[gnso-ff-pdp-may08] Thread hijacking

[Dave Piscitello <dave.piscitello@xxxxxxxxx>]

Quick point of order.

Mike has asked that we distinguish threads using subject lines.

This discussion has veered from the original thread. It's a good discussion but 
if possible, I'd like to ask that we try to separate threads as much as 
possible.

Can I reclaim my original thread subject: DTeam - framework for proposal?




On 8/7/08 2:37 PM, "Eric Brunner-Williams" <ebw@xxxxxxxxxxxxxxxxxxxx> wrote:



Howdy George,
> ...
> Anyhow, if I had a client who committed credit card fraud, and I
> registered domains on their behalf, I'm the one that takes the loss,
> not Tucows. Thus, the onus is on me to know my clients, etc., I can't
> push the responsibility on to others.
>

I may not have stated the problem clearly. There is the possibility of a
race condition existing where two or more instances of the registrar
client (each a distinct "reseller"), and the "window of opportunity"
can't be known to any instance of the registrar client at the time of
the add, only at the time of delete.

The same applies to registry-specific discounts that rely upon global
state, e.g., first 100 at $x, thereafter at $x+$y.

Restated, you (registrar X sub-account) can't know what any other
registrar X sub-account is doing, and therefore you can't know what the
status of registrar X's threshold value is, whether you think you are
responsible or not.

> The actual threshold that was used for the AGP/tasting solution was
> enormous (i.e. 10%). If folks are seeing a fraud rate of greater than
> 10%/yr on new registrations, perhaps they should not be registrars, as
> they've got bigger problems, and shouldn't be shifting those problems
> to everyone else.
>

That's a nice value judgement, and I've heard it elsewhere, but I'm
testing a registrar client and I'd prefer if people with values other
than stable, robust, well-tested registrar systems had the courtesy to
go out of business themselves and not inflict those values on others.

> Shifting the discussion back to Fast Flux ...
>

The point of mentioning it (yours) was that something mitigated
something else without side-effect.

The point of mentioning it (mine) was that the above representation of
absence of side-effect may be a question of perspective, or at least my
being smarter after I hit my thumb with a hammer.

Which may have some non-zero relation to "fast flux".

Eric