Strong parallels can be drawn between today's open-source peer-to-peer networks (AKA decentralized networks) and the personal computer industry of the early 1980's. Now, as then, it is all about standards, protocols, and Application Program Interfaces (API's). Yes, this is all very boring stuff. Since I don't know or care to know (for now) much about the details of these things, I'll stick to explaining the big picture, as I see it.
Before I begin talking about the importance of open-source peer-to-peer networks, let me pave the way with a history lesson. Back in the early 1980's, the microcomputer industry was offering consumers several different microcomputers, each running it's own operating system. The phrase "personal computer" was also used as an alternative to the word "microcomputer" back then, until IBM hijacked that phrase. Among these competing computers were the Commodore PET, the Commodore 64, the Apple line of computers, and a line of Tandy Corp. computers, which included the TRS-80 Model One (on which I learned to program Benton Harbor Basic). The monumental problem associated with these multiple lines of computers was that they couldn't use each others' software. This meant that a company that wanted to write software would either be stuck writing for only one computer line, or it would have to write a very different version of its software for each computer line.
Then, IBM came along with a modular microcomputer that it marketed as the "personal computer", stealing the name of a general type of computer, thereby implying that no other personal computers had ever existed or been important. Bill Gates managed to convince IBM to use his DOS operating system, at which point his delusion of grandeur took over, and he began working to force every PC owner to run only Microsoft Software. Initially, that was a good thing, because he forced nearly all the software development companies to write software for one platform. But, after some point in time, it stifled innovation and left most people using a not-so-great (in my opinion) operating system for decades. Ultimately, Microsoft lost its stranglehold on the PC software industry, and now there are many, many organizations writing software that runs on the PC, including other operating systems like Linux. The beauty of IBM's modular computer design was that it allowed thousands of companies to create hardware that was compatible with the PC. For this reason, IBM PC hardware standards continued on, long after IBM lost control of its PC hardware. Ultimately, Bill Gates' and IBM's gift to the PC industry was a common platform that let developers write code and create hardware that would work with nearly every computer sitting on everyone's desk.
If you are not familiar with the concept of a peer-to-peer network, you can read more about it in my earlier article on alternative social networks. To summarize, a peer-to-peer network is a group of users' computers talking directly to each other, rather than talking to some company's central server. When a user logs onto a peer-to-peer social network, instead of connecting to a central server, as he would by logging on to Facebook, he is connecting to a network where he can see data stored on every other computer connected to that network. Today, many so-called "peer-to- peer" networks exist, for example, Diaspora. But, most are actually hybrid peer-to-peer networks, because users are not connecting directly to other users' computers. When a user logs on to Diaspora, he connects to a server run by another user who has decided to share his server (his computer running the network's server software) with other users' computers (or peers) for the purpose of enabling them to access content on the Diaspora network. Many peers, perhaps hundred or thousands, are connected to his server, which routes information back and forth between them. His server is also connected to other servers, which are connected to the peers they serve. By the way, this hybrid design makes it easier for an outside entity, like a government, to shut down the network, because far fewer computers must be seized to bring this type of network to a stop than a pure peer-to-peer network. At any rate, peer-to-peer and hybrid peer-to-peer networks require no central company server to operate.
Currently, we have several different "peer-to-peer" networks springing up, all hoping to be the "next Internet", or at least fill a major nitch in it. I'm thinking mainly about two kinds of networks. The first is social networks, like Diaspora, Friendica, Mastodon, Pleroma, and others. The second kind is networks that are designed primarily to host websites. Some of these websites can be distributed social media websites. Some examples of website-hosting, distributed networks are ZeroNet, Freenet, and Blockstack. All three of these replace the current underlying Internet structure of DNS servers and public-key infrastructures with their own solutions.
Just as none of the early competing developers of microcomputer hardware and software were able to make themselves the one-and-only developer of microcomputer hardware or software, the developers of individual peer-to-peer networks are unlikely to make any one of their networks the "next Internet". With multiple, competing networks, we have been seeing (that is, until very recently) competing, incompatible solutions, just as we did in the early 1980's microcomputer industry. The solution, I think, is the same one that was begun in the microcomputer industry by Microsoft and IBM--to agree on a common standard (in this case, maybe a common peer-to-peer protocol), and let users choose the peer-to-peer network they prefer to use to connect to the Federated (conglomerated) peer-to-peer network, the "next Internet".
In case I haven't already made it clear enough, the reason to use peer-to-peer networks, instead of centralized networks with central servers, is to take away the power to control the network from giant companies like Facebook, Google, and Amazon. These companies have a habit of using their power for their own benefit, not that of their users. The goal of peer-to-peer networks is to put the power back into the hands of individual users. Unfortunately, things are a little more complicated than that. While the fundamental structure of peer-to-peer networks (especially the open source variety) theoretically makes it more difficult for networking software developers to subvert their networks, it is not impossible. Still, this is a huge step in the right direction. The next step after that should be to give the user a choice about which network, with it's associated software, to use. This is necessary to prevent the possibility of one network's developers simply supplanting Facebook and Google to become the next social media or Internet tyrants.
Let's say consumers were to flee from Facebook, Google, and Amazon toward peer-to-peer networks and mostly congregate on one platform, say for example, ZeroNet. Now ZeroNet developers are in the same position Google once held. They are in a position to decide how the new Internet works. As with every group of people since the beginning of time who have had a large amount of money thrown at them--and anyone who is in a position to control the Internet will have a large amount of money thrown at them--ZeroNet developers will give in to the temptation to take the cash at the expense of the people who are using their network. I'm sure today's ZeroNet developers will argue that they have no way of doing this. Let me tell you, when a large amount of money is up for grabs, people find ways of doing things. Just look at Google, Facebook, and Amazon. They originally started out to benefit us all. Now, their goals are murky at best.
This corruption of noble goals can be put off, perhaps for decades, by establishing a way for all peer-to-peer networks to communicate with each other. In 2018, the World Wide Web Consortium recommended that ActivityPub become the protocol (or standard) for linking peer-to-peer social networks together. This protocol allows Diaspora servers to talk to the servers on Friendica, Mastodon, etc. So, for example, when someone posts a cat picture on Diaspora, everyone on the other social networks listed above can see it. The importance of this is that, if the programmers of Diaspora ever decide to become tyrants, Diaspora users can switch to using Friendica, Mastodon, or some other network. And, hopefully they will be able to take all their data with them, including perhaps even their posts of cat photos. In other words, no one group of developers can control the federated peer-to-peer network, because users are not forced to use any particular peer-to-peer network and its software.
I can't resist making the observation here that, because Google now has more than half of Internet users using it's Chrome browser, it has begun to dictate how the Internet itself operates. See my earlier article that talks about this. Among other things, this allows Google to bias Internet traffic toward very large corporations and away from smaller websites run by smaller companies or individuals. This is only one of the things these companies are doing that are not in their users' best interests. This is why we don't want to simply trade the Google/Facebook/Amazon monopoly for a new peer-to-peer-based monopoly.
I know this article has been long and filled with twists and turns, and I'm sorry about that. But, the fact is that the possibility of a next Internet is a complicated topic. It involves technology, business strategies, marketing, economics, psychology, and ultimately politics. I could have gone on for several more pages, but I've decided to stop here and just say this. In my opinion, we need a standard or protocol that allows small peer-to-peer networks to connect to other such networks. This is the approach that will lead to individual users having the most freedom, privacy, and visibility on the next Internet, whatever it turns out to be.
Copyright © 2018-2019 The Cheapskate's
Guide to Computers and the Internet. All rights reserved.