mp | Centralisation, reversibility, and restarting
href="//cdnjs.cloudflare.com/ajax/libs/highlight.js/9.15.6/styles/atelier-dune-light.min.css"><br>-->
home
blog
code
about
privacy
Menu
home
blog
code
about
privacy
Centralisation, reversibility, and restarting
22 May 26
Technical progress tends toward centralisation: telephone multiplexing<br>made calls easier and centralised the entire telephone industry; GitHub<br>made coding easier and centralised the world of software development.<br>Technical centralisation is mostly reversible. The economic and<br>political centralisation that rides alongside it is not. The real<br>binding force is neither the technology nor the data but the interfaces<br>imposed to make it all work together. One of the best ways to ensure<br>centralising tendencies can be reversed is to find ways to keep those<br>interfaces simple.
This is a restart of a blog. It used to be made in a fairly customised<br>way, as described in one of the earlier<br>posts1, but other than that it was<br>hosted, built, and deployed on GitHub. Your one-stop shop for all things<br>programmery. GitHub made things easy, which I guess they did as part of<br>a general policy or business aim or whatever of bringing people<br>together to them. When everybody is busy scurrying all over the<br>place to assemble bits and pieces from here and there, being a one-stop<br>shop is a great way to become the go-to shop.
And that would be an understandable aim for any business, and an<br>understandable means of achieving it. Reduce friction by enabling people<br>to do more and more disparate and distinct things in one single place.
But the ease of this also encouraged people not to worry about the<br>details they used to have to worry about in order to fit the disparate<br>pieces together within their own bespoke workflows. That’s also a<br>process that has been effectively happening forever. Innovation very<br>often strives to make things easier. And often does so in ways that<br>don’t generally need to be reversed. It might be possible to go back,<br>but there’s no need.
In an era of mobile phones and wireless connectivity, nobody needs to<br>really know how a telephone exchange really works. Like now, in this<br>present moment. We’ve moved on. And it’s great that nerds still maintain<br>knowledge and ability to operate telephone exchanges, but the loss of<br>that knowledge would have relatively little direct effect on the<br>abilities of people throughout the world to keep on innovating,<br>extending, and enriching. And yet telephone exchanges still provide a<br>good starting point for this, so here we go.
Telephone exchanges, multiplexing, and centralisation
Telephone exchanges were made obsolete because somebody came up with the<br>idea of “multiplexing” telephone lines. In the exchange era, every<br>connection between a pair of telephones was a single line. People pulled<br>plugs out and pushed them in to make those connections. They exchanged<br>the ends of cables. A “multiplex” was a way of combining many signals<br>together within a single line. Instead of each call flowing along a<br>dedicated line, a single line could be used to route multiple calls<br>simultaneously. Multiplexing made far more efficient use of the scarce<br>resource of telephone cables. It also needed de-muliplexers at the other<br>end to extract only the desired part of a signal.
One effect of multiplexing was increased centralisation. Telephone<br>switching systems were always hierarchical, and so always manifest their<br>own forms of centralisation. This was necessary to enable long-distance<br>calls to be switched ever upward from local, regional hubs to more<br>centralised ones controlling flows over entire regions, and back down on<br>the other side.
Multiplexing required physical multiplexers and demultiplexers. And<br>bigger multiplexers are more technically complex. The technical needs at<br>different hierarchical levels became (far) more pronounced. That<br>required a greater degree of centralisation, even if only for telephone<br>companies to be able to run the most central, technically advanced parts<br>of a multiplex hierarchy. And the benefits of greater centralisation<br>then fed back proportionally more to those companies able to centralise<br>the most. Which were naturally those companies that were already big<br>enough to do so. And so the technical advance of telephone multiplexers<br>centralised the entire telephone system both technically and<br>economically. And then further feedbacks between economics and politics<br>led to additional political centralisation of telephone systems, and so<br>on. Just one of many cases in which economies of scale2 combined with<br>efficiencies of technical centralisation to become inextricably coupled<br>with increased economic and political centralisation.
The centralisation of telephone systems has been problematic in many<br>political areas throughout the world, but even in such places any<br>problems are likely viewed as a price worth paying for the technical<br>advances that started with multiplexing and end up with the internet<br>everywhere on...