Re: [gnso-rap-dt] Front-Running

[George Kirikos <icann+rap@xxxxxxxx>]

Hi Jeff,

On Wed, Aug 5, 2009 at 10:36 AM, Neuman, Jeff wrote:
>
> I had a question on the methodology of the report.  Maybe I missed it, but 
> does Ben reveal which (or even how many) registrars were used in the testing?

As I implied yesterday, it was a very simple report (that's why I
caleld it a "little study"). If you go to the first page of the
report:

http://www.icann.org/en/compliance/edelman-frontrunning-study-16jun09-en.pdf

"I formed a list of web sites to be tested based on top organic search
results for domain‐related search terms (e.g. “register a domain”,
“check whether a domain is available, “domain availability”, and “get
a domain name”). I checked whether each site in fact provided a domain
search function, and I discarded any site that did not provide such a
function."

He later goes on to say that 200 distinct sites were tested. He does
not reveal how many of those were registrars. Indeed, he does not even
reveal whether he was logged in to the registrar, or whether he was
changing IP addresses when he was doing the lookups, or the "quality"
of the names that he did lookups.

If one wanted to be harsh, one could simply conclude people doing
front-running (if they exist) don't care for low quality domain names
that are looked up from IP addresses belonging to Harvard. :)

Let's give a different scenario, a high value registrant (let's call
him a "Whale"), who is "cookied" and logged in to a registrar (or a
WHOIS provider) checks for the availability of that domain name. A
domain name is found to be available. Would the front-runner (if they
exist) treat the whale differently than they would the Harvard
researcher doing random lookups?

Or to put it in financial terms, I go into a ghetto and see a few bums
discussing IBM's stock price, and how it's a steal at $100/share, and
that they're investing in it. Do I trade based on that information?

Contrast that with an alternative scenario, I'm at a dinner party in
the Hamptons and a M&A guy from Goldman Sachs is on his cell phone in
the bathroom, and I happen to overhear as I pass in the hallway that
they're finalizing the acquisition of Acme Inc. by IBM. Is the quality
of information in this case better or worse? :)

If a registrar is taking a peek at the registrations and lookups of
certain companies (say Google, Time Warner, Disney, Microsoft, etc.)
those provide a much higher quality (and value) of information than a
registrar peeking at a lookup by a grandmother in South Dakota. (of
course the Googles and Time Warners of the world are probably a lot
more sophisticated to take appropriate counter-measures in how they do
availability checks, to ensure their good ideas are not taken by
others)

Sincerely,

George Kirikos
416-588-0269
http://www.leap.com/