[governance] What is Network Neutrality

[Meryem Marzouki]

Dear Nancy and all,

Thanks for providing examples in your developments. They help a lot  
in clarifying a common understanding of concepts and the terminology  
used to formalize them.

Le 11 janv. 09 à 19:54, nancyp at yorku.ca a écrit :

> A neutral network must ensure that data is reachable from anywhere  
> on the
> internet network.

I would say, "...that data is reachable from anywhere on the internet  
network, and with the same accessibility level when placed under the  
same conditions". This might probably be better formulated, but what  
I mean by "same conditions" is same technical (especially bandwidth)  
capacities at both ends: upstream (sender side, e.g. content host)  
and downstream (receiver side, e.g. user access).

> The most famous example of non-neutrality in Canada occurred during  
> the Telus
> labour dispute (2005). (Benkler, 2006)  Telus blocked access to a  
> pro-union
> website by blocking the server on which it was hosted. Researchers  
> at Harvard,
> Cambridge and the University of Toronto, Munk Centre, OpenNet  
> Initiative found
> that the action resulted in an additional 766 unrelated sites also  
> being
> blocked for subscribers. (OpenNet Initiative Bulletin 010, 2005).

I simply don't understand why some people are reframing such an old,  
well known issue of content censorship by technical filtering means,  
in new clothes (a net neutrality issue), thus adding to the overall  
confusion?!
Such "collateral damages" are typical of technical filtering through  
the blocking of an IP address, due to IP sharing practices (a very  
common practice allowing, schematically, to host many virtual  
machines on one actual machine computer).
FYI, an older case occurred in France, back in 2001, when a complaint  
was filed in France against French ISPs who refused to filter a hate  
site hosted in the US (the site was called Front14) without legal  
order. While the case was in process, Renater, the public ISP for all  
universities and research centers took the decision to filter it by  
its IP address, which resulted in blocking access to some 8000  
websites having nothing to do with front14, but simply hosted by the  
same provider (this is documented in French in a press article:  
http://www.transfert.net/a7481). I personally testified in this court  
case on behalf of my organization (see my testimony in French: http:// 
www.iris.sgdg.org/documents/filtrage/memoire-filtrage.html).
Such content filtering practices are well documented in the Council  
of Europe report accompanying a Recommendation adopted by the CoE on  
March 2008, on "measures to promote the respect for freedom of  
expression and information with regard to Internet filters".
Recommendation at: https://wcd.coe.int/ViewDoc.jsp?Ref=CM/Rec(2008)6
Report at: https://wcd.coe.int/ViewDoc.jsp?Ref=CM(2008)37&Ver=add
The report and the recommendation have been prepared by the CoE group  
of specialists on human rights in the information society. As I  
already commented (http://www.edri.org/edrigram/number6.7/coe-good- 
bad-ugly), this recommendation is "the best text to date that could  
be expected from an intergovernmental institution on this issue,  
breaking off the usual rhetoric of "technical filtering panacea" to  
fight illegal or harmful content."

> Another important example of the unreachability of data took place  
> in March 2008
> at York University. A professor using the on-campus network  
> attempted to reach
> an internet website located somewhere in Europe that was important  
> to his
> research.  After repeated attempts, he still could not reach the  
> site so he
> contacted the IT (Information Technology) department at the  
> university. They
> were mystified why this would be the case. When the professor went  
> home,
> however, he found that he could reach the website. After several  
> days the IT
> department found out that the university’s bandwidth supplier  
> Cogent had
> severed a peering relationship with a bandwidth provider in Europe  
> called Telia
> which was the bandwidth network provider for the website that the  
> Professor was
> trying to reach. (Miller, 2008) Cogent did not proactively inform the
> University of the issue and the loss of connectivity.

This is, is my view, a typical net neutrality issue, contrarily to  
the above example. The Cogent vs. Telia case resulted from the  
breaking of the peering agreement between them, thus creating  
unreachability of data for some people. This breach of the peering  
agreement had nothing to do with censorship purposes, but everything  
to do with commercial disputes. Moreover, what makes it a net  
neutrality issue is that this is an arbitrary interference with the  
internet architecture itself (by cutting some routes), rather than a  
dedicated action (even if causing "collateral damages") against a  
particular content, aiming at censoring it.

This latter case is rather easy to understand and to analyze, since  
it caused a completed unreachability of content: a whole portion of  
the internet was entirely unreachable by an other whole portion, due  
to the cut of all possible routes between the two portions. Less easy  
to understand and to analyze are cases which Parminder is trying to  
define. These cases are more subtle, in that data are still reachable  
by all portions of the Internet, while with different ease or comfort  
(e.g. with different speeds). In other words, in Cogent vs. Telia  
case, "data was NOT reachable from anywhere on the internet  
network" (to refer to your definition of a "neutral network"), but in  
the more subtle cases which Parminder are referring to, "data is  
STILL reachable from anywhere on the internet network, BUT NOT with  
the same accessibility level, EVEN THOUGH placed under the same  
conditions" (to refer to my amendment of your definition).

Now the main issue is to better understand what could be such subtle  
cases. We know about, e.g. the Comcast vs. BitTorrent case. Could  
anyone give other examples of such cases in order to clarify the  
issue? Does anyone know of any contract which limits, or degrades,  
access for some applications or services? I do know that some ISPs  
forbids some protocols, typically those ISPs which are also mobile  
telecom operators forbids P2P and VOIP traffic in their mobile  
internet access offers, or those cable companies that forbids IPTV  
traffic: this is indeed a net neutrality issue, but one causing the  
total reachability of data when transmitted through these protocols,  
not a different accessibility level by artificially degradating  
performances.

While breaches of net neutrality might indeed lead to censorship, or  
more precisely to dominance of some kind of content (mainstream) over  
others (alternative, independent, non commercial, whatever you call  
it). But this is definitely different from targetted censorship of  
specific content (even with "collateral damages"). This difference is  
fundamental because each case has different causes and implications,  
and is to be dealt with differently in terms of legislation/ 
regulation, if only because the involved actors to be regulated are  
not necessarily the same: in net neutrality cases, the main actor of  
net neutrality violation is the network operator (carriage provider),  
while in content censorship cases the main actor is the ISP (access  
or hosting provider). Even if a same company could operate as both  
network operator and ISP, the regulation addresses a given function.  
In one case (net neutrality), it's a commercial/competition issue,  
while in the other (censorship), it's exclusively a civil rights  
issue. I'm not ignoring that a commercial issue might have  
consequences on the full benefit of civil rights (e.g. free speech  
llimitations) and that this should be taken into account when  
analyzing a given situation, but still, the situation should be  
addressed differently, and this involves different legislation corpus  
in each case.

> Reachability is a net neutrality question. The policy concept of  
> common carriage
> evolved into net neutrality through deregulation in the context of  
> a transition
> from analog to digital communications. Unreachability of internet  
> data may be
> due to geo-political factors, uncontrollable technical reasons as  
> well as human
> error, but unreachability due to hidden arbitrariness in commercial  
> peering is
> unacceptable.

Fully agree, be they hidden or not BTW, and provided that we address  
not only unreachability but also degraded reachability.

> The problem of the lack of transparency in commercial internet
> interconnection is largely a US problem as the US is the main  
> battleground for
> carriers refusing peering.

Maybe the main, but certainly not the only one, as vertical  
integration is also increasing elsewhere.____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:
     http://lists.cpsr.org/lists/info/governance