Re: BCP97bis

[John C Klensin <[e-mail]>]

--On Monday, October 18, 2021 08:50 +1300 Brian E Carpenter
<[e-mail]> wrote:

> On 18-Oct-21 07:04, John C Klensin wrote:
>> 
>> 
>> --On Sunday, October 17, 2021 17:13 +0000 "Salz, Rich"
>> <rsalz=[e-mail]> wrote:
>> 
>>>   *   To make a stable reference in an unstable world, add a
>>> date.
>>> 
>>> Or add text "retrieved on xxx xxx xxx" like Wikipedia
>>> often does.
>> 
>> That makes it clear what is being referenced, which is good.
>> Absent other provisions/ requirements, it does nothing to
>> guarantee that the content will be there when someone goes
>> looking for it years (or decades) later.
> 
> Indeed. I've used that formulation in academic publications,
> and in my mind it's an abbreviation for "I apologise if this
> citation is no longer valid when you happen to look for it,
> and I would have preferred a properly archived reference but
> I was too lazy to find one."

And I have been a reviewer and on an editorial board for
academic publications where references like that have been
spotted and interpreted exactly as the abbreviation you suggest.
The question is then raised as to whether the reference is
important enough to the article that people will actually want
to look it up and study it.  If the answer to that question is
"yes", the draft goes back to the author with "try harder"
instructions.

> Which isn't really good enough for a normative reference in
> a standard, of course.

Exactly.

    john