Re: [Ietf-and-github] Warren Kumari's Discuss on draft-ietf-git-using-github-05: (with DISCUSS and COMMENT)

["STARK, BARBARA H" <[e-mail]>]

>> I agree with Martin here: the point of having this go through a WG and consensus is that we think that if you are using Github this is a good set of practices and that we recommend them in that case. So while it's at their discretion that doesn't mean we don't have an opinion.

> There were also real concrete improvements that came about because of the WG process.  The most notable example is the “light weight” mode, that culminated in Barbara being added as co-author.  No amount of presentations at the wgchairs lunch could have achieved that.

I just now managed to get through all my emails from yesterday and today and would like to post something quickly before it gets overtaken by others' comments.
First of all, I think Martin is doing a fabulous job and fully support all the changes he's made. If he ever does something I *don't* support, I'll make it known. 
As someone with strong tendencies to pragmatism, I think the most important thing is to get these recommendations out. WGs are using GitHub. Independent of the existence of other git mechanisms (I really like the Atlassian suite of tools, with Jira and Bamboo integration -- but they totally require someone to administer them), GitHub is what many WGs are using (to various degree). Independent of the fact that GitHub only runs on IPv4, WGs are using GitHub. A dogmatic stance of "because GitHub is IPv4-only IETF MUST NOT provide guidance or tools to the WGs that are using it" would be beyond my comprehension. Any suggestion that IETF forbid use of GitHub is untenable, IMO.

Therefore, guidance *for those WGs making use of GitHub* is needed. And that's what this draft provides. I'm not sure how many different ways this draft can say that it isn't requiring or forcing any WG to use GH. And it isn't requiring or forcing those who do use GH in simplistic ways (e.g., just draft authors make use of it and other list participants don't) to institute heavy-handed policies. I use GH with a co-author without WG-chair-blessing in one WG. In a WG I chair, one set of draft authors is using a repository in an account I set up for the WG; no-one else is participating there, and the other draft author chose not to use GH. This is all perfectly ok under the provided guidance.

BTW, I don't (competently) write code. When I was first forced to start using git (for document management, in Broadband Forum), it evoked significantly more terror in me than, say crawling through a cave or rappelling off a 30 m cliff. But it's where things are headed. The more advice and useful info we can provide, the better off everyone will be.

I honestly don't care what label is attached to it. The advice is useful for those using GH. Therefore, I'm confident it will be used.
Barbara