--On Wednesday, 02 April, 2003 08:27 +0200 Harald Tveit
Alvestrand <harald(_at_)alvestrand(_dot_)no> wrote:
idiot isps who configured route filters but did not bother to
maintain them. darwin at work. the subject is
uninteresting, as the study of stupidity is an exceedingly
for those of us who are endlessly fascinated by watching
evolution in action - what was the previous usage of 69/8 that
led to those filters being installed?
(parenthesis: similar things have reportedly happened to
people using .info for email - there turns out to be a number
of MTAs in the world who have hardcoded all the non-2-letter
TLDs, assuming "there will be no more", and routinely toss
mail from/to the newer ones. They, too, deserve the pain they
get; unfortunately they, like the route filterers, don't get
all the pain they cause.)
See draft-klensin-name-filters-00.txt for a longer explanation
of this issue and some cases other than mail-tossing. I'm
working on a version that fills in the empty sections and
clarifies anything that I can find that isn't clear; comments
and suggestions welcome.
The evolutionary implications of this one are, however, somewhat
less clear than those of the 69/8 case. In the 69/8 case, the
ISPs who did the filtering were ultimately subject to attack by
their own customers who, presumably, would sooner or later go
elsewhere (or cause other harm to the ISP) if the problem wasn't
fixed. An MTA-operator whom I'm paying to receive and deliver
my mail is (or should be) subject to roughly the same pressures.
But, in the TLD case, the victims are mostly those who have
registered in the new domains and the help desks of registrars
and unrelated servers, neither of whom have significant leverage
on the offending ISPs in the non-MTA cases.
It might also be part of a good case for not creating new gTLDs
except where they are needed for technical reasons (pretty much
a null set), but please hold any discussion on that subject on
the whining-at-ICANN list of your choice. Spending excessive
time on that one of course leads to another example of evolution
in action :-(.