Design Meeting: Path Inference Experiment

[JoshBashed]

Summary

Proposing a T-lang design meeting to discuss the goals, scope, and experiment plan for RFC 3444 "Path inference" and the prototype implementing leading-dot syntax (e.g. .Variant, .{ … }, .(…)). The aim is to align on the mental model and any syntax concerns (e.g. behind -Zpath-inference).

Agenda

• What exactly are the goals and intended mental model of path inference?
• Should we explicitly deny generics in enums (.Pending<f64, Foo>(0.0);)?
• Should we include alternative syntaxes in the experiment (e.g. :Variant, :{ … } or _::Variant, _ { ... }), if so which ones?

Background reading

• RFC draft
• Zulip thread

About this issue

This issue is a lang-team design meeting proposal for scheduling a discussion on the path inference experiment, its goals, scope, and gating strategy.