Re: #1416: USEPRO 3.9: Reinjection and Injection-Date

[Russ Allbery <rra@xxxxxxxxxxxx>]

Frank Ellermann <nobody@xxxxxxxxxxxxxxxxx> writes:
> Russ Allbery wrote:

>> if we want to use the normal arrival-based expiration, we need to
>> require it somewhere.

>> If we disallow late injection in general, both injection and
>> reinjection, then we don't need Injection-Date.

> Yes, but that's not realistic.

It should be a warning sign for a train of logic if the thing that one
declares not realistic is how all of Netnews currently works and what RFC
1036 requires.

It's realistic; Netnews has worked that way for decades.  The question is
rather whether the benefits from changing it are worth the costs, and
proceduraly, whether it's worth reopening the topic rather than working
with what we've got now in USEFOR.

> Users can poll their server, and then go offline for some time creating
> articles with different days, waiting in their outbound for better
> times.  Then they go online again and try to post their articles.
> That's one scenario where Injection-Date could be better than Date.

Or said off-line posting agent can rewrite Date when the message is
actually injected.  That's what they have to do now.  You lose information
for humans, but the protocol works.

> Either I'm too clumsy to get this right, or both USEPRO drafts don't
> discuss the function of a history file yet.  S-o-1036 9.2 explains how
> that's supposed to work, proposing a minimum of N = 7 days for "seen",
> with a Date older than now - N considered as "stale".

> If it's really not yet there USEPRO needs a similar section with the
> same RECOMMENDED N = 7 days minimum.

We already discussed this years ago and reached a working group consensus
to not document any specific retention time, but instead to document the
restraint on retention time (namely that you have to reject any article
with a Date older than your retention time).  I'd prefer not to revisit
that decision again.

In practice, people use all sorts of different retention times, ranging
from three days to a year (and in some specialized cases, even shorter
than three days -- I've used a retention time as short as a day before).

-- 
Russ Allbery (rra@xxxxxxxxxxxx)             <http://www.eyrie.org/~eagle/>