[gnso-rpm-wg] Directly from INTA's website: What the TTAB has to say about sample size

[George Kirikos]

Hi Paul,

On Fri, Sep 1, 2017 at 8:57 PM, icannlists <icannlists at winston.com> wrote:
> George, once again you leave me befuddled. You are insisting that the INTA survey data be up to your high standards (which standards you assert through vague references to your personal opinions about statistical theory but no citations to any recognized authority) or else the survey be "sent to the trashcan", while at the same time insisting that your opinions - which you admit have no surveys attached to them at all- be given full weight. You really do need to make up your mind about the importance of surveys in order to support positions in this WG. You can't have it both ways.

I didn't rely *just* on my "personal opinions" about statistical
theory. Go read the first post of this thread again, or read the
subject of all the emails in this thread:

http://mm.icann.org/pipermail/gnso-rpm-wg/2017-August/002438.html

My citation is published work from INTA's own journal! That should be
enough. If you want more, go to Google and search for articles on
survey size. Obviously the judges in the TTAB cases (all public and
cited in the article) would have had that basic statistics
knowledge/evidence/authorities when they made their
rulings/conclusion. Go and download the cited cases, and read them at
your leisure.

Or are you asserting that INTA made a mistake when they allowed that
article talking about sample size to be included in their very own
journal?

In the first email of the prior thread:

http://mm.icann.org/pipermail/gnso-rpm-wg/2017-August/002408.html

Confidence interval calculations could be done easily from the
supplied links. They're standard calculations (Kurt even calculated
the 18% figure in his followup email).

Raise your hand if you've been published in a peer-reviewed journal,
like I have, in relation to statistics/econometrics/quantitative
finance?

http://www-2.rotman.utoronto.ca/~kan/research.htm
http://www-2.rotman.utoronto.ca/~kan/papers/mom1.pdf
http://www.powerfinance.com/convexity/ [no "stats" in this one, but it
will give you a sense of the depth of my math knowledge; my company
does own Math.com, remember?]

and completed Ph.D. level courses in those areas? (I never finished my
dissertation, as I got too distracted with making money outside
academia)

That doesn't make me an "authority" in statistics, but anyone with an
iota of statistics education and objectivity should be able to see
through this INTA study. I'm sure the lawyers in those TTAB cases
argued long and hard, maybe even paying "experts" or "authorities" to
attempt to assert that their small surveys had validity -- they lost.

I'm not "trying to have it both ways". If someone uses a survey as the
only foundation that supports their position -- it better be a
statistically valid survey, if that's all they are relying upon to
support their position.

However, some positions (depending on the nature of the question at
hand) don't require a "survey" at all, to be a positive contribution.
I already gave 2 examples in the prior email (the ETRP, and the tiered
pricing issue).

In another PDP (the IGO one), I didn't need to "take a survey" to
debunk the position that some IGOs were taking, that they were unable
to use the UDRP (because of the mutual jurisdiction clause, which
could affect their claimed immunity). Do you know what won the day?
Going out and doing research, and actually finding examples of UDRPs
where IGOs expressly participated! Same for their claimed "inability
to waive immunity" -- it's kind of funny how that argument was
debunked, after a little bit of research brought to light examples of
the World Bank (an IGO) filing lawsuits in US court. I could go on,
and on, but if you want other examples, feel free to email me outside
this list.

One doesn't need to "take a survey" to debunk the INTA study, because
of its inherent flaws. Nor is one required to support my position on
that topic. All of math, statistics, and INTA's own publication in its
journal (which I expressly cited) about survey size support that
position.

What scientific citations have you introduced that support your strong
belief that this INTA study should be given any weight, other than the
study being supportive of your own position?

Sincerely,

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