[gnso-rds-pdp-wg] Five models of RDS (was Re: Apologies, and some reflections on requirements)

[Ayden Férdeline]

Thank you for taking the time to outline the architecture of the existing WHOIS
service and how it has evolved over time, Andrew. With this background and your
extremely useful diagrams I now have a much clearer understanding of to what
extent different parties wield control (or not) over the data of registrants. I
am not going to be able to enter into any conversation where we are debating the
finer technical points of how a protocol functions, but I can comment on more
focused policy applications. Certainly I do not want to contribute to any
degradation of what has made the Internet what it is today – a place for
limitless innovation and where users trust it as safe and secure environment for
commerce and communication.
I have tried my best not to approach this working group with any pre-conceived
ideas as to what I would like the final outcome to be (partially because I do
not have the institutional knowledge to even formulate these ideas in the first
place ), but I lay my cards out on the table and say it is very important to me that
the Internet’s growth is not impeded in any way. I worry, looking at Model V in
particular, that we could end up in a situation where we all end up wishing a
different approach could have been taken. We could very well end up with an
outcome where the reasons behind establishing this working group are not
addressed, where ICANN or another party is stuck with huge costs, and where the
Internet’s growth and evolution is permanently stifled.
There are six issues that I see with Model V, four of which Mark has already
outlined: questions around jurisdiction, privacy, cost, and service level
agreements/overall accountability. I would also like us to consider potential
liability for the manager of the federated store for data breaches (and
liability for those who have retrieved data from the RDS and have lost it once
it was in their possession), along with the potential anti-trust implications of
there being one, privately-owned database which all gTLD registries must use.
What I like about the Internet, as we know it today, is that it is for everyone.
There is no central authority that permits different classes of Internet
activities, which means it is possible for anyone to connect to the network and
to build new parts of it. This innovation (which happens organically, without
requiring permission) has come about because the Internet’s technologies are
based on interoperability and reusable building blocks that might have been
built for one purpose but have come to support some other important function at
a later date. This is what I want to continue happening. A distributed RDS which
anyone can improve through cooperation and collaboration, because its success
depends on its continued relevance and utility, and not on some arbitrary clause
in a contract that prioritises 'incumbation' (for lack of a better word) over
innovation.
I understand that we are not yet at a stage where we should be deliberating on
what the RDS should look like (if we even conclude one is required), so before I
continue to outline my strong opposition to Model V, may I please clarify: where
did the idea for this federated RDS originate from? Is it just because in the
wish list of possible requirements we have to consider, a few suggested a
federated system? Or is there something more concrete in the works? (And if so,
can you please link me to any IETF discussions on this?) Thank you.
Thank you again, Andrew, for taking the time to go back to basics. The diagrams
you have produced of how WHOIS-related protocols and services have evolved over
time have left me much more informed of the differences between 'thick' and
'thin' registries than I was a few days ago.
Best wishes,
Ayden





On Fri, Jul 1, 2016 5:52 AM, Mark Svancarek via gnso-rds-pdp-wg gnso-rds-pdp-wg at icann.org wrote:
That's why I'd be concerned 😃

Sent from my Windows Phone
--------------------------------------------------------------------------------

From: Rod Rasmussen
Sent: ‎7/‎1/‎2016 6:27 AM
To: Mark Svancarek
Cc: Andrew Sullivan ; gnso-rds-pdp-wg at icann.org
Subject: Re: [gnso-rds-pdp-wg] Five models of RDS (was Re: Apologies, and some
reflections on requirements)

ICANN doend’t “run” most of the things coming under it’s remit or in the
ecosystem it’s policies guide. Any sort of RDS system regardless of model would
likely fall into this category as well. I thinks a few of the large registries
run 5 9 operations for example…

> On Jun 30, 2016, at 11:15 PM, Mark Svancarek via gnso-rds-pdp-wg
<gnso-rds-pdp-wg at icann.org> wrote:
>
>
> That's why I mentioned my concern about ICANN SLA. I don't think they are
ready to run a 99.999% online service.
>
> Sorry for the confusion.



Ayden Férdeline Statement of Interest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/gnso-rds-pdp-wg/attachments/20160703/52595ae9/attachment.html>