Re: #1416

["Charles Lindsey" <[e-mail]>]

In <[e-mail]> Russ Allbery <[e-mail]> writes:

>"Charles Lindsey" <[e-mail]> writes:

>> That left just one small worry. I wanted (version IC) Injection Agents
>> to _always_ insert an Injection-Date (except when it was already
>> present, of course). Russ argued (version IR) that this would sometimes
>> cause existing implementations to behave oddly, and proposed a
>> less-intuitive rule for when Injection Agents should insert it.

>I think there's also a remaining point of disagreement over whether
>injection agents SHOULD reject articles with a stale Date.

>I just submitted draft-ietf-usefor-usepro-10 including the (I believe
>non-controversial) new history section and corresponding rewordings in the
>duties of a relaying and serving agent.  Below is the remaining diff for
>my solution for #1416.  Please note if any of the below is uncontroversial
>so that I can commit those sections and reduce the diff to only the
>disputed portion.

90% of what follows is agreed, and it might be helpful to get it into an
ID so that we can see it all in context, and then argue over the small
remaining difference. Such concerns as I have with the rest of it are all
in the "niggles" category.

>             <t>It SHOULD reject any proto-article whose Newsgroups header
>             field does not contain at least one &lt;newsgroup-name> for a
>@@ -770,8 +829,14 @@
>             the source of the article and possibly other trace information
>             as described in Section 3.2.8 of <xref target="USEFOR" />.</t>
> 
>-            <t>The injecting agent MUST then add an Injection-Date header
>-            field containing the current date and time.</t>
>+            <t>If the proto-article already had an Injection-Date header
>+            field, it MUST NOT be modified or replaced.  If the
>+            proto-article had both a Message-ID header field and a Date
>+            header field, an Injection-Date header field MUST NOT be
>+            added, since the proto-article may have been multiply injected
>+            by a posting agent that predates this standard.  Otherwise,
>+            the injecting agent MUST add an Injection-Date header field
>+            containing the current date and time.</t>

That is the principle point of difference between IR and IC. I would
prefer it to say

>+            <t>If the proto-article already had an Injection-Date header
>+            field, it MUST NOT be modified or replaced.  Otherwise,
>+            the injecting agent MUST add an Injection-Date header field
>+            containing the current date and time.</t>

If that change is made, there are probably a few consequential changes
and niggles elsewhere, but that is the essence of it.

In that case, there is a small possibility that, until all agents have
been upgraded, a few users may occasionally see an article they have seen
before. See <http://www.imc.org/ietf-usefor/drafts/issue-1416> for the
somewhat complicated chain of events that can give rise to this.

OTOH, without my change, most articles will never receive an Injection-Date
at all, and that will continue to be the case undefinitely (unless Posting
agents routinely start to do it, which I consider unlikely).

-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: [phone] Fax: [phone]   Web: http://www.cs.man.ac.uk/~chl
Email: [e-mail]      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5