[NNagain] Net neutrality and Bufferbloat?

[Livingood, Jason]

> Misapplied concepts of network neutrality is one of the things that killed
> fq codel for DOCSIS 3.1

I am not so sure this was the case - I think it was just that a different AQM was selected. DOCSIS 3.1 includes the DOCSIS-PIE AQM - see  https://www.rfc-editor.org/rfc/rfc8034.html and 
https://www.cablelabs.com/blog/how-docsis-3-1-reduces-latency-with-active-queue-management. I co-wrote a paper about our deployment during COVID at https://arxiv.org/pdf/2107.13968.pdf. See also https://www.ietf.org/archive/id/draft-livingood-low-latency-deployment-03.html.

> Finally, some jurisdictions impose regulations that limit the ability of
> networks to provide differentiation of services, in large part this seems to
> be based on the belief that doing so necessarily involves prioritization or
> privileged access to bandwidth, and thus a benefit to one class of traffic
> always comes at the expense of another.

Much regulatory/policy discussion still frames networks as making decisions with scarce bandwidth, rather than abundant bandwidth, and prioritization in that view is a zero-sum game. But IMO we're no longer in the bandwidth-scarcity era but in a bandwidth-abundance era - or at least in an era with declining marginal utility of bandwidth as compared to techniques to improve latency. But I digress.

To go back to the question of reasonable network management - the key is that any technique used must not be application or destination-specific. So for example, it cannot be focused on flows to the example.com destination or on any flows that are streaming video [1]. 

Anyway - I do not think new AQMs or dual queue low latency networking is in conflict with net neutrality. 

Jason

[1] Current rules differ between wireless/mobile and fixed last mile networks; currently the MNOs have a lot more latitude that fixed networks but that may be sorted out in the current NPRM. My personal view is there should be a unified set of rules of all networks.