Thanks for the details, Russ. On Tuesday, May 28, 2019 at 11:53:09 AM UTC-4, Russ Cox wrote: > > On Mon, May 27, 2019 at 7:08 PM Matt Farina <matt....@gmail.com > <javascript:>> wrote: > >> 1) when a company runs a project without much publicly documented process >> but does as they choose, isn't that a sign of a company run project? >> 2) The go team at Google has had processes that are not public. One >> example is the proposal review process. There has long been a group at >> Google that decides these. For a long time this wasn't documented publicly >> but happened. The public documentation on it came after the decision on go >> modules. >> > > This is demonstrably false. We launched the proposal process at Gophercon > in 2015 <https://talks.golang.org/2015/how-go-was-made.slide>, well > before modules. We did it explicitly to open the process of contributing > ideas - as opposed to code - and make it more accessible to new > contributors. You can see the full commit history > <https://www.google.com/url?q=https%3A%2F%2Fgithub.com%2Fgolang%2Fproposal%2Fcommits%2Fmaster%2FREADME.md&sa=D&sntz=1&usg=AFQjCNE4YrifErkcQQ6ckzSBOjgcXhIoQA> > > of the process description online. I tried to streamline the doc to make it > more approachable in 2018, but I did not make semantic changes in that edit. >
Let me be a little more clear. In the docs (see the version prior to the 2018 update at https://github.com/golang/proposal/tree/a16a937b3b39c4c42f063842407c30c4c451b524#process) there was no documentation on the proposal reviewers on the Go team. How proposals were reviewed was not *documented*. Numerous people, myself included, found this by seeing comments on issues that referred to this group. Saying something in a speech at a Gothercon is not documentation. People who were not paying attention, were not there, or forgot we also not able to come upon this information. This is why I referred to it not being documented publicly. > > 3) how has no one outside of Google qualified for the core team and why >> aren't more companies who are heavy users in on owning it? >> > > It's unclear what you mean by "core team." If you mean the set of people > who can approve (+2) and submit code changes, then as Ian said, there are > more people outside Google than inside Google at this point. > I think this highlights a documentation gap or place where people have different understandings. Is the organizational hierarchy of Go documented somewhere? Lots of phrases are thrown around and different people can throw different meanings on them. When I refer to that group I'm thinking of things like the Kubernetes Steering Committee. It is the core set of people responsible for the decisions on the project. This is a smaller group than the committers. It is the decision/direction makers for the project. This group, as far as I'm aware, is all Googlers. It may not be documented and there may be assumptions. This is what I expect from a company run project. > > On Mon, May 27, 2019 at 9:17 PM Matt Farina <matt....@gmail.com > <javascript:>> wrote: > >> As a concrete example: Cloudflare pretty heavily uses Go. When a >>> cloudflare-employee started stepping up to work more and more on the Go >>> crypto stack, they got hired by Google to do it fulltime. At least from the >>> outside, that seems to what happened with Filippo Valsorda. >>> >> >> That's fantastic. This shows some open source in action. It also >> highlights that this isn't open governance. If it did, Filippo joining the >> core team would have been a separate event from joining Google. >> > > We did give Filippo the ability to +2 (review, approve, and submit) crypto > CLs while he was still at Cloudflare, so in that sense it *was* a > separate event. Filippo being at Google now means that working directly on > Go can be his full-time job now. My understanding is that he had > responsibilities at Cloudflare beyond contributing to the Go project. > > A better example might be the various engineers who work at companies like > Intel, ARM, and Microsoft on Go support for those companies' processors and > operating systems. Many of them can now +2 code change as well, and they do > so within their area of expertise. > Thanks for sharing the additional detail. This has more do to with ownership than reviewer/approver status. I want to make something clear, I'm not suggesting Go change it's governance in any of these messages. I'm only attempting to provide evidence that Go is a Google owned project. Many open source projects are company owned. I use a bunch of them. I'll let other people have a debate on how things *should be.* -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/55d512bb-5afd-4487-9a5c-cc9f48ba683d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
