[rfc-i] Valid email addresses [last call "On Authors, Contributors, Editors, and overload."]

Paul Hoffman paul.hoffman at vpnc.org
Tue Jan 10 12:48:47 PST 2012

On Jan 10, 2012, at 12:36 PM, Heather Flanagan wrote:

> On 1/10/12 12:14 PM, Paul Hoffman wrote:
>> On Jan 10, 2012, at 12:07 PM, Brian E Carpenter wrote:
>>> Maybe this is a side issue, but I am aware of a quite recent case
>>> where an author insisted on listing an obsolete email address in
>>> an RFC, because it was his address at the former employer where
>>> he did the work.
>>> We know that email addresses are not for ever, but should there be
>>> a default policy requiring a valid address at the date of publication?
>> Good point, and it is not just email. Should the policy be that the 
>> stated affiliation be valid at the time of publication, regardless
>> of what the affiliation was during the document development?
> (removing RSE hat for a moment)
> Requiring current affiliation makes me twitch a bit.  If Joe Smith does
> his work on an RFC as part of his employment contract with Cisco, and
> then during the AUTH48 process goes to work for Juniper, putting Joe
> Smith, senior engineer at Juniper as the affiliation seems very bad
> form.  That said, I think it is the author's call, and he and his
> previous employer can discuss any breach of contract.

The discussion so far has been about "valid address" and "stated affiliation", not "current affiliation". As you point out, insisting on "current affiliation" would be difficult, but it is also unnecessary. Many IETF regulars with known affiliations use no affiliation instead, which is just fine. Brian's original question about "valid address", meaning an email address that still exists. Often an email address at a previous affiliation will bounce, which makes it useless as an identifier.

--Paul Hoffman

More information about the rfc-interest mailing list