| 1 | Decision-Making Process in the Git Project |
| 2 | ========================================== |
| 3 | |
| 4 | Introduction |
| 5 | ------------ |
| 6 | This document describes the current decision-making process in the Git |
| 7 | project. It is a descriptive rather than prescriptive doc; that is, we want to |
| 8 | describe how things work in practice rather than explicitly recommending any |
| 9 | particular process or changes to the current process. |
| 10 | |
| 11 | Here we document how the project makes decisions for discussions |
| 12 | (with or without patches), in scale larger than an individual patch |
| 13 | series (which is fully covered by the SubmittingPatches document). |
| 14 | |
| 15 | |
| 16 | Larger Discussions (with patches) |
| 17 | --------------------------------- |
| 18 | As with discussions on an individual patch series, starting a larger-scale |
| 19 | discussion often begins by sending a patch or series to the list. This might |
| 20 | take the form of an initial design doc, with implementation following in later |
| 21 | iterations of the series (for example, |
| 22 | link:https://lore.kernel.org/git/0169ce6fb9ccafc089b74ae406db0d1a8ff8ac65.1688165272.git.steadmon@google.com/[adding unit tests] or |
| 23 | link:https://lore.kernel.org/git/20200420235310.94493-1-emilyshaffer@google.com/[config-based hooks]), |
| 24 | or it might include a full implementation from the beginning. |
| 25 | In either case, discussion progresses the same way for an individual patch series, |
| 26 | until consensus is reached or the topic is dropped. |
| 27 | |
| 28 | |
| 29 | Larger Discussions (without patches) |
| 30 | ------------------------------------ |
| 31 | Occasionally, larger discussions might occur without an associated patch series. |
| 32 | These may be very large-scale technical decisions that are beyond the scope of |
| 33 | even a single large patch series, or they may be more open-ended, |
| 34 | policy-oriented discussions (examples: |
| 35 | link:https://lore.kernel.org/git/ZZ77NQkSuiRxRDwt@nand.local/[introducing Rust] |
| 36 | or link:https://lore.kernel.org/git/YHofmWcIAidkvJiD@google.com/[improving submodule UX]). |
| 37 | In either case, discussion progresses as described above for general patch series. |
| 38 | |
| 39 | For larger discussions without a patch series or other concrete implementation, |
| 40 | it may be hard to judge when consensus has been reached, as there are not any |
| 41 | official guidelines. If discussion stalls at this point, it may be helpful to |
| 42 | restart discussion with an RFC patch series (such as a partial, unfinished |
| 43 | implementation or proof of concept) that can be more easily debated. |
| 44 | |
| 45 | When consensus is reached that it is a good idea, the original |
| 46 | proposer is expected to coordinate the effort to make it happen, |
| 47 | with help from others who were involved in the discussion, as |
| 48 | needed. |
| 49 | |
| 50 | For decisions that require code changes, it is often the case that the original |
| 51 | proposer will follow up with a patch series, although it is also common for |
| 52 | other interested parties to provide an implementation (or parts of the |
| 53 | implementation, for very large changes). |
| 54 | |
| 55 | For non-technical decisions such as community norms or processes, it is up to |
| 56 | the community as a whole to implement and sustain agreed-upon changes. |
| 57 | The project leadership committee (PLC) may help the implementation of |
| 58 | policy decisions. |
| 59 | |
| 60 | |
| 61 | Other Discussion Venues |
| 62 | ----------------------- |
| 63 | Occasionally decision proposals are presented off-list, e.g. at the semi-regular |
| 64 | Contributors' Summit. While higher-bandwidth face-to-face discussion is often |
| 65 | useful for quickly reaching consensus among attendees, generally we expect to |
| 66 | summarize the discussion in notes that can later be presented on-list. For an |
| 67 | example, see the thread |
| 68 | link:https://lore.kernel.org/git/AC2EB721-2979-43FD-922D-C5076A57F24B@jramsay.com.au/[Notes |
| 69 | from Git Contributor Summit, Los Angeles (April 5, 2020)] by James Ramsay. |
| 70 | |
| 71 | We prefer that "official" discussion happens on the list so that the full |
| 72 | community has opportunity to engage in discussion. This also means that the |
| 73 | mailing list archives contain a more-or-less complete history of project |
| 74 | discussions and decisions. |