---
title: T-spec meeting 2024-07-11
tags: ["T-spec", "meeting", "minutes"]
date: 2024-07-11
discussion: https://rust-lang.zulipchat.com/#narrow/stream/399173-t-spec/topic/Meeting.202024-07-11
url: https://hackmd.io/60etUTkfRnOBiR9NLJTrVg
---

Attendees: Joel Marcey, Connor Horman, Mara Bos, Eric Huss, Monadic Cat, TC, Pierre-Emmanuel Patry, Lukas Wirth

Agenda:

* `t-spec` Team Member Addition Proposal
* Spec Project Goal?
* Inline Assembly Chapter Conversion

## Adding `t-spec` team member

Joel: Connor has been officially contracted by the rust foundation. Should we add Connor to the `t-spec` team ? Should this be done in the context of just "Connor" or "as a contractor" ? We don't have to decide *now*.

Connor: I will need at least write access to the repositories in order to merge pull request.

Joel: If we add Connor to the `t-spec` team Connor would get those permissions by default.

Connor: I did speak to Ray (other lead of lccc), he didn't expect any conflict of interest on that side. I don't think I'm the right person to analyze that in regards to the Rust Project Side.

TC: We'll create a t-spec-contributors team as the place to start here to give Connor what he needs quickly.

## Spec Project Goal

Joel: Do we want to submit a spec-base [project goal](https://github.com/rust-lang/rfcs/pull/3672). It may be too late but we possibly could have until July 12th to submit something.

Joel: Do we want to have a small project goal or move forward with our current plan for 2025?

Mara: Library team doesn't have a project goal.

Joel: We don't really know yet what teams will do, it is too nebulous at this point.

Eric: Once we'll have things rolling we will be able to create goals.

TC: Having a goal would be on net good. My main reservation had been that I'd prefer at this point that Joel and Connor spend the time doing rather than on more planning.

Eric: I'd prefer flying under the radar but I don;t have any strong preference.

TC: We could essentially copy and paste our plan into a goal.

Connor: We do already have a goal that is not formalized, the process that we agreed upon in June. What is the state of the reference, how could we commit to that fully, we need to formalize those plans as a goal.

TC: We'll talk about it offline. We could probably do a goal about creating a "draft version of the spec". It'd be good to have something up to increase the visibility of the team a bit.

## Inline Assembly Chapter Conversion

Joel: We are converting Rust Reference documentation into a format for the Rust specification. Connor is leading the first conversion that we will model for the others - Inline Assembly.

Eric: Announcement, integration is not done. I ran into a frustrating problem.

[reference#1523](https://github.com/rust-lang/reference/pull/1523)

Connor: *Presenting the spec*.

Eric: The green color could lead to problems for people with visual impairments.

Connor: I need to modify the CSS.

Pierre-Emmanuel: We should probably adapt the color for color blind people.

Connor: Striping happens automatically

Joel: Question for the group: Each of the sections got a tag, I imagine for linking purpose. Do we want those tag publicly visible ?

Connor: In c++ those tags are handy to point to a specific item and link to it.

TC: Connor, what was your process for doing this?

Connor: I had in mind the things from the layout chapter.

[...]

Mara: Paragraphs are too small.

TC: One thing that makes the FLS hard to read is the granularity, so it seems good this is separated into somewhat larger ideas. Having to actually created named identifiers to each section also makes this more important.

Joel: Is Connor on the right track?

Mara: Yes.

Joel: We need to ensure we got the correct trajectory(?).

TC: Skimming it, high level, this looks right.

Eric: I got a question about the "dynamic" word in the tag.

Connor: It comes from dynamic semantic, description on how things run at runtime.

TC: Let's remove it for now and just follow the lexical structure. We can readdress this along with any updates to that.

Mara: The document is moving back and forth from the compiler and the program's perspective a few sentences appart.

Connor: Yeah, I mostly kept it separated by section, and I was clear when I needed to address the compiler.

TC: In terms of review, after we make the changes discussed here, and this looks good to us, we'll ask Amanieu for review. What comes next after this?

Connor: Maybe atomics.

Mara: I can help review that.

(The meeting ended here.)