committees wants to make a better language ▪ Even if no two of us can agree on what that is! ▪ It is Consensus by Committee ▪ Not Design by Committee ▪ I don’t speak for the Committee ▪ No one does ▪ Polls Implicit Dangling Resolution (4/6/9 Consensus Against) ▪ P2742 Indirect Dangling Identification (4/6/9 Consensus Against) ▪ P2750 C Dangling Reduction (4/6/9 Consensus Against) ▪ P2821 span.at() ▪ P2878 P2878 Reference Checking (R2 5/12/18 Consensus Against; now on R5) ▪ P2951 Shadowing is Good for Safety (not yet discussed) ▪ P2955 Safer Range Access (not yet discussed) 47P2951 SHADOWING IS GOOD FOR

consists of regular Jane programmers  …with an atypical passion for C++  Decisions are made via consensus  Meetings aren’t held on gigantic snowy mountains  …it’s a meeting room in a hotel 10© 2024 reliable, well-documented software. 3. Industry standard technology. Where there's industry consensus on best practices, we should take advantage. Innovation in tooling and style is not our purpose

easier to continue modernizing the Code Generator. We did it without risk! And we did it with the consensus and participation of dozens of teams. Prologue© 2018 Bloomberg Finance L.P. All rights reserved critical space. The wins came with NO RISK! They were accomplished with overwhelming community consensus.© 2018 Bloomberg Finance L.P. All rights reserved. © 2024 Bloomberg Finance L.P. All rights reserved

chasing our tails •Recall ^^x reflects on x, [:y:] unreflects (splices) y; [:^^x:] is x •Large consensus on introspection query: contemplation is great •Fear and loathing about code generation • Expansion objects deemed acceptable •Consequence: • Severely limited: can’t do 3D with 2D abilities • No consensus on “how much” generation is just enough •No clarity on necessary primitivesWhere Do We Want To Be

code they contain! ● ✓ Projects should be “cattle,” not “pets”! 7Why dependency management? 8Consensus: Managing dependencies == way too hard Q: Which of these do you find frustrating about C++ development maintainers ○ Core CMake developers ○ Toolchain and OS vendors ○ You? ● Share and confirm consensus via public ISO C++ Tooling Study Group (SG15) papersCPS plans 77 ● Continue with MVP in CMake

time Adobe 80exhaustion mind the power dynamics Adobe 82exhaustion who gets to be part of consensus? Adobe 83exhaustion being home for bedtime is a hard-to- beat perk Adobe 84exhaustion men can

under 2 out of 3 major compilers Analysis revealed that even within WG21 there is no obvious consensus about the standard in this regard 😱Adapting Existing Lib to become Deterministic ✅ 1 Our Take

Requires: Libs: -L${libdir} -L${sharedlibdir} -lz Cflags: -I${includedir}Why this scope? - Easiest consensus - It will deliver the most value, the sooner, for users and the community - Largest contributor

discussion point from previous work. The discussion points brought up do not necessarily represent a consensus view. 📖gist.github.com/dietmarkuehl/file-lets-hash-things-over-md2024 Victor Ciura | @ciura_victor

performance reasons. • The error handling mechanism was addressed by C++’s Direction group around 2016 (contracts) and again in 2018. • The exception mechanism was addressed by Herb Sutter (Chair) in in 2019: P0709: Zero-overhead deterministic exceptions: Throwing values • The exception mechanism was debated in Prague (last WG21 meeting) with a paper by James Renwick, Tom Spink, Björn Franke: Low-cost Low-cost Deterministic C++ Exceptions for Embedded Systems • I expect a major change in this mechanism or alternatives in the near future. MOTIVATION 4• Part 0: Prologue: What are Exceptions? • Part