[gnso-rds-pdp-wg] Dangers of public whois
John Horton
john.horton at legitscript.com
Tue Feb 14 13:59:57 UTC 2017
Nathalie and others,
I wanted to take a moment and explain why I'm strongly opposed to requiring
email or other registration in order to view thin or thick details. For the
reasons outlined below, I think it's antithetical to the open and
decentralized nature of the internet, and constitutes a form of internet
surveillance.
First, putting aside repressive regimes, private networks and edge cases,
one of the hallmark principles of the internet is that it's open; you don't
have to register or justify your need to access information on the
internet. And, it's decentralized. Historically, its open nature has
included not only being able to see a website, but also the registration
details for the website's domain name. And, whatever governments may do
(which isn't the question here), there's no centralized internet
surveillance or registration authority for internet users generally.
If we impose a scheme where there is a central organization with the
authority to a) require registration and b) centrally control access, and
c) (as has been proposed) require the user to provide a reason for their
access, that organization then also has the ability to d) make judgment
calls about what reasons are valid and which are not and e) maintain data
on who accessed what RDS data, for what reason, for how long and why. Note
also that at least one version of the EWG report said that f) the
organization would be empowered to levy punitive measures against internet
users who accessed more data than the RDS deems appropriate.
So: you have a system that surveils internet users who access some
information and maintains data on their use of that data. Let's think about
the following scenarios from the point of view of openness,
decentralization and civil liberties.
- A journalist (or blogger) is writing an investigative article and
wants to find out who is behind a domain name. If we require registration
and disclosure of the reason, that in essence creates a situation where the
RDS de facto is monitoring that journalist and determining if their basis
for conducting the investigation is worthy. It also allows the RDS the
ability to monitor the journalist's use of the domain name registration
data. This potentially chills free speech.
- Consider a political activist who wishes to expose corruption by an
elected politician and wants to access RDS information to show, for
example, conflicts of interests in the politician's business operations.
Once the political activist has to disclose who they are, let alone why
they are accessing the information, that not only chills legitimate
political activism but also potentially opens up a route for government
abuse (e.g., if a government agency were able to subpoena the list of who
accessed RDS information for which domain names and why).
- Academic researchers periodically review Whois/RDS data; requiring
them to register before reviewing data and disclose why they are doing the
research potentially empowers the RDS to monitor academic research and
determine its worthiness.
- Imagine that a cybercrime network is under investigation (as they are
wont to be); requiring law enforcement to register -- particularly if there
is a log of which domain names they reviewed RDS for -- can potentially
compromise the investigation if that information is disclosed. Would
registrants have the right to be informed every time that someone
registered to review their RDS details?
For one central entity to possess that much power over internet users is
something that I think we should avoid, and it's antithetical to the
principles of openness and decentralization. There are other well-known
solutions to spam and inappropriate contacts; forcing all other legitimate
activities to grind to a screeching halt -- particular under the umbrella
of a surveillance scheme -- is a cure worse than the disease.
I recognize and agree that we should try to find constructive solutions to
this that require some compromise, and I'm grateful not only for the
expertise that Stephanie and others have brought to this group, but also
that Benny and others have pointed out some of the problems with Whois
details being inappropriately used (e.g., for spam). However, I wanted to
outline my strong concerns about creating a centralized registration and
surveillance scheme over one subset of internet users as part of the
solutions.
John Horton
President and CEO, LegitScript
*Follow LegitScript*: LinkedIn
<http://www.linkedin.com/company/legitscript-com> | Facebook
<https://www.facebook.com/LegitScript> | Twitter
<https://twitter.com/legitscript> | *Blog <http://blog.legitscript.com>*
| Google+ <https://plus.google.com/112436813474708014933/posts>
On Tue, Feb 14, 2017 at 4:10 AM, nathalie coupet via gnso-rds-pdp-wg <
gnso-rds-pdp-wg at icann.org> wrote:
> Hi Allison,
>
> Would you be able to carry out your investigations normally if access to
> WHOIS thick were restricted only by the need to enter an email?
>
> With regards to privacy by design, instead of pushing for the
> implementation of this concept inside the realm of WHOIS where it is
> foreign, since it is an engineering concept, why not advocate for its
> implementation at the design level of the Internet, where it belongs?
>
>
> Nathalie
>
>
> On Tuesday, February 14, 2017 12:38 AM, allison nixon <elsakoo at gmail.com>
> wrote:
>
>
> This car metaphor isn't complete without also stating that some car owners
> purchase them for the sole purpose of running over people!
>
> Some car owners purchase fleets of cars to run over as many people as
> possible. Even though they re-use their name on every single vehicle
> registration, the subpeona takes so long that the city can no longer
> automatically block the cars as they enter, and need to wait for them to
> run over a few people before they can do anything about it.
>
> This metaphor has obviously been tortured past the point of absurdity,
> I'll leave it alone now.
>
> I've mostly been lurking for the whole duration of this group, and please
> forgive me if I'm missing something massive here, but I get the impression
> that most people here don't spend a lot of time doing investigations. But
> this is my life. If I needed a subpeona for every single historical lookup,
> pivot, and reverse search, I would get zero done due to a lack of legal
> authority. Many if not most of the people doing the heavy lifting in
> anti-cybercrime efforts are private citizens with no government issued
> authority. It seems that the general expectation here is that limiting
> access to people with badges is OK, but I'm telling you there is a severe
> lack of those skillsets and it will be years before we see widespread
> technical literacy among the police. Whatever system results, private
> citizens need a path for unrestricted and automated access. And if we want
> to talk protecting privacy, I think criminally motivated violations of
> privacy are far more likely to affect everyone's day to day life right now,
> and automated WHOIS lookups are used heavily especially in anti-phishing
> and anti-spam operations.
>
> With the status quo, I can go on fishing expeditions through the WHOIS
> data and turn up hundreds of domains used for the same type of malicious
> activity, and predict with a high accuracy which domains will be malicious
> before they are used for anything. It sometimes turns up domains owned by
> innocent people, and I doubt privacy minded people would like that, but the
> reality is I rarely ever encounter WHOIS data that is convincing PII. It's
> almost all fake. And if it's not fake, it's a company's public contact
> info, or it's a foolish person who turned down WHOIS privacy protection,
> and will change their WHOIS as soon as the spam starts flowing.
>
> Have there been any studies on what percentage of WHOIS data is real and
> correct? Can we ever expect to have meaningful data when registrars are
> allowed to take Bitcoins over Tor as payment? At what point does "privacy"
> become an empty argument when some of these Internet hosting/registrar
> companies clearly profit from facilitating abuse, and network defenders
> block entire TLDs due to the saturation of abuse?
>
> From my vantage point, I see great benefit from seeing patterns in the
> fake data submitted by fraudsters, and I see few harms from the privacy
> side of things, because people seem to generally realize that "123 fake st"
> is a perfectly acceptable WHOIS entry.
>
> I also recognize this situation is completely absurd. Every aspect of this
> is surely an abuse of the original system. But it seems like building a
> pyramid from the top down, restricting access to supposed "PII" that is
> unlikely to contain PII, to the detriment of legitimate efforts that also
> seek to enhance privacy by preventing criminal theft of private data like
> bank account numbers.
>
>
> On Mon, Feb 13, 2017 at 9:14 PM, Sam Lanfranco <sam at lanfranco.net> wrote:
>
> I have to strongly agree with Alex that whatever the criteria are for thin
> data, they cannot include that thin data "is transitive" in some sort of
> bread crumb trail manner.
>
> Everything is potentially transitive in that sense. I observe a vehicle
> but all I get is make, model and license plate, and in most jurisdictions
> that is all I get. It is the vehicle owner's "thin data". Of course I can
> hang around, see that the car has a baby seat, witness a woman or man
> putting a child in the car, assume that she/he has legitimate access to the
> car, follow the car and assemble more personal information (lives at; works
> at; shops at; visits;) The license plate didn't facilitate that crumb train
> discovery, but no license plate would hamper legitimate seeking of
> information about who owns the car (issuing a parking ticket, LEA
> investigation, etc.) . License plate is part of thin data with no gated
> access. Of course, this will change in the era of the digital vehicle.
> Depending on security, and authorization, one will be able to just ask the
> car, and ask about a lot of things...like whose cell phone was in the
> passenger's seat last night, when I was supposed to be alone )-:
>
> There needs to be a similar balance (license plate but no owner's name
> unless wanted, like Sam's Curry Pizza Barn logo, phone number and website
> URL painted on the side).
>
> More Important, have we made progress (convergence) on the working
> principles that should be brought to bear in building a thin data set. A
> lot of time has been spent looking at good case and bad case scenarios.
> What operational principles have been distilled from all these examples?
> What is the balance between thin data inclusion and exclusion, and design
> and technical solutions that can be used to prevent (for example) robotic
> harvesting? There is another frontier here, and that is what governments
> will do to restrain or enable certain uses of thin data? While ICANN needs
> to be aware of what is going on there, that part is beyond ICANN's remit,
> but those policies will help shape some of the context within which ICANN
> deals with the thin data task.
>
> Sam L
>
>
> On 2017-02-14 1:23 AM, Deacon, Alex wrote:
>
> All,
>
> So it seems the debate has progressed from “thin data” to “thick data”
> (i.e. data that includes email). I know we are all super excited to talk
> about “thick data” but I don’t think we are there yet (are we? Hopefully I
> didn’t miss the party…)
>
> Focusing on thin data for the moment I struggle to understand how it is
> personal data. I do not believe it is. As for the odd logic proposed by
> some that the property of privacy is transitive (i.e. Because “thin data”
> can be used to link/point/discover other data then “thin data” equals
> “personal data”) I just don’t buy it.
>
> I don’t disagree with much of what was expressed in this thread, however
> we must keep in mind that balance and proportionality are important
> concepts in many (all?) data privacy laws. Any arguments that imply that
> no such balance exists (or should exist) is obstructive IMO.
>
> Alex
>
>
> On 2/13/17, 5:42 AM, <gnso-rds-pdp-wg-bounces at icann .org
> <gnso-rds-pdp-wg-bounces at icann.org> on behalf of michele at blacknight.com>
> wrote:
>
> I agree and I know from how I’ve used various email addresses that
> they are actively being harvested and spammed.
> Also it’s one of the biggest sources of complaints we get from
> our clients (registrants)
> It’s definitely not an “edge case”.
> Regards
> Michele
> --
> Mr Michele Neylon
> Blacknight Solutions
> Hosting, Colocation & Domains
> https://www.blacknight.com/
> http://blacknight.blog/
> Intl. +353 (0) 59 9183072
> Direct Dial: +353 (0)59 9183090
> Social: http://mneylon.social
> Some thoughts: http://ceo.hosting/
> ----------------------------- --
> Blacknight Internet Solutions Ltd, Unit 12A,Barrowside Business
> Park,Sleaty
> Road,Graiguecullen,Carlow,R93 X265,Ireland Company No.: 370845
> ______________________________ _________________
> gnso-rds-pdp-wg mailing list
> gnso-rds-pdp-wg at icann.org
> https://mm.icann.org/mailman/ listinfo/gnso-rds-pdp-wg
> <https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg>
>
> ______________________________ _________________
> gnso-rds-pdp-wg mailing list
> gnso-rds-pdp-wg at icann.org
> https://mm.icann.org/mailman/l istinfo/gnso-rds-pdp-wg
> <https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg>
>
>
> --
> *----------------------------- ---------------*
> "It is a disgrace to be rich and honoured
> in an unjust state" -Confucius
> ------------------------------ ----------------
> Dr Sam Lanfranco (Prof Emeritus & Senior Scholar)
> Econ, York U., Toronto, Ontario, CANADA - M3J 1P3
> YorkU email: Lanfran at Yorku.ca Skype: slanfranco
> blog: http://samlanfranco.blogspot.c om
> <http://samlanfranco.blogspot.com/>
> Phone: 613 476-0429 cell: 416-816-2852
>
>
> ______________________________ _________________
> gnso-rds-pdp-wg mailing list
> gnso-rds-pdp-wg at icann.org
> https://mm.icann.org/mailman/l istinfo/gnso-rds-pdp-wg
> <https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg>
>
>
>
>
> --
> _________________________________
> Note to self: Pillage BEFORE burning.
>
> _______________________________________________
> gnso-rds-pdp-wg mailing list
> gnso-rds-pdp-wg at icann.org
> https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg
>
>
>
> _______________________________________________
> gnso-rds-pdp-wg mailing list
> gnso-rds-pdp-wg at icann.org
> https://mm.icann.org/mailman/listinfo/gnso-rds-pdp-wg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/gnso-rds-pdp-wg/attachments/20170214/6017e873/attachment-0001.html>
More information about the gnso-rds-pdp-wg
mailing list