Markus Schirp

The challenge is that the `coverage_criteria` will be supported per subject, and that would not fit within the CLI flags without becoming a source of headache for the user. Curious...

> presence and sunk cost of existing tooling Can you elaborate? From my perspective, RES would still use the Makefile to trigger mutant, but not have to re-implement a mutant...

I still think that feature should exist, so re-opening it.

> Will this include [].freeze and {}.freeze? I would be quite interested in that mutation. Yes these are literals.

And another thought pro this mutation: Mutant is about to reduce unused semantics in the code. If mutability of a build-in datastructure is seen as a "feature that is by...

> I guess my concern is the effort to add .freeze to literal strings now to kill mutations, when the strings will be frozen AFAIK in Ruby 3 by default....

> I18n.t('class_title', klass: klass) Would it be a bug if some library mutates the `'class_title'` ?

You can alwas whitelist these via rspec: ``` before do allow(I18n).to receive(:t) do |selctor, options| expect(selector).not_to be_frozen end.and_call_original end ```

Maybe this is also finally a case where I'd support adding conditionally disabling a mutation operator.

> Yes it would be a bug, but one that would not encourage me to add .freeze to hundreds of them with Ruby 3 under development. You can use a...