Re: Errata 5173 on RFC8200

[Fernando Gont <[e-mail]>]

On 9/8/19 11:44, Ole Troan wrote:
>>> I think this exposes a larger problem with the errata system.
>>> The errata system is like a bug reporting system, where both the bug report and the proposed solution is merged.
>>> That creates a mess. As in this case we have 4 errata, that is solved by one single change.
>>>
>>> A cleaner end result would have been one erratum with the problem description and the new text.
>>> Where the new text was a result of IETF consensus.
>>> Is there any way we can achieve that?
>>
>> In short, no. As you pointed out the errata system has the issue and the resolution conflated. I think this is something we need to think about and provide input about before someone designs the proposed in-line errata system.
> 
> Right, with the inline errata system you cannot accept 4 errata with the same resolution.

The thing here is that there were separate errors submitted in different
errata, that would lead to separate patches.

What you propose is to patch multiple things on the same patch/commit.
-- not that I necessarily disagree.

One can use the first errata, put the text in verifier's notes, and I
guess, for the rest, HFDU (or so) and point to the first errata.

-- 
Fernando Gont
SI6 Networks
e-mail: [e-mail]
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492