Happy new year to you all. We have kicked off 2018 by continuing our work on data protection and privacy issues. In particular, we are preparing for the 25 May 2018 enforcement date for the European Union's General Data Protection Regulation (GDPR). As I've previously written, we are working to ensure compliance with this law while maintaining access to WHOIS to the greatest extent possible.
Our work in this area began when we asked the community for user cases and created a matrix of these different use cases about the personal data that gTLD registries and registrars collect, transmit or publish pursuant to ICANN agreements or policies. In the absence of a WHOIS policy, the user stories are essential for describing the many different uses of the WHOIS system. The matrix informed discussions about whether there were potential compliance issues under ICANN's agreements with registries and registrars because of the new law, and aided in our engagement with data protection authorities.
On 2 November 2017, we published a Statement from Contractual Compliance, which indicated ICANN org would defer taking compliance action against any registry or registrar for noncompliance with contractual obligations related to the handling of registration data. To be eligible for this deferral, we asked ICANN's contracted parties and stakeholders to follow this process to submit proposed interim models for compliance. We've published those community-proposed models here.
In parallel, we engaged the European law firm Hamilton to provide its legal analysis of these issues. The three-part assessment found in its first memo [PDF, 253 KB] that the WHOIS service in its current form must change. In the second part [PDF, 577 KB], Hamilton answered community questions about the law's applicability and scope. In its third analysis [PDF, 440 KB], Hamilton described how processing data within the scope of WHOIS could be changed to become compliant with the GDPR. We asked for your feedback on these analyses and published your input here.
In December, I wrote that we were working to develop interim models for collecting registration data and implementing registration directory services that may be compliant with both the law and ICANN's contractual agreements. To be clear, these proposed models are meant to facilitate discussion and a final model decided on to be an interim solution. They do not replace any existing ICANN policy development work or policies.
Today we published [PDF, 624 KB] for community input those three proposed discussion models for collecting registration data and implementing registration directory services. These models reflect discussions from across the community and with data protection authorities, legal analyses and the proposed models we have received to date. Please provide your input on these models. The input from the community will contribute to assessing the viability of each of the models. From that input either variations or modifications to one of these models will be identified at the end of January for the path forward. To help inform this, please provide your feedback by 29 January 2018. Please send your feedback to email@example.com.
The three models are summarized at a high-level below. The models differ based on what contact information is displayed in the public-facing WHOIS, their applicability, the duration of data retention and what data is not displayed in a public-facing WHOIS:
- Model 1 would allow for the display of Thick registration data, with the exception of the registrant's phone number and email address, and the name and postal address of the technical and administrative contacts. To gain access to these non-public data points, third parties would be required to self-certify their legitimate interests for accessing the data. This model applies if the registrant is a natural person, and the registrant, registry, registrar and/or the data processor is in the European Economic Area.
- Model 2 would allow for the display of Thin registration data, as well as the technical and administrative contacts' email addresses. To access the non-public information registries and registrars would be required to provide access only for a defined set of third-party requestors certified under a formal accreditation/certification program. There are two variations on how this model would apply. Model 2A applies to registrants who are both natural and legal persons, where the registrant, registry, registrar and/or the data processor is in the European Economic Area. Model 2B would apply to registrants who are both natural and legal persons, where the registrant, registry, registrar and/or the data processor is regardless of location, that is on a global basis.
- Model 3 would allow for the display of Thin registration data and any other non-personal registration data. To access non-public information, a requestor would provide a subpoena or other order from a court or other judicial tribunal of competent jurisdiction. This model would apply to all registrations on a global basis.
Please click here to see the models [PDF, 624 KB].
We will share these models as we continue our engagement work, including with the Article 29 Working Party.
As always, we'll continue to keep the community apprised of the various discussions we have. We've also received a range of correspondence relating to the GDPR. We urge you to visit our data protection/privacy page to view the latest correspondence, proposed models from the community, and other materials relevant to this discussion.
Happy 2018 and we look forward to all the work with the community over the coming year.