North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

Re: Lazy network operators - NOT

  • From: Paul Vixie
  • Date: Sun Apr 18 11:08:19 2004

> > ... Margin pressure makes it impossible for most "broadband" service
> > providers to even catalogue known-defect customer systems or process
> > complaints about them.
>
> What is the estimated cost per subscriber of such an operation in your
> opinion and where should it be to make it feasible? Off-the-shelf
> automation can accomplish this for pennies per subscriber per month,
> keeping the catalogs up to date and informing users automatically.

let me drive deeper into what i mean by "margin pressure".  it means every
penny-per-month has already been allocated ("pre-spent") and if some group
(like the abuse desk if there even is one) wants even one of those pennies-
per-month then they will need SVP approval, which they aren't going to get.

> After deployment there is a smallish support burst, but after the levels of
> infection plummet and stay at levels two orders of magnitude lower than
> prior situation, queues will shorten and customers will be significantly
> more happy.

you know that, and i know that, and we agree on that.  but the SVP in
question does not know that and cannot be convinced of that.  this short
sightedness used to be a uniquely US/Canada business phenomenon but i now
see that we've exported it to europe and are working hard to get it into
asia and latin america now.  if you want an SVP to say yes, then to cover
their own ass they have to be able to prove that revenue will increase in
the medium/long term or that costs will increase in the short term.  your
story -- which i agree with -- is that at a slight cost increase in the
short term they can reduce costs in the medium/long term.  that won't fly,
anywhere in the world that "broadband" has taken hold.  "revenue neutral,
cost increase" is many words as that SVP will be able to speak before they're
standing on a street corner selling pencils (...again).

> I would expect the community who uses similar blackhole criteria as you to
> be fairly insignificant to the spammers revenue stream. So the stream must
> be cut at the source, not just fending off the 1% somewhere.

you are all going to have to do this.  questions which remain are: (1) will
we each maintain our own private list of broadband networks or will there be
a good/fast/cheap/global aggregator of such information?  and (2) what is
the magnitude and slope of the volume-over-time of spam-from-"broadbanders"
that will make the average edge/customer e-mail relay operator do as i've
done?  and (3) when or if IETF MARID ever produces results, will they be
useful/relevant and if so will they cause people to relax their "broadband"
filters or will other forms of unwanted traffic (that MARID doesn't police)
have risen to the point where "it isn't just spam any more, sorry."?