Stream: t-compiler/wg-meta

Topic: MCP process RFC ready?


nikomatsakis (Apr 03 2020 at 12:28, on Zulip):

Hey @T-compiler/WG-meta and cc @oli -- I put in some time to the Major Change Process RFC and I think it's basically ready. I would appreciate it if folks want to talk a look:

https://hackmd.io/VhVSTwgiTcmEslbpC7qCOQ?view

It includes a [revised template] for major change proposal issues, @oli, that I based off of the starting point you landed (but added a few things and changed the style).

nikomatsakis (Apr 03 2020 at 12:28, on Zulip):

Er, the here is a link to the revised template -- editing the comment seems to mess up Zulip notifications. =)

nikomatsakis (Apr 03 2020 at 12:31, on Zulip):

Ah, one minor thing I wanted to check with you on, @Santiago Pastorino, was around how to include MCP things in the triage meeting agenda. I wrote that they should ping the proritization wg, but maybe people can just directly add them to some hackmd?

pnkfelix (Apr 03 2020 at 14:04, on Zulip):

whoa, can zulip stream names have spaces? "#t-compiler/major changes"

nikomatsakis (Apr 03 2020 at 14:04, on Zulip):

yes, see e.g. #project leads (public)

Santiago Pastorino (Apr 03 2020 at 14:06, on Zulip):

pnkfelix said:

whoa, can zulip stream names have spaces? "#t-compiler/major changes"

hard to adapt to modern stuff, I wonder if they allow emoji :P

pnkfelix (Apr 03 2020 at 14:06, on Zulip):

Q: if a change has already gone through the RFC process, but implementation details were left unspecified there, am I right in thinking that the implementation details may well be expected to go through MCP?

Santiago Pastorino (Apr 03 2020 at 14:06, on Zulip):

we are from the no spaces in file names era :P

pnkfelix (Apr 03 2020 at 14:06, on Zulip):

(i'm thinking specifically of cargo-report-future-incompat here)

Santiago Pastorino (Apr 03 2020 at 14:06, on Zulip):

nikomatsakis said:

Ah, one minor thing I wanted to check with you on, Santiago Pastorino, was around how to include MCP things in the triage meeting agenda. I wrote that they should ping the proritization wg, but maybe people can just directly add them to some hackmd?

this is very interesting, I guess we would need to add a WG-prioritization tag

Santiago Pastorino (Apr 03 2020 at 14:07, on Zulip):

and for whatever reason people can just tag the WG

Santiago Pastorino (Apr 03 2020 at 14:07, on Zulip):

MCP being one but there may be others

Santiago Pastorino (Apr 03 2020 at 16:22, on Zulip):

@nikomatsakis let me know if you want me to do something about this but maybe we should have a WG-prioritization label on github

Santiago Pastorino (Apr 03 2020 at 16:23, on Zulip):

the docs you linked look good to me, have fixed some small typos

nikomatsakis (Apr 03 2020 at 17:08, on Zulip):

pnkfelix said:

Q: if a change has already gone through the RFC process, but implementation details were left unspecified there, am I right in thinking that the implementation details may well be expected to go through MCP?

I thnk so

nikomatsakis (Apr 03 2020 at 17:08, on Zulip):

Something else that was raised is "where is the line between this and a RFC"

nikomatsakis (Apr 03 2020 at 17:08, on Zulip):

I think the way I'm thinking of it is:

nikomatsakis (Apr 03 2020 at 17:09, on Zulip):

for some compiler impl detail, people should start with MCP

nikomatsakis (Apr 03 2020 at 17:09, on Zulip):

we may guide them to an RFC if it seems useful

nikomatsakis (Apr 03 2020 at 17:09, on Zulip):

this is kind of the "So you want to do X" flowchart I was talking about

nikomatsakis (Apr 03 2020 at 17:09, on Zulip):

I wonder if it wants to be part of this RFC...

Last update: Sep 28 2020 at 16:15UTC