How the net was lost

Those who currently struggle to maintain what is called “Net Neutrality” on the internet I think have taken too limited an approach to their struggle. What they ask is to maintain an existing status quo that had already been eroded from the original promise and potential of the internet against those who wish to change it even further. This to me leaves for a poor negotiating position when congress loves to bridge difference with half measures, and even limited compromise between the current status quo and proposed changes would still be disastrous. This would be much like North American civil libertarian’s discussing which of the remaining of the first 10 amendments they will be forced to accept being discarded versus those they think they can still actually preserve. This to me is a long term losing position to occupy.

In the beginning, the internet was a peering arrangement where all nodes were treated equally, and anyone could interconnect from any one node to another. This was the network of peering built upon public standards that anyone could freely implement. Other commercial networks also existed, some built on the layered OSI model. All, however, were implemented in some proprietary fashion, or otherwise built around some controlling model of centralized traffic routing, rather than that of essentially equal peers, and as a result diminished over time.

The internet eventually spread to the general population through modem dialup. This changed the internet from being a semi-closed environment connecting just a few hundred or thousand commercial and government institutions into something interconnecting millions. The speeds and bandwidth of analog modem dialup naturally limited what individuals could do over dialup links, but outside of technological limitations, the internet imposed no additional discriminatory practices nor did those ISPs who offered direct internet access through dialup at the time. While closed garden proprietary dialup service providers like Game Master, CompuServe, and America Online, came and went, people remained free to use direct dialup networks for both consuming and producing content on a peer basis. There was a time in fact that I ran my own domain and mail servers out of my own location on a dialup connection.

With the widespread introduction of broadband, over cable and DSL, came the first real discrimination on the internet. Just when finally there was enough easily deliverable bandwidth to go around to enable the millions of dialup users to more directly participate on the internet, it was closed off from them. At the physical layer, peering was closed by artificial uplink “bandwidth caps”, which restricted their ability to produce and distribute. At the application layer, broadband providers actively discriminate by blocking certain ports and services, particularly in regard to email. At the legal layer, broadband service agreements offered through monopoly telco and cable companies restrict what services and applications people can run.

Even during the age of dialup, when bandwidth was scarce except for a few locations, a model for service hosting and co-location appeared. This allowed someone who had a peering agreement, which already was very expensive, to then distribute and share the cost of bandwidth by renting space and/or servers on a rack to others. With the introduction of capped, application layer and legally restricted broadband, hosting became the last refuge for what the original internet was about; peering by equals.

This division between consumers and producers means only a limited few are privileged to directly publish on the internet. Yet—even though they pay considerably more for that privilege and their connectivity already, and even though consumers pay directly for their connectivity as well—the current internet backbone peer providers wish to collect additional charges, and otherwise artificially constrain traffic to hosting facilities and companies as they please, much like they do with those they consider consumers. The death of internet peering means that hosts will be billed based on their popularity as well as the bandwidth they consume and have paid for. It also reduces all hosting arrangements into a question of pure economic value, rather than considering the social value of sites that exist for non-commercial purposes or that otherwise do not charge. Finally, the death of Net Neutrality means providers could selectively choose to make some sites (commercial competitors, those who publish information that they disagree with, etc) entirely unreachable if they so choose.

The internet flourished and grew precisely because nobody was in control of traffic. That millions now are classified as passive consumers already is an affront to the dream of an active community where everyone has opportunity to participate and publish. The remaining struggle over Net Neutrality today is simply one of how small and how privileged a minority will still retain the ability to publish, and hence how much it will cost to still exercise former rights as reclassified as a limited privilege at the discriminatory whim of a few large corporations.

The internet today is already divided between a large number who are only allowed to consume and a small number who are permitted to produce. Rather than simply fight to preserve this already unequal status quo, it would be far better to challenge it by fighting to actively restore the rights of all internet users. In the worst case of such an effort, the current status quo then becomes the logical compromise position, rather than the starting point in any forced negotiation. Today, those fighting for Net Neutrality are already backed to the edge of a cliff. The telecoms want them to step a further ten feet over the edge, but they (the telecoms)are probably quite willing to accept a compromise where those defending Net Neutrality are asked to step only 5 feet off instead. It would be far better to push forward rather than to simply try to stand still.