On Mon, Sep 23, 2024, at 07:21, Tiago Freire via Std-Proposals wrote:
> Here’s the summary of the discussion for those who missed it.
 
<snip 1-5 but they are good points>

You are missing 6: modules have been in the standard since C++20 and what we want everybody to use. #include isn't quite depreciated/obsolete but it shouldn't be used. There is little interest in making #include better because we want those efforts to go making modules work in the real world. As such if you want to propose this you need to convince the committee there are significant issues with modules such that for some environment they can never work.

I realize modules have been slow to roll out. There are still early adopters holding off because important parts are not ready. The rest of us waiting for the early adopters to figure out what morality guides should be - given extensive real world experience here is what you should do to avoid the pain we faced. However there is reason to think that before the most trusted figures on the committee could ram this through into the standard those issues will be solved. Sure there will be many people waiting to adopt modules after that, but they won't have adopted the standard version of pragma once either.