[AusNOG] Fwd: [Internet Australia - members] Net neutrality

[Paul Brooks]

The way I've experienced it,  they're are several flavours of Net Neutrality, and the distinction falls into the realm of religious wars and debates over a few glasses of red.
The radicals/fanaticals believe every single packet is sacred, and there should be no variation in handling, no prioritisation of anything - the transport network should be a strict FIFO dumb pipe. The really radicals believe even the QOS/DIFFSRV  bits or dot-1q tags should be ignored.

Moderates believe there should be no discrimination according to source or destination, but it's OK to prioritise real'time traffic over best-effort, OK to selectively drop non-real-time stuff like email or FTP in congestion conditions, and OK to respect the DIFFSRV bits the source host intended. Not OK to selectively prioritise your own VoIP service over another VoIP service, or deprioritise a competing service. Perfectly OK to boost internal management/monitoring traffic to ensure the network can be managed during link failures.

Radical NN advocates tend to be policy "experts" that have never run a network in anger. I fall into the moderate camp. 

In the past I've considered we didn't need explicit Net Neutrality policy or regulation in Aus because we have anti-competitive conduct laws. Deprioritise my service over your own and I'll have the ACCC on your case.

 I also have no idea why it is an issue for NBNCo. Their job is to make sure there is enough access bandwidth that priorities don't matter, it all gets through uncongested regardless.


On 25 November 2015 8:52:53 am AWST, Mark Cheeseman <mark-lists at cheeseman.org> wrote:
>Thanks Jared,
>
>Agreed on that paper. The introduction put me off when it made the 
>assertion that equal = fair and failed to justify that assertion later 
>in the paper. And the paper comes to no conclusion that I can see.
>
>I do have to question why the head of NBNCo, providing as it does layer
>
>2 services, is calling for a discussion about things that aren't (or 
>should not) be his concern.
>
>Holly's post yesterday summed up the discussion quite well: "First - 
>network neutrality is one of those terms whose meaning is in the eye of
>
>the beholder (mixed metaphor, sorry)"
>
>Fair treatment of traffic doesn't equate to equality in just passing 
>packets. Why wouldn't an ISP (or company) prioritise voice traffic over
>
>video, over file transfers? The ISP is delivering services to the 
>customer. Why not deliver what they the customer wants; what they think
>
>is important? That defines fair treatment of traffic!
>
>I've been watching this thread with interest and have avoided
>responding 
>until now, mostly because I have been on the road and reading the 
>conversation on a phone. There is a lot that needs to be thrashed out
>in 
>this discussion and the most important bit just now is what "Net 
>Neutrality" actually means.
>
>Good to see the conversation on this list; my main concern is that the 
>concept of Net Neutrality is ill-defined and if we don't take a stance 
>on defining it, it will be defined either by the media, or by prior art
>
>(ie, the USA).
>
>-Mark
>
>On 25/11/2015 9:26 AM, ausftth at mail.com wrote:
>> Vijay Sivaraman wrote:
>>> I would like to weigh in (as an academic) on the net neutrality
>discussion - we have recently authored a position paper that surveys
>perspectives from around the world.
>>> This is to appear in the ACM Computer Communications Review (CCR)
>journal in Jan 2016, and is freely available at:
>>> http://www2.ee.unsw.edu.au/~vijay/pubs/jrnl/16ccr.pdf
>>> I am most welcome to any comments on our article, including
>pertinent things that we may have omitted, and your views on what we
>(the research community) can do
>>> do shape the net neutrality debate in Australia.
>> What a terrible, sloppy paper. It states that CDNs are a form of
>prioritization and conflates CDNs with network neutrality violations.
>It also confuses congestion with quality and implies that peering is a
>form of prioritization.
>>
>> The paper also presents value judgements without any corraborating
>research and takes an explicit non-network neutrality stance.
>>
>> I do admire the level of out of touchedness with reality that leads
>the authors to surmise that the only reason consumers are against paid
>prioritization is due to the deals being stuck in backrooms.
>>
>> The paper also posits a tragedy of commons by underinvestment in
>broadband networks unless network neutrality is abolished. This is an
>outlandish position to take given that it has been shown in the US that
>network neutrality legislation has had no effect on profitability or
>investment. The main observable effect has been that any lingering
>peering issues have been cleared up, which is a clear win.
>>
>> The proposal in chapter 4 is beyond silly. The proposal only benefits
>the ISP by introducing a billing point for rent seeking. It's a losing
>proposition for the CSP and an unnecessary headache for the consumer.
>It's very much not a win-win-win, but a win-lose-lose proposition.
>>
>> Furthermore there is no feasible way to implement the proposal in
>current access networks and the proposal assumes an ISP network without
>oversubscription, congestion or chokepoints. Any such network would
>immediately make the proposal redundant.
>>
>> Also, do we really need to contribute to link rot and source
>obfuscation by using URL shorteners in academic papers?
>>
>> I can only assume the writers have no real world network or industry
>experience. I do wonder where they get their funding from, tho.
>>
>> In closing, this is not a research paper, it is an opinion piece. I
>can only recommend retracting the paper.
>>
>> Jared
>> _______________________________________________
>> AusNOG mailing list
>> AusNOG at lists.ausnog.net
>> http://lists.ausnog.net/mailman/listinfo/ausnog
>
>_______________________________________________
>AusNOG mailing list
>AusNOG at lists.ausnog.net
>http://lists.ausnog.net/mailman/listinfo/ausnog

-- 
Paul Brooks
Trident Subsea Cables

Sent unplugged 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20151125/b6d2697e/attachment.html>