Scenario: there's a better new alternative to a standard, but it's less conveniently reliable for users than the old standard due to slow adoption

Specific example: some different nostr clients experimenting with ways of referencing media other than DNS-based URLs, while DNS-based URLs still tend to be the ones that embed or display properly in most clients

Reasoning: devs and users are being lazy supporting legacy DNS without adopting the new thing

Instinct: stop using legacy DNS for that use case, forcing adoption on your end to punish incompatibility by drawing it out

Result of instinct: you have less reach and dumb people think it's your fault when your content doesn't load

Giving up: just using DNS and not trying to push things faster than devs naturally go

A more balanced strategy: using DNS, but recommending against it and more importantly, sticking with services and devs who support the better thing, until the better thing is common enough to switch from DNS

The point: you can use a legacy standard while also refusing to use implementations that don't support the newer standard, if you're looking for a balance between crippling your reach or giving up. Same principles apply in other examples of such scenarios