^^^Any attempt by a Heist Action to modify protected text instead does nothing. Text is protected if it appears after a caret (^) and before a dollar sign ($), with no other carets or dollar signs in between.$$$

This text is not protected as it contains both a ^ and a $.

Or rather, the text that is protected is “) and before a dollar sign (”.

Good point – it’s hard to word this to work as we’d want! Hopefully this new wording works.

I would not say that it has changed the way that the rule interprets at all, to my eyes. It may even be worse as the matching is now subjective, as parentheses can contain parentheses.

Ultimtately this is starting to become what I feared when I started drafting my CfJ: which is to say, not a CfJ, in that rather than fixing the proximate issue attemtping to address fundamental underlying issues that should really be handled by proposal.

The intent of my CfJ was only ever to be a good-enough patch while proper proposals readdressed how the mechanic works, because fundamentally all strings-of-symbols approaches I think will end up on this exact kind of issue. I’d like to see proposals that reimagine the way that layered protection works

On that basis I’ll probably vote against this, because it’s misconceieved - we shouldn’t be iterating on the curly brace formulation, we should be reimagining the mechanic of protection, and also grasping the nettle of how the ruleset handles copy-pasted mediawiki markup.

It crosses my mind that the best fix would probably be to have a symbol that can be placed anywhere in a sentence and protects the entire sentence – that way, we don’t need to mess around with matching pairs. That might be too big a fix to do by CFJ, though.

FWIW, I’d say that this wording does work – text can be inside the outer set of “parentheses” and that protects it just as well as being inside the inner set. But I’d also think it should only be temporary until we can find a better solution.

(And I’ve now submitted the better solution as a proposal, “Golden Bolts”.)