Re: [netmod] [Technical Errata Reported] RFC8342 (5514)

[Juergen Schoenwaelder <[e-mail]>]

On Fri, Oct 05, 2018 at 09:53:32AM -0700, Andy Bierman wrote:
> On Fri, Oct 5, 2018 at 9:12 AM, Lou Berger <[e-mail]> wrote:
> 
> > My personal opinion (with any hat on) is that it isn't appropriate to make
> > a technical change that impacts implementation in an errata.
> > Clarifications of original intent, corrections of inconsistencies and
> > editorial corrections are perfectly appropriate.  I'm happy to learn that
> > this intended use/scope of errata is wrong.
> >
> >
> Strongly agree.
> Errata cannot be used to change technical decisions.
> It can only be used to correct text that is incorrect.
>

So far so good but we all know that at the end its a judgement call.

The intention of the text was to ensure that there is always a defined
origin value. One way to achieve that is to have an origin defined at
the root. But there are obviously other possibilities to achieve the
intended goal, as the example demonstrates. While we can for sure add
an origin at the root in the example, it serves no purpose. In other
words, the example demonstrates that we failed to reinforce this by
saying "every config data node needs to have a defined origin value
and one way to achieve that is to define origin's are the roots of the
subtrees" but instead we created the rule "all roots of the subtrees
must have a defined origin value", which in certain approaches to
maintain origin metadata and generate origin attributes is more a CLR.

But sure, if people want to close this discussion on formal grounds,
so be it. We will just have introduced a CLR.

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
Fax:   [phone]         <https://www.jacobs-university.de/>