[GNSO-RPM-WG] Q#15 Transparency for the TMCH

Sat Oct 12 21:56:57 UTC 2019

Michael, Rebecca, Jason, all

I think Marie and Greg have done a great job explaining the views of IP
owners, along with the nature and type of harms that are at risk.

I am supplementing their response to address some of the other issues or
questions
raised by Michael, Rebecca, and Jason, and to suggest a path forward that
may be capable of reaching consensus.

*The Rules of the Original TMCH Proposal Did Not Change During
Implementation to Create a "Secret" Database*

In terms of the 'original intent' question that Jason raised, I previously
served the ICANN committee (the Implementation Recommendation
Team) that developed the original proposal and recommendation for an
IP Clearinghouse (the name of which was later changed to TMCH); I can
assure you that the intent from the very beginning was for TMCH data
to remain confidential.

To demonstrate this point, the following language appears in the Final IRT
Report:

"Access to and use of such data must be restricted to trademark owners (who
will be permitted to access and use only their own data), ICANN, new gTLD
registries and registrars, for the sole purpose of performing the data
validation functions for new gTLD registries and the implementation of
RPMs, e.g., in relation to...[TM Claims etc.]. Ownership of any and all
data submitted to or generated by the IP Clearinghouse must remain
exclusively with the entity providing such data."

The concept of having a database where these assets are left out in the
open for third-parties, including criminals (or even competitors), to
access the proprietary data, would never have gotten off the ground within
the IP community to provide the support needed to obtain the adoption of
the TMCH by ICANN in the first instance.

The issues we are discussing now (regarding confidentiality of TMCH
data) are not new, and were identified and discussed during the
implementation phase of the new gTLD program.

But there was never a period of time when the "status quo" (as expressed in
the various draft Applicant Guidebook(s)) included an open TMCH where
trademark data was freely accessible. While these issue were contemplated,
there was never sufficient traction for this concept to be
incorporated into the official rule-making, so there wasn't a big switch
behind the scenes at the 11th hour.

*Distinction Between Searching National Trademark Office Databases and the
TMCH*

The TMCH is a proprietary database which houses some of the most valuable
and important assets that organizations own. This is the underlying basis
for the need of confidential treatment of the associated data elements.

There are now over 45,000 trademark records recorded in the TMCH; without
confidentiality, there would never be the current level of TMCH adoption by
the IP community.

If the TMCH data was not confidential, the targeted, aggregation of
this data could be used to identify jurisdictions in which the rights
holder may not have registered its trademarks or in which it has not chosen
to defensively register domain names. It would also expose gaps in IP
protection strategies because it would reflect which trademarks may be
considered more valuable than others in terms of online brand protection,
or on the other end of the spectrum, may expose competitive intelligence.

Since the TMCH is a unique and proprietary database, the risk exposure
relates to the aggregation of trademark data for a specific purpose, e.g.
to protect brands in the DNS through the TMCH. A similar level of
inference can not be gained by simply searching the national trademark
office databases.

And we know that having this level of inference about IP owners
online brand strategies can be leveraged to further criminal activity, such
as counterfeiting and DNS abuse (phishing attacks, etc.).

So these factors, which lay at the very foundational design of the
TMCH, are the primary distinctions between the trademark data that is
available in the various public searchable national trademark office
databases and the targeted, aggregated data held by the TMCH for the
purpose of online brand protection.

*Using TM Claims to Misuse and Mine TMCH Data*

I do not find the recent question that Michael raised concerning the misuse
of data access, e.g. "data mining", to be a valid argument for
transparency.

To be clear, I am not saying there are not valid arguments for
transparency, but I am referring to the practice of mining the
database for the purpose of obtaining confidential data.

Accordingly, the TMCH should be structured in a manner that minimizes data
mining of trademark registration patterns. I haven't dug into the contracts
as of yet, but there should be contractual safeguards (either within the
Registrar Accreditation Agreement (RAA), or the Registry Agreement
(RA)), to ensure that TMCH data is not mined and to provide
enforcement against abuse of access.

The justification of confidentiality is based on the concept of minimizing
abuse of data, so that the distribution of data is limited to situations
where the data is needed to implement TMCH functionality, including
for technical reasons, e.g. uptime/availability. The WG should consider
putting forward a recommendation if sufficient enforceable safeguards
against data mining do not already exist.

*Paradigm for Building Consensus*

Based on the concerns that Rebecca, Michael, and Jason have articulated, I
believe the appropriate venue to address those concerns are the dispute
resolution procedures at the Registry-level, including the Sunrise Dispute
Resolution Policy (SDRP).

For example, Greg has recently submitted a proposal (in relation to the
design marks topic) that sets forth a new ground, (e.g. the trademark mark
was obtained in a pre-textual manner) upon which to bring a complaint under
the SDRP policy.

On our last call I extended an olive branch to Michael to work on
developing additional solutions. Rebecca and I recently successfully
engaged in such an exercise to help build consensus on the TMCH.

I'm not sure if it helps you feel any better, but my previous proposal to
provide transparency on Reserved Names, was not adopted because of a
similar confidentiality issue. So in conclusion, I encourage you to accept
my offer to work together to develop solutions that have a better potential
to reach consensus.

Cheers,
Claudio

On Fri, Oct 11, 2019 at 5:36 PM Paul Keating <paul at law.es> wrote:

> Marie,
>
> Regarding defensible arguments:
>
> Because while it's very easy to find out whether a particular word (or a
> dataset of 50 or 100 words) is in the TMCH, it's impossible to see the
> totality of words included in the TMCH. *This prevents effective
> oversight - as we've run into time and again in discussing the extent of
> registrations of GIs, the extent of registrations of design marks, etc*.
> It's only easy to search when you know specifically what you're looking for.
>
>
> Sent while on the run.
>
> On 11 Oct 2019, at 20:37, Marie Pattullo <marie.pattullo at aim.be> wrote:
>
> I appreciate the replies, but have laid out a straightforward, fact-based
> "defensible and realistic argument for the harm that would result from
> transparency", while I don't believe I've seen fact-based arguments as to
> the purported harm that transparency would solve.
> I wish you all a great evening,
> Marie
>
> -----Original Message-----
> From: Michael Karanicolas <mkaranicolas at gmail.com>
> Sent: Friday, October 11, 2019 9:15 PM
> To: Marie Pattullo <marie.pattullo at aim.be>
> Cc: Jason Schaeffer <jason at esqwire.com>; gnso-rpm-wg at icann.org
> Subject: Re: [GNSO-RPM-WG] Q#15 Transparency for the TMCH
>
> Because while it's very easy to find out whether a particular word (or a
> dataset of 50 or 100 words) is in the TMCH, it's impossible to see the
> totality of words included in the TMCH. This prevents effective oversight -
> as we've run into time and again in discussing the extent of registrations
> of GIs, the extent of registrations of design marks, etc. It's only easy to
> search when you know specifically what you're looking for.
>
> And while I understand your perspective is different... I feel like at
> some point you need to make a defensible and realistic argument for the
> harm that would result from transparency, since the brand strategy
> information you mention is already, essentially, publicly available.
>
> On Fri, Oct 11, 2019 at 3:01 PM Marie Pattullo <marie.pattullo at aim.be>
> wrote:
> >
> > Thanks, Michael.
> > If it is as easy as you say, then what purpose would a public TMCH serve?
> > As for your suspicion, while of course I cannot speak for you, I can for
> me: that it absolutely not the case.
> > And Paul, at the risk of simply quoting you: nothing in your comments
> moves the needle of the opposition for us either. That's the joy of a
> debate between people with different perspectives.
> > Best to all,
> > Marie
> >
> > -----Original Message-----
> > From: Michael Karanicolas <mkaranicolas at gmail.com>
> > Sent: Friday, October 11, 2019 8:49 PM
> > To: Marie Pattullo <marie.pattullo at aim.be>
> > Cc: Jason Schaeffer <jason at esqwire.com>; gnso-rpm-wg at icann.org
> > Subject: Re: [GNSO-RPM-WG] Q#15 Transparency for the TMCH
> >
> > Hi Marie,
> >
> > The problem with the claim of commercial harm as you frame it here is
> that it is trivially easy to find out which marks from a particular
> portfolio are registered in the TMCH through trial and error. As you
> mention - the totality of the portfolio is publicly available. So if you
> have that dataset, it's pretty simple to just search for each word and see
> when a trademark claim notification pops up (indeed - this is how
> journalists found all those generic words). So any interested party who
> wanted to find out about a company's "commercial/enforcement strategy"
> could just spend an hour or two searching for each registered mark, and get
> the complete picture.
> >
> > I think that's the reason why Jason painted the elaborate scenario he
> did, is because many of us are wracking our brains trying to find out how
> there's a legitimate confidentiality interest when this information is
> available to anyone who cares to look. It's good to know that the scenario
> he mentioned hasn't come up... but if that's the case it begs the question
> as to what the basis of your opposition is, given that the strategic
> information you refer to is already available to anyone who spends a couple
> of hours looking?
> >
> > Because I have to admit, I'm growing increasingly suspicious that the
> argument to keep it secret is not rooted in any actual need for
> confidentiality, but rather in a desire to avoid the kind of public
> oversight which we are advocating.
> >
> > Best,
> >
> > Michael
> >
> > On Fri, Oct 11, 2019 at 2:25 PM Marie Pattullo <marie.pattullo at aim.be>
> wrote:
> > >
> > > Thanks Jason,
> > >
> > >
> > >
> > > I can only speak for myself but no, this isn’t our main concern and I
> apologise if I was unclear.
> > >
> > >
> > >
> > > A branded goods company may own many trade marks – some, 100s (e.g. in
> the FMCG sector), some, one or a handful (SMEs, start-ups). In the EU, it’s
> pretty straightforward to find out what TMs are registered and by whom –
> indeed, it’s basic IP strategy to conduct a search of TM Registers to find
> out whether the TM you want to have is available, as no-one wants to face
> needless oppositions, a wasted marketing budget etc. So, as many in the WG
> have noted, if you want to find out which marks a right holder owns, you
> can.
> > >
> > >
> > >
> > > The TMCH is a totally different thing. A brand that chooses to (pay
> to) put a TM into the TMCH does so for either commercial reasons (it may
> apply for a registration in a new gTLD) or enforcement reasons (to have
> notice if someone else is trying to register such a name). Brands with
> multiple TMs are unlikely to put all of them into the TMCH, so disclosing
> which ones it has chosen to put there also discloses its
> commercial/enforcement strategy – i.e. the TMs about which it is most
> concerned in the new gTLD space of the DNS.
> > >
> > >
> > >
> > > It may not want to announce that to the market for a variety of
> reasons, either to its competitors (who will then know which of its 100s of
> TMs it considers most important in the new gTLD space) or to
> registries/registrars (we all remember some of the “premium pricing”
> concerns in the new round).
> > >
> > >
> > >
> > > As for enforcement, I stress that the TMCH only gives the brand the
> possibility to be told if someone wants a DN including their TM: it clearly
> does not result in an automatic block on such a registration, for which
> there could of course be multiple valid reasons – not least different class
> and/or no confusion. However in the case of a malicious/bad faith (etc.)
> registration then yes, the brand would be on notice and could then choose
> to (pay to) take appropriate enforcement action, including (if appropriate)
> following the URS route. The brands that I know are not in the business of
> launching enforcement actions they know are unfounded; they’re in the
> business of making products that consumers trust and choose to buy.
> > >
> > >
> > >
> > > I can’t comment on your scenario below as that has never been brought
> to me as a concern; the above has. And we fail to see any benefit in the
> TMCH being public.
> > >
> > >
> > >
> > > I hope this is clear, and again apologise if I was not before.
> > >
> > >
> > >
> > > Marie
> > >
> > >
> > >
> > >
> > >
> > > From: GNSO-RPM-WG <gnso-rpm-wg-bounces at icann.org> On Behalf Of Jason
> > > Schaeffer
> > > Sent: Friday, October 11, 2019 6:52 PM
> > > To: gnso-rpm-wg at icann.org
> > > Subject: [GNSO-RPM-WG] Q#15 Transparency for the TMCH
> > >
> > >
> > >
> > > Hi All,
> > >
> > >
> > >
> > > Sorry for not being able to address this on yesterday’s call – I was
> bounced from the Zoom and couldn’t reconnect from my mobile. After much
> consideration of the chat comments and statements during the past two calls
> I’ve parsed together the following strategy concerns that might be at
> issue. It seems the opposition to an open/transparent TMCH have set forth
> the following rationale:
> > >
> > >
> > >
> > > Large brand owners like the ability to file in “remote” jurisdictions
> with non-searchable DBs to gain a priority date that can later be used for
> priority presumably in the US under a Section 44 application under the
> Paris Convention. For example, Party A could “secretly” register in say
> Mauritius or Jamaica (jurisdictions that I believe do not have searchable
> DBs), and then file a Section 44 with the USPTO when ready to “go public.”
> In this scenario, large brand owner obtains the earlier priority date in
> the US without having disclosed its “secret” brand plans or strategy.
> Alternatively, there may also be a similar use case with a registration in
> the EU. This is what I understand the opposition was presenting in the
> comments last week and again yesterday.
> > >
> > >
> > >
> > > This use of the “secret TM priority” filing method is one thing, and I
> understand the ostensible business intelligence concerns to require such a
> tactic. However, we are discussing this in the context of the TMCH. How can
> these parties obtain Sunrise protection without proof of use? This is even
> more challenging in the case of an EU registration that might not be based
> on actual use.
> > >
> > >
> > >
> > > If the project is top secret and subject to heightened secrecy, how
> can the party then show legitimate use of the mark and obtain TMCH
> protection while maintaining its purported heightened secrecy? What is the
> POU and what is the declarant stating to the TMCH? Is this the position
> that the opposition is proffering to block transparency of the TMCH?
> > >
> > >
> > >
> > > In addition, if I am correctly capturing how the method is employed in
> practice, it would be helpful to know if this practice is really in
> widespread use or rather an outlier, because I’m not yet seeing an
> important justification that outweighs the important benefits of an open
> TMCH.
> > >
> > >
> > >
> > > Thanks,
> > >
> > >
> > >
> > > Jason
> > >
> > >
> > >
> > > _______________________________________________
> > > GNSO-RPM-WG mailing list
> > > GNSO-RPM-WG at icann.org
> > > https://mm.icann.org/mailman/listinfo/gnso-rpm-wg
> > > _______________________________________________
> > > By submitting your personal data, you consent to the processing of
> your personal data for purposes of subscribing to this mailing list
> accordance with the ICANN Privacy Policy (
> https://www.icann.org/privacy/policy) and the website Terms of Service (
> https://www.icann.org/privacy/tos). You can visit the Mailman link above
> to change your membership status or configuration, including unsubscribing,
> setting digest-style delivery or disabling delivery altogether (e.g., for a
> vacation), and so on.
> _______________________________________________
> GNSO-RPM-WG mailing list
> GNSO-RPM-WG at icann.org
> https://mm.icann.org/mailman/listinfo/gnso-rpm-wg
> _______________________________________________
> By submitting your personal data, you consent to the processing of your
> personal data for purposes of subscribing to this mailing list accordance
> with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and
> the website Terms of Service (https://www.icann.org/privacy/tos). You can
> visit the Mailman link above to change your membership status or
> configuration, including unsubscribing, setting digest-style delivery or
> disabling delivery altogether (e.g., for a vacation), and so on.
>
> _______________________________________________
> GNSO-RPM-WG mailing list
> GNSO-RPM-WG at icann.org
> https://mm.icann.org/mailman/listinfo/gnso-rpm-wg
> _______________________________________________
> By submitting your personal data, you consent to the processing of your
> personal data for purposes of subscribing to this mailing list accordance
> with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and
> the website Terms of Service (https://www.icann.org/privacy/tos). You can
> visit the Mailman link above to change your membership status or
> configuration, including unsubscribing, setting digest-style delivery or
> disabling delivery altogether (e.g., for a vacation), and so on.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/gnso-rpm-wg/attachments/20191012/053f6d10/attachment-0001.html>