At Fri, 7 Nov 2003 18:01:21 +0100, BELOEIL Luc wrote:
> 
> Rob, I do think that "badly configured network" is an operational issue.

Right.

> Is your point that "this possibility" is more dangerous than possible
> operational issues of other methods, and thus you think that well-known
> address method should not be used ? I hope I'm clear... I'm trying to
> understand ;+)

Stipulating that opinions differ on whether this is a serious issue,
and that I'm only writing about this again because Luc asked me for
clarification:

Yes, the concern was about fault isolation.  One can do damage with
any mechanism if one tries hard enough, but I do see a difference
between mechanisms in which dumb mistakes tend to aggregate and
mechanisms in which they do not.  In both cases I assume that the
users directly affected are unhappy, the question is how much
collateral damage there is to the larger network.  YMMV.
#----------------------------------------------------------------------
# To unsubscribe, send a message to <[EMAIL PROTECTED]>.

Reply via email to