Define / concretize recommendations for I-D

[michaelaA19]

[Nick Merrill] Is there more a need to focus on other issues and solutions re: equality in access, e.g development of ISPs in the global South in order to address consolidation?

[Response] This is certainly something we can incorporate into the suggestions for further research and/or recommendations. However, the issue of development of local ISPs extends beyond just the global south (e.g. increasing competition in the ISP market more broadly). It is also likely to require actions that extend beyond the remit of the IETF. But it is certainly something to be incorporated into the I-D as an area for further consideration!

[michaelaA19]

[Nick Merrill] Focus on what is distinctive, what IRTF can/should do, end user, recommendations (is it for metrics/etc)? The I-D needs to be more aggressive about the assertion that the IETF/IRTF MUST do something about this, and be more specific on what exactly it should do.

[Response] We can certainly be more aggressive in the recommendations, and specifically for actions that the IRTF/IETF can take to address the issue. The aim is for this document to set out a framework for understanding the potential consequences of centralisation to the end-to-end principle and how it can negatively impact end-users. From this framework, we aim to promote developers and researchers working in the IRTF/IETF space to consider how their work could have a negative impact on end-users, with the focus here being on centralisation and the end-to-end principle. In terms of the ‘how’, we will integrate more concrete suggestions into the recommendations.

[michaelaA19]

[Tobias Fieberg] The I-D makes more the impression of the beginning of a position/policy paper than a draft (even for IRTF values of draft). It might be better to see how far you can cut down the document to really focus on the essentials. E.g., what might be a good document would be something organizing (read: Taxonomy) all the different ways/factors/perspectives playing into the centralization and fragmentation (i believe the two to be connected) of the Internet; Stretching from walled gardens to political agendas an interesting train of thought;

[Response] We will be integrating more concrete recommendations for centering the end-user perspective, looking at RFCs that are cited in the I-D (including 9260 and 8890) to identify any gaps. We can then ascertain whether it will fit for evolving into a text that contain recommendations and therefore be an RFC.

[michaelaA19]

[Andrew Campling] Does this community have any competence to say more than that? It’s not multistakeholder; arguably other places have more ability/competency to say more on this to to then inform the IETF’s actions rather than the IETF itself making these decisions (e.g. IETF more of a receiver of these decisions and/or convenor of discussions)

[Response] We would suggest that the IETF could be doing more on this topic even if it is not a mulitstakeholder space. The IETF can do something even if it t may not be making decisions, such as hosting more discussions on protocol design and impact to augment or curtail concentration of power.

[michaelaA19]

[Arnaud Taddei] How do we continue the work here and by which angle do we tackle it? Want to keep it user-centric e.g. proposing finding solutions on end-user

[Response] Yes, we want to keep it user - centric and find solutions that centre the end user. We will be adding more proposed concrete solutions to the I-D that we can consider for how to continue this work.

[michaelaA19]

[Vittorio Bertola] How can we make it so that what we discuss at this part of the IETF actually impacts what the rest of the IETF does?

[Response] One way we aim to do this is by reviewing current RFCs that focus on the end user and identifying relevant questions/recommendations and any gaps in relation to fragmentation, centralisation and consolidation. The main questions guiding our analysis will be “how are the recommendations or guiding questions included here relevant to centralisation or fragmentation, specifically how are they relevant to the principal connections of concern: 1) the advancement of single points of failure 2) impact on user choice”?