Re: draft-ietf-usefor-usepro-02

["Charles Lindsey" <chl@xxxxxxxxxxxxxxxx>]

In <41BE00BA.48C6@xxxxxxxxxxxxxxxxx> Frank Ellermann <nobody@xxxxxxxxxxxxxxxxx> writes:

>Charles Lindsey wrote:

>> Will the rest of the Working Group please tell me whether
>> it wants me to continue to reply in detail

>Maybe s/perverse/strange/ or odd, peculiar, idiosyncratic.

>I'm still not sure what you have in mind with a gateway saving
>Message-IDs as comments.  Maybe replace it by ten repetitions
>of "above all avoid loops".  Or add some jokes:  "If you're a
>news to avian carrier gateway make sure that you're the only
>one."  "While it's funny to gateway news to IRC you'll be shot
>if somebody else gateways IRC back to news (and a Message-ID
>in a comment is no excuse in this case)".

The best means for preventing loops is to ensure that Message-IDs are
propagated correctly. That is the first line of defence, but it sometimes
fails because message-ids get lost/changed/whatever somewhere in the other
medium. So we need a backup method.

The commonest backup is to insert an extra header in the gatewayed
article, such as:

X-Gatewayed-By: <name of gateway>

which the named gateway can then check for before gatewaying any message.

That is the second line of defence, but if fails if the other medium does
not allow X-headers of that nature, or even any headers at all, or if it
cannot be relied upon to propagate such headers reliably. So we need a
last-ditch backup, for use only in these really nasty situations.

And the only last-ditch possibility left is to include something in the
body of the message, such as:

THIS ARTICLE HAS ALREADY PASSED ONCE THROUGH <name of gateway>.

Which will look pretty ugly to people who see that in the body, but what
else could be done to prevent the loop?

-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133   Web: http://www.cs.man.ac.uk/~chl
Email: chl@xxxxxxxxxxxxxxxx      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