[rfc-i] Referencing Internet Drafts

Heather Flanagan (RFC Series Editor) rse at rfc-editor.org
Fri Jun 16 12:49:39 PDT 2017

On 6/16/17 12:37 PM, Bob Hinden wrote:
> Heather,
>> On Jun 16, 2017, at 11:16 AM, Heather Flanagan (RFC Series Editor) <rse at rfc-editor.org> wrote:
>> On 6/15/17 11:38 AM, Bob Hinden wrote:
>>>> On Jun 15, 2017, at 1:37 AM, Julian Reschke <julian.reschke at gmx.de> wrote:
>>>> On 2017-06-15 02:03, Brian E Carpenter wrote:
>>>>> ...
>>>>> This triggers one of my hobby-horses, I'm afraid.
>>>>> If I refer to an I-D because it's useful reading but not yet published, I'm happy with it being tagged as "Work in progress". A smart reader may even have the idea to look for a corresponding RFC.
>>>>> If I refer to an I-D because it has historical value, I would prefer it to be tagged as "Unpublished draft" and have an exact version number (and date). Calling it "Work in progress" is either inappropriate or simply untrue.
>>>>> Today we have no metadata to separate the two cases.
>>>>> ...
>>>> +1
>>> I tend to agree as well.  Though I am not sure “unpublished draft” is exactly right since the draft is available online.  We treat Internet Drafts as temporary, and at the same time permanent.  Starting to seem more like the later.
>>> Why don’t we include a link to the Internet Draft in the reference like we do for RFCs?
>> Historically, it's to enforce the face that I-Ds are not equal to RFCs,
>> and because I-D URLs were ephemeral (since I-Ds would disappear after
>> six months). The latter is no longer true. How do you feel about the former?
> Certainly I-Ds are not equal to RFCs, though there are a bunch of sub cases depending on if the ID is just an individual submission w/ no review, an active w.g. document, or approved by a stream for publication.  I conclude that calling them “work in progress” may or may not be correct when an RFC is published with a reference to an ID, and certainly not true 10 years later.
> I don’t think we want to have labels for each stage of an IDs life because it changes over time.  Perhaps calling them something like “working draft” or similar might be better than “work in progress”.  It doesn’t imply it is changing.

Process question: Does this suggest we need to do something special to
update RFC 2026 (which is where the "Work in Progress" term for I-Ds
comes from)? (See Section 2.2).

> Since they don’t appear to be ephemeral any longer, including a link in the reference might be useful.

I'm fine with that.


More information about the rfc-interest mailing list