no, but if a documentation tool does support multiple dialects (for example, the DocC dialect, and that used by the standard library / SwiftNIO), it would need to know what dialect to parse the doccomment as. to make this concrete, the traditional means of creating asides in Swift is with an UnorderedList, but DocC uses BlockQuotes instead. the situation is even more messy with “Parameters”/“Returns” detection.
because aside and parameter detection generates a lot of false positives, it is not desirable for tooling to accept both forms at the same time; it’s better to be able to specify which syntax the tooling should use.
you would be surprised at the amount of documentation that is already falling through the cracks because of syntax differences like these.
example: system/filedescriptor/duplicate(as:retryoninterrupt:) , the documentation for the target parameter was lost because it was written in backticks.
of course tooling would have to start supporting block directives in order to use @version, there’s no other way to bootstrap this. but compared to the current situation, where there is no tooling directive at all, this would at least be an improvement.
DocC has already introduced a lot of syntax without much consideration for how it fits into or impacts the existing ecosystem, and a lot of this new syntax (like the aside keywords, which seem to have been assigned completely by fiat) isn’t even documented as far as i am aware.
i don’t see how this is relevant. inherited documentation shares the same doccomment as the declaration it was inherited from, so it would use the same syntax as the original doccomment. this is an argument for introducing a version directive like @version, since it may be a different syntax than what the module itself uses.
i don’t think it is reasonable to expect people to employ source formatters to migrate documentation. those of us who work on source tooling tend to be far more trusting of them than the general community, i have found.