universal basic internet

date: 2020-01-27

tags: technology walkaway

internet costs money because you need loads of people to maintain and operate a ton of shit so that your cat videos can reach every corner of the globe in <1s. that shit uses nearly as much energy as the fucking airline industry. the pleroma network[1] is a breath of fresh air in a world of mega-corporate silos, but stacking massive piles of kernel, system, networking, and application code on top of each other doesn't address the fundamental cost of maintaining the internet infrastructure that it depends on.

1: pleroma network

secure scuttlebutt[2] has its own issues (electron, javascript), but it actually addresses the cost of fixed infrastructure to an extent. i agree with the majority of the design decisions behind the protocol, but i think software efficiency needs to be a higher priority. this train of thought has led me to discover a future internet protocol called named data networking[3]. this is one of the more promising alternatives to TCP/IP which side-steps a lot of the major issues with routing in a roaming mesh, among other things. because NDN is an information-centric network, the routing systems can be much more flexible. there is no need to maintain telephone-like streams to exchange data, rather the fundamental unit that the network cares about is a logical block of useful data.

2: secure scuttlebutt

3: named data networking

NDN can easily be adapted to fetch data from many more sources than a TCP/IP without adding large amounts of application code that must maintain accurate routing information in order to replicate information among peers. systems like ntorrent[4] are much simpler to implement on NDN than the equivalent protocol running on IP. this also applies to ideas such as sneakernets[5]

and IPoAC[6], which can be directly integrated into the network system intead of acting as separate systems. a gossip system[7] could also enable replication between mobile devices as they are carried from place to place along people's routes to work and to social engagements. in fact, it is likely that a social gossip network could provide much higher bandwidth than TCP/IP at the cost of increased latency.

4: ntorrent

5: sneakernets

6: IPoAC

7: gossip system

changing network protocols on a macro level is unlikely to happen any time soon. if there is no collapse scenario (which i certainly hope to be the case), then the energy-intensive systems that are employed on the internet today are likely to keep running as long as they are operative. however, NDN is an interesting approach for solarpunks and eco-village (micro)infrastructure[8] because it can be operated with a very small amount of low-powered infrastructure, especially a social gossip-based variant. there are also some interesting applications for latency tolerant gossip networks in space that could work with essentially no changes from terrestrial solutions.

8: eco-village (micro)infrastructure

so back to the internet cost problem: it is unlikely that a market-based solution to ditch massive and profitable investments in favor of a peer to peer social gossip network will ever happen. this is probably true even if NDN is widely adoped at some point. however, it is certainly possible for a community of high-tech low life's to build a global gossip network that costs next to nothing to operate. this is how we get to post-scarcity. not by creating policies that require free public internet or other resources, but by creating solutions that permanently lower the cost to something that anyone can afford.