Design meeting: "Marker effects"

[traviscross]

Summary

@nikomatsakis will have a document for us to talk about "marker effects".

About this issue

This issue corresponds to a lang-team design meeting proposal. It corresponds to a possible topic of discussion that may be scheduled for deeper discussion during one of our design meetings.

[traviscross]

We ended up not doing this on 2025-04-30. In the planning call today, we talked about rolling the discussion on avx2-like target features and const trait impls into one discussion, which we've now rescheduled for 2025-06-18.

See also:

• Design meeting: SIMD multiversioning #309

[traviscross]

cc @veluca93

[traviscross]

cc @oli-obk @fee1-dead @yoshuawuyts @RalfJung @Jules-Bertholet @compiler-errors @Nadrieril @workingjubilee

[RalfJung]

Do you see some relation between those features or is that just to discuss more than one topic in the meeting?

[workingjubilee]

I believe the insight they are attempting to start from, here, is that both are about "a context where it is valid to evaluate f(x)", and especially that both are now touching on how you write traits that describe them.

[RalfJung]

"a context where it is valid to evaluate f(x)"

So in other words, a general effect system (albeit without user-defined effects)... that is tempting, I just hope it doesn't lead to feature creep and delaying const traits another 3 years.

Unfortunately I won't be able to attend the design meeting.