Today, I stumbled upon this blog post entitled Why the Fediverse is stuck (warning: SFW post but NSFW elsewhere on the site). This point stuck out at me:
It may seem contradictory that all these 'decentralised' protocols are based on a very centralised structure such as the DNS, when one could use a DHT, like torrent is doing.
This is something I've said often about the Fediverse, and why it's only a temporary solution to a bigger, more complex problem.
The author later goes on to say:
Could SMTP, XMPP, SIP have been used to build the fediverse?
The answer is: yes, and they would have worked MUCH better.
How do I know? Because they are widely used, and they DO work much better. As easy as that.
I once again agree. All those protocols are perfectly serviceable and could be great standards to build social networks upon. I certainly hope it happens.
But what's going on with those protocols right now?
The value of SMTP is experiencing a rennaissance, especially as big corporations realize there's value in sending newsletters, and that email outperforms social media in terms of engagement. Good for email!
Unfortunately, just like social media, email has become centralized. By far, Google is the dominant email provider. As of 2019, there were 1.5 billion Google mail users. If Google mail were considered to be a social network, it would rival Facebook.
Regarding XMPP, there's no doubt in my mind that it's the best messaging technology today – yes, better than Matrix, Signal, and Telegram. This is why both Google, Apple, amongst others, continue to build technology on top of it.
The problem is, how many people use an XMPP client? Anyone can use Conversations, Gajim, or Swift.IM, but who is using any of those options? It seems as though XMPP is a good standard to build a centralized messaging app on, but nobody's building apps that federate.
Finally, let's consider SIP. Back in the 2000s, I remember several VoIP providers allowing you to work with any SIP client of your choosing. Those providers got crushed by Skype, which is now dying a slow death after Microsoft bought them and is trying to push everyone to Teams.
However, there's still hope for SIP. Jitsi uses the protocol, and it's become successful within the open source community – but not enough to displace the likes of Zoom.
During the last decade, I've noticed similar patterns with standardized protocols:
- If they're wildly popular, they get centralized
- If they're superior technology but don't achieve a good network effect, companies build with them – they get centralized
- If they're good enough and stick around, maybe they'll get a second life
Let's consider what's happening right now. Each protocol has either become centralized or doesn't have the network effect to be used widely.
But what is decentralized and has a wide enough network effect for use? That is to say, what collective protocol(s) have momentum?
Earlier this week, I wrote about how easy it was to create a network effect that rivals Instagram on a new Pixelfed instance. Maybe I'm a weird edge case who just got lucky?
However, I also have a Mastodon account on a different instance. It 336 followers and receives hundreds of interactions per day. In comparison, I have a Twitter account that has over 8,600 followers and a fraction of interactions.
In other words, the Fediverse might use messy, badly documented protocols. However, when it comes to decentralizing the web – without need of blockchain – the Fediverse has momentum.
Now let me now confess something. Off and on, for five years, I've been working on different technologies to decentralize social media. I've been making progress. What's more, I believe they will eventually be better than current Fediverse apps. They're not ready yet.
But you know what's here now? Mastodon, Matrix, Pleroma, Pixelfed, Peertube... I can go on. It's better to start decentralizing now than to wait years for better technology.