Loïc Denuzière

> It of course follows that the constraints would have to be copied over and still checked, not just ignored. Which is tricky for "allows ref struct" since F# will...

I like this. I think the third proposed syntax, with `internal new` listed like a member, is the best. Putting constructor information between the type name and the equal sign...

@kspeakman I still prefer to have a type for which I can be certain that if I have a value of this type, then it is correct. But I hear...

Presumably these problems are related to having some union cases, or some record fields, private? If we simply allow making construction private regardless of case/fields, then these problems shouldn't arise....

I kinda like the idea, although it feels a bit weird to have `match` followed immediately by a pattern.

> But Don also considered `match? … with … -> …` and left a rationale in [#705 (comment)](https://github.com/fsharp/fslang-suggestions/issues/705#issuecomment-1157739765). `match!? ... with ... -> ...` in a computation expression would be......

MinimalAPI-style host builders are becoming pervasive in the .NET ecosystem, and they expect `args` as first argument. Aspire, for example, uses one. And the lack of `args` in F# can...

There are so many arbitrary choices that are necessarily made in something like this, and I'm a bit uncomfortable with having such specific choices being blessed by the standard library....

@Happypig375 My point isn't which specifics you chose; my point is that there are so many specifics to choose 😄 Something like this will necessarily be opinionated in what kind...

@smoothdeveloper If this is allowed then I assume it will be subject to the same caveats as `as this` regarding initialization soundness. For example this throws an exception: ```fsharp type...