Re: [OPSAWG] draft-ietf-opsawg-tacacs-06 Contributions, Status and Plans

[Alan DeKok <[e-mail]>]

> On May 15, 2017, at 1:00 PM, t.petch <[e-mail]> wrote:
> I echo part of what Alan says, that for a WG document, the editors
> should reflect the consensus of the WG.  The problem I see is the lack
> of consensus, not with people disagreeing, but with an absence of people
> agreeing.

  The only agreement I see was at one of the early WGLCs.  There was wide-spread support for publication of the draft, when it was just not ready.

  i.e. if it had been approved by the WG and gone to Security Area or Transport Area review, it would have been bounced back with comments much like the ones I've made.

> Alan made a number of comments in October last year, Alexander made some
> in  November but I did not see much follow up from anyone else to either
> set of comments.
> 
> Trouble is, do the editors incorporate comments that one person has made
> and noone else has agreed or disagreed with?  There is no good answer.

  People argue on the list...

> In other WGs, I have seen ping-pong, one person comments, comments
> incorporated, someone else then disagrees, disagreements incorporated
> into a new revision, first person comes back, changes incorporated into
> a newer revision and so on, circling around a lack of consensus.

  Chairs / ADs should make a final decision in such a situation.  Or, the document should be dropped as having insufficient consensus.

> Changing editors, unless it is to someone remote from the subject, is
> unlikely to change things..

  The problem I see is that most people supporting the document are just too close to it.  There has been wide-spread support for publication when it just wasn't ready.

  Plus, (speaking as an implementor) I had no idea what to do when reading early versions of the draft.

  And other implementors aren't speaking up and saying what they've done, or how to address issues in the document.

  It really does look like the majority of supporters want a TACACS RFC, but they have no opinion about what the content should be.  For me, that raises questions about why we need a TACACS RFC.

> I did look at Alan's comments, agreed with some, disagreed with others,
> ditto Alexander's, but was disinclined to do more with noone else
> chipping in, especially as several more did chip in in the initial
> stages of should we adopt this, and what status should it be.

  Part of the problem is that it's hard to have an ongoing conversation around my reviews.  The comments are so numerous that discussing each one publicly would require hundreds of messages.

  Which for me is a sign that the draft is just nowhere near ready for publication.

  Alan DeKok.