clients that obscure original pre-edited notes will ultimately be deemed by consensus to be the superior clients. 

 People think 'to replace' when they hear/see edit, and that's where people on Nostr are think 'lol no'. But if it's a new post that gets added on top of the OP, and repeats as we see it's history, and replies being attached to whoever version they replied to, is fine imo. 

 Good point