[gnso-rpm-wg] Inferences (was Re: Mp3, Attendance, AC recording & AC Chat Review of all Rights Protection Mechanisms (RPMs) PDP Working Group)

[George Kirikos]

Hi again,

Just to followup on my prior email, The Analysis Group *already*
adjusted the stats to attempt to take into account the "mining" theory
that Jeff spoke about. i.e. the 93.7% abandonment figure that we've
been talking about is a figure obtained *after* making the
adjustments! Without the adjustment (which eliminated 62.2% of the
observations), the abandonment rate would have been 99%! See:

https://community.icann.org/download/attachments/64066042/Analysis%20Group%20Revised%20TMCH%20Report%20-%20March%202017.pdf?version=1&modificationDate=1490349029000&api=v2

(a) page 17 (footnote 55)

""As discussed in Section IV, there are two registrars that averaged
downloads of more than 20 trademark strings per download, which is
large compared to the average of fewer than five trademark strings in
the downloads of other registrars. We also exclude downloads made by
ICANN’s monitoring system. The exclusion of the two registrars does
not significantly impact our results. Inclusion of the two registrars
shows that 99% of registrations are abandoned and 0.5% of completed
registrations are disputed."

(b) page 18, note [2]

"[2] A bulk download is defined as a download from the TMCH of
multiple strings by the same registrar with exactly the same time
stamp. Downloads by two registrars are excluded from this analysis
because of a potentially high prevalence of bulk downloads (98.7% and
81.9% of downloads, respectively) by each of these two registrars. The
average size of the “bulk downloads” by these two registrars
(approximately 23 and 35 strings, respectively) is much larger than
the average “bulk download” size of other registrars (other registrars
in the Claims Service data download 5 strings or less on average).
This exclusion results in an exclusion of 62.2% of the observations in
the original Claims Service data received from IBM after excluding
downloads by ICANN's monitoring system."

Sincerely,

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



On Wed, Jul 12, 2017 at 4:10 PM, George Kirikos <icann at leap.com> wrote:
> Hi folks,
>
> Just following up on some statements from today's transcript:
>
> On Wed, Jul 12, 2017 at 3:51 PM, Terri Agnew <terri.agnew at icann.org> wrote:
>> Adobe Connect chat transcript for 12 July 2017:
> ...
>>   Jeff Neuman:My belief is that we have a huge rate of people abandoning is
>> because (i) registrars were mining the system, (ii) registrants were mining
>> the system to see what was valuable, and to a lesser extent as a result of
>> the claim (either legitimiate or not)
>>
>>   Jeff Neuman:But I cant prove any of those theories
>>
>>   Kathy Kleiman:@Jeff: we have gathered evidence already from registries;
>> there is probably more
>>
>>   Jeff Neuman:There just is no way to do so on a backwards basis
>
> If one had access to the raw data (and presumably The Analysis Group
> would have had it, to generate their reports), one could filter such
> "mining" by examining abandonment rates by (1) registrar and (2) by
> time relative to the launch date.
>
> i.e. presumably those were "mining" the system were not spreading out
> their queries across all registrars equally. It would be easy to
> identify the outliers. Furthermore, registrants would be also focused
> on a few registrars that permit those bulk lookups, and they wouldn't
> spread their queries over time equally --- they'd be focused at the
> launch (i.e. the beginning, first few days, etc.) of a TLD.
>
> Of course, if need be, one could anonymize the data by registrar (i.e.
> Registrar 1, Registrar 2, etc.), if there are any concerns about
> revealing their individual abandonment rates.
>
> Registry operators and registrars both have technology to block WHOIS
> access to those who are "mining" that data. One can use similar
> detection techniques in this instance.
>
> Sincerely,
>
> George Kirikos
> 416-588-0269
> http://www.leap.com/
> _______________________________________________
> gnso-rpm-wg mailing list
> gnso-rpm-wg at icann.org
> https://mm.icann.org/mailman/listinfo/gnso-rpm-wg