[Gendispatch] Re: New Version Notification for draft-rsalz-2026bis-11.txt

[Brian E Carpenter <[e-mail]>]

Hi Rich,
On 16-Oct-24 09:07, Salz, Rich wrote:
>>>> I think there are bogons in there, e.g. RFC 3450 (chosen at random) references 2026 but doesn't cite the reference. But quite a few to check...
>   
>>> Are there? Do we udate references when RFC N becomes RFC Nbis? I thought we just assume that the obsoleted-by tag is good enough.
> 
>> We don't. Possibly we should, but that's an RSWG topic. I just wanted to observe that to do complete due diligence, we'd need to look at all of them to see if any of them are significant.
> 
> I do not think I understand what you are saying.
> 
> Suppose that RFC N references 2026, but isn't on the updated-by list[2]. Are you saying that we need to see if any of the nearly 300 in that list [1] actually do update 2026? If that is what you are saying, I disagree. 

No, it's more subtle. Is RFC N significantly affected by the changes made to RFC 2026? If so, does something have to be done about it? The answer is probably "No" but how can you tell without checking?

The reason I'm sensitive to this is that it recently happened to me, as editor of a draft that obsoletes an RFC - we had to add two "Updates: " and a couple of sentences to fix up some collateral damage to two dependent RFCs. This will show up in the metadata for those two RFCs.

Of course you could decide not to bother with this, with a small risk of leaving a few trip hazards.

     Brian


> We're not verifying that the RFC Editor/Publication Center did their job correctly and that the updated-by  field in [2] is correct. I am not aware of any bis-type draft that has done this; are you?
> 
> Another option is that you could be pointing out that many RFC's reference 2026 (as opposed to, say, BCP 9). I don't think that is a problem either, since if anyone fetches RFC N, and then goes to look up RFC 2026 (in either the datatracker or the RFC Editor website), they'll see that it is obsoleted by this new document. Sure, there are probably copies of 2026 floating around that will not have that information, but this is the Web and there's not much we can do about that. I think if RFC N references 2026, then when RFC N-bis is published the 2026 reference will be updated. Unless and until then, there is nothing to do.
> 
> If you are not saying either of the above, please explain it to me.
> 
> [1] https://datatracker.ietf.org/doc/rfc2026/referencedby/
> [2] https://www.rfc-editor.org/info/rfc2026
> 
>