Gurshabad,
In reviewing your feedback for technical issues, I identified an item to be
addressed. Quoting from your note:
The draft notes that, “the data verified by the VSP MUST be stored by
the VSP along with the generated verification code to address any
compliance issues.”
Registrants whose personal information will be shared in this way have
no control over these aspects, which negatively impacts their privacy.
In consideration of this comment, I plan to revise this sentence in the
Introduction:
“The data verified by the VSP MUST be stored by the VSP along with the
generated verification code to address any compliance issues.”
in a subsequent draft to be:
“The VSP MUST store the proof of verification and the generated verification
code; and MAY store the verified data.”
I will also add text the Security Considerations section to ensure that the
storage of the data that was verified by the VSP is stored securely and in line
with applicable privacy laws and regulations.
—
JG
James Gould
Distinguished Engineer
[email protected]
703-948-3271
12061 Bluemont Way
Reston, VA 20190
Verisign.com <http://verisigninc.com/>
On 10/3/18, 9:14 AM, "regext on behalf of Gould, James"
<[email protected] on behalf of [email protected]>
wrote:
Thanks for the review, Gurshabad. I'll consider your feedback in the
context of technical issues with the draft. The registration of domain names
in some jurisdictions may be subject to various requirements that involve
verification by a party other than the registry. The draft is intended for
interoperability and is independent of the verification process. The EPP
extension takes no position on specific verification processes which are a
"local matter" for the implementation. For these reasons, I disagree with your
assertion that implementation of the proposed extension will impact domain
registrants' right to freedom of expression, right to freedom of assembly and
association, and right to privacy. Rather the EPP extension, as the draft
describes, is a proposal for a standard framework for the common industry
practice of providing proof of verification to meet legal and/or security
requirements within a given top-level domain.
—
JG
James Gould
Distinguished Engineer
[email protected]
703-948-3271
12061 Bluemont Way
Reston, VA 20190
Verisign.com <http://verisigninc.com/>
On 9/20/18, 5:02 PM, "regext on behalf of Gurshabad Grover"
<[email protected] on behalf of [email protected]> wrote:
Hi all,
Firstly, thanks to James Gould and the regext working group for
developing draft-ietf-regext-verificationcode.
As part of efforts in the Human Rights Protocol Considerations (HRPC)
group, I have reviewed the human rights considerations (using RFC 8280
as a framework) in draft-ietf-regext-verificationcode-03. Please find
the review below. Hope that some of these concerns can be addressed
moving forward.
HR Review: Verification Code Extension for the EPP
==================================================
An assessment of human rights considerations in
draft-ietf-regext-verificationcode-03.
Introduction
------------
The ‘Verification Code Extension for the Extensible Provisioning
Protocol (EPP)’ ([VCEPP], hereinafter also referred to as the “draft”)
describes an extension in the EPP object mappings which supports adding
a verification code provided by a third-party known as the Verification
Service Provider (VSP). If the registrant’s data is not “verified” by
the VSP, it may be prohibited from requesting the execution of EPP
transform commands.
This is a review of the human rights considerations in the the draft.
(See [RFC 8280], ‘Research Into Human Rights Protocol Considerations’).
We believe that the draft does not make adequate considerations for
human rights: The implementation of the proposed extension will impact
domain registrants’ right to freedom of expression, right to freedom of
assembly and association, and right to privacy.
Human Rights Considerations
---------------------------
# Privacy ([RFC8280], section 6.2.2)
From [RFC8280]: “Could your protocol improve data minimization?” and
“Does your document identify potentially sensitive data logged [...]
and/or for how long that data needs to be retained for technical
reasons?”
The draft leaves the precise data shared with the VSP “up to the
policies of the locality” and outside the scope of the draft. In the
'domain-registrant' type of verification, both the registrant contact
information and domain name are sent to the VSP. The registrant contact
information is sensitive personal information, including name, address,
telephone number and email address. [RFC5733]
The draft notes that, “the data verified by the VSP MUST be stored by
the VSP along with the generated verification code to address any
compliance issues.” The information is retained and “may be accessed at
a later time.”
Registrants whose personal information will be shared in this way have
no control over these aspects, which negatively impacts their privacy.
# Content Agnosticism ([RFC8280], section 6.2.3)
From [RFC8280]: “Does your protocol make decisions based on the payload
[...]? Does your protocol prioritize certain content or services over
others [...]?”
Section 1 of the draft states that the “Verification Service Provider
(VSP) is a certified party to verify that data is in compliance with the
policies of a locality.” Therefore, the extension facilitates the
prioritization of certain individuals based on how the VSP judges
whether the registrants’ data is compliant. Registrants’ data which is
not in “compliance” with the local regulations will not receive a
verification code from the VSP, which gives the VSP the power to
restrict individuals from registering/modifying a domain name, and
limits their right to freedom of expression.
# Censorship Resistance ([RFC8280], section 6.2.6)
From [RFC8280]: “Can your protocol contribute to filtering in such a way
that it could be implemented to censor data or services? [...]”
By design, the extension facilitates filtering: the VSP will receive the
power to “verify” data on arbitrary grounds when determining whether
registrant data is in “compliance” with local regulations. The draft
also identifies examples of these scenarios: the VSP will check whether
“the domain name is not prohibited”, or whether the “registrant is a
valid individual, organization, or business in the locality.” On these
grounds, the VSP can accept or reject attempts to register domain names.
Thus, the extension explicitly provides filtering and censorship
abilities to the VSP, which are inimical to the registrants’ freedom of
expression.
# Open Standards ([RFC8280], section 6.2.7)
From [RFC8280], “Are you aware of any patents that would prevent your
standard from being fully implemented [RFC6701] [RFC8179]?”
There is an IPR declaration filed by Verisign Inc. for an older version
of the draft mentioning the relevant [PATENT]. The declaration
facilitates mostly open development, giving minimal assertion rights
over the license for Verisign Inc.
From [RFC8280]: “Is your protocol fully documented in such a way that it
could be easily implemented, improved, built upon, and/or further
developed?”
Several key details that will form the complete mechanism for the
verification code are not included in the draft (as it only describes
the extension), but have been offered in the [PATENT]. This includes a
description of the grace period within which “the requirement set of
verification codes may be sent before the object becomes non-compliant”,
and a clear depiction of the flow of the request detailed in Fig. 1 of
the [PATENT].
# Decentralisation ([RFC8280], section 6.2.14)
From [RFC8280]: “What is the potential for discrimination against users
of your protocol?”
There is immense potential for discrimination against registrants whose
data will be subject to “verification” by the VSP. The proposed
extension provides the VSP powers to discriminate against registrants
based on the information it receives, which may include names and
addresses, and which may be further compared to other “local data
sources” [VCEEPP] to indicate legal sex, religious affiliation, or
ethnicity.
From [RFC8280]: “Can your protocol be used to negatively implicate users
(e.g., incrimination, accusation)?”
There is potential for negative implication: a malicious registrant may
submit a request for a domain name considered legally “prohibited” by
the locality, accompanied with someone else's personal information.
Registrant information is sent to the VSP, an entity that may be
associated with law enforcement, which creates potential misuse for
false incrimination/accusation.
# Reliability ([RFC8280], section 6.2.14)
No fallback mechanism is specified within the draft for when the VSP is
offline or unavailable. In the event that the VSP is not available to
process the sent object, the registrants’ requests will not be
performed.
Similarly, when considering how long the VSP takes to process a
particular request, Section 0057 of the [PATENT] states that a “pending
compliance status indicator may indicate that the object is not
compliant by [...] a time, or grace period, in which the requirement set
of verification codes may be sent before the object becomes
non-compliant.” Therefore, if the VSP does not respond within the grace
period, the object will be considered “non-compliant”. The proposed
extension thus creates a problematic assumption of “non-compliance”,
which is at odds with the presumption of innocence.
# Confidentiality ([RFC8280], section 6.2.15)
From [RFC8280]: “What options exist for [...] implementers to choose to
limit the information shared with each entity?”
No such mechanisms are present which allow the registrants to limit the
information that will be accessed by the VSP.
# Integrity ([RFC8280], section 6.2.16)
From [RFC8280]: “Does your protocol maintain, assure, and/or verify the
accuracy of payload data? Does your protocol in any way allow the data
to be (intentionally or unintentionally) altered?”
The XML Signature maintains authenticity and integrity of the
verification code. However, we are not sure whether the object can be
altered by the VSP before appending the verification code to the shared
object. If so, it creates the potential for false implication (also see
section on Decentralisation) by the VSP.
# Outcome transparency ([RFC8280], section 6.2.19)
From [RFC8280]: “Are the effects [...] fully and easily comprehensible
[...]?”
As highlighted earlier in the review, many details (some out of the
scope of the draft) may be opaque to registrants with the implementation
of the proposed extension, including but not limited to, the exact data
shared with the VSP.
Additionally, as the [PATENT] notes, a “rejection of the verification
request may be transmitted [...]” by the VSP. Therefore, while the
proposed extension technologically facilitates a legal mechanism, there
are no built-in measures that facilitate human rights related to legal
transparency, remedy, or due process. For example, the VSP has no option
to append an explanation for sending back a rejection.
Conclusion
----------
In summary, the proposed extension in its implementation allows the VSP
to have enormous censorship and discriminatory power. The arbitrary and
opaque mechanism to “verify” compliance with local regulations gives
disproportionate powers to impede freedom of expression, right to
privacy, freedom of association, and the right to non-discrimination.
Some of the concerns highlighted in this review have already been
pointed out in the past in and outside the IETF. [NLLZMSG][ART-19] We
hope that the WG considers the potential adverse human rights impact
that the proposed extension creates before recommending its
standardisation. We understand that some of the concerns in the review
do not stem purely from the proposed extension, but due to the
unavoidable consequences of implementing the extension. The WG could
consider adding the details of those to the draft as well, so as to make
its impact on human rights clearer.
References
----------
[RFC8280] ten Oever, N., Cath, C., “Research into Human Rights Protocol
Considerations”, RFC 8280, October 2017,
<https://www.rfc-editor.org/info/rfc8280>.
[VCEEPP] Gould, J., “Verification Code Extension for the Extensible
Provisioning Protocol (EPP)”, draft-ietf-regext-verificationcode-03,
April 2018,
<https://datatracker.ietf.org/doc/draft-ietf-regext-verificationcode/>.
[IP-DECL] Ward, M., “Verisign Inc.'s Statement about IPR related to
draft-gould-eppext-verificationcode”, October 2015,
<https://datatracker.ietf.org/ipr/2703/>
[RFC5730] Hollenbeck, S., “Extensible Provisioning Protocol (EPP)”, RFC
5730, August 2009,
<https://www.rfc-editor.org/info/rfc5730>
[RFC5731] Hollenbeck, S., “Extensible Provisioning Protocol (EPP) Domain
Name Mapping”, RFC 5731, August 2009,
<https://www.rfc-editor.org/info/rfc5731>.
[RFC5732] Hollenbeck, S., “Extensible Provisioning Protocol (EPP) Host
Mapping”, RFC 5732, August 2009,
<https://www.rfc-editor.org/info/rfc5732>.
[RFC5733] Hollenbeck, S., “Extensible Provisioning Protocol (EPP)
Contact Mapping”, RFC 5733, August 2009,
<https://www.rfc-editor.org/info/rfc5733>.
[PATENT] Waldron, J., et al. “Domain Name Operation Verification Code
Generation and/or Verification”, United States Patent Serial No.
14/868,972, September 2015,
<https://patents.google.com/patent/US20170093793A1/en?oq=United+States+Patent+Serial+No.+14%2f868%2c972>.
[NLLZMSG] ten Oever, N., “Re: [regext] I-D Action:
draft-ietf-regext-verificationcode-01.txt”, IETF Mail Archive: regext,
April 2017,
<https://mailarchive.ietf.org/arch/msg/regext/EIK0E7ins874q8DmP37yg7AzAQI>.
[ART-19] Article 19 Digital, “Corporate actors must not facilitate human
rights violations through new Chinese rules”, Article 19, December 2016,
<https://www.article19.org/resources/corporate-actors-must-not-facilitate-human-rights-violations-through-new-chinese-rules/>.
Acknowledgements
----------------
Thanks to Paul Kurian for research inputs. Thanks to Amelia
Andersdotter, Mallory Knodel, Sunil Abraham and Swaraj Barooah for their
feedback. Please note that acknowledgement here does not necessarily
imply endorsement of the contents of the review.
--
Thank you.
Gurshabad
_______________________________________________
regext mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/regext
_______________________________________________
regext mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/regext
