Re: [Ietf-and-github] Éric Vyncke's Abstain on draft-ietf-git-using-github-05: (with COMMENT)

[Martin Thomson <[e-mail]>]

Hi Éric,

I understand the reasons for your abstention.  I hope that these responses help, even if I'm guessing that they won't address your concerns.

On Tue, Mar 10, 2020, at 22:02, Éric Vyncke via Datatracker wrote:
> Is there a reason why "GitHub" is used rather than simply "Git" or similar
> system ? At least, it is explicit to be the commercial web site "github.com".
> Alternatives such as GitLab and BitBucket are only briefly mentioned.

Because that is what people are using.  And we've seen significant payback from that.  Both in terms of using common - and therefore well-maintained - tools and in terms of accessing an existing community of contributors.

I personally do this in full appreciation for the drawbacks, both in terms of centralization and technology.  For the former, git is good about that, and we've a strategy (in -configuration primarily) for mitigating risks associated with centralization.  As for IPv6, I'm not prepared to stand on principle in this case, but respect the choice of others to do so.

FWIW, the alternatives are pretty good, and most have all of the same capabilities (some are even better in some ways).  Working groups probably won't suffer too badly for a choice of an alternative service.  But for me access to a wider community of contributors is - unfortunately - a determining factor.

> General comment: it appears that the two 'git' documents have very similar
> content. Was it considered to better split them or merge them?

This was debated at some length in the working group.  This one will end up having BCP status.  I suspect that the other will have less long-term utility as it is more tactical.
 
> Section 3.2: a private repository is not free AFAIK with github.com. Which
> party will pay for those repos ?

Private repositories are offered for free, with some limitations.  https://github.blog/2019-01-07-new-year-new-github/