One thing I like about connecting to linker artifacts is that it restricts
the problem from how to build some module interface to how to build an ABI
equivalent interface for a particular binary in a particular ecosystem.
It's still not a trivial problem, but it's one we have some experience in
providing solutions for.

On Mon, Apr 11, 2022, 17:40 Daniel Ruoso via SG15 <sg15_at_[hidden]>
wrote:

> Hello,
>
> As discussed in previous meetings, we need to solve the problem on how
> to solve the discovery of the metadata for modules. This paper uses
> the "linker arguments" idea.
>
> It makes the caveat about "interface-unit-only module libraries" as
> well as to the fact that this will eventually be superseded by proper
> package management (although I think it will remain useful as a
> lowest-common-denominator option even then).
>
> https://isocpp.org/files/papers/P2577R0.pdf please take a look and let
> me know what you think.
>
> daniel
> _______________________________________________
> SG15 mailing list
> SG15_at_[hidden]
> https://lists.isocpp.org/mailman/listinfo.cgi/sg15
>