From: Andrew Gierth (andrew@erlenstar.demon.co.uk)
Date: Thu May 09 2002 - 13:26:34 CDT
>>>>> "Henry" == Henry Spencer <henry@spsystems.net> writes:
Henry> Mind you, there's a potential problem with that: if a reader
Henry> agent *is* checking, it won't believe Xref headers with a
Henry> different name, no matter how much care has been taken to
Henry> ensure that they are valid. I have no data on how widespread
Henry> such checking is.
I don't believe there is significant use of non-matching Xref headers
(at least, I don't recall ever seeing any system set up that way).
What _does_ happen is sites putting a dummy path entry in, so that
the headers look like
Path: 0!news.example.com!...
Xref: 0 example.group:1234
which would look right to a client that checks, but which introduces
a potentially non-unique path entry.
In this instance, though, the non-uniqueness of the path entry causes
no harm that I know of (since it's in addition to the site's real entry).
There are variations on this; e.g. "sn-us" as the path entry (short but
probably still unique).
-- Andrew.