 The history of social media in 5 waves: 1st wave: Focus on protocols. Examples: Email, finger, Usenet, BBS, listserv, IRC 2nd wave: Focus on web. Examples: Classmates.com, LiveJournal, Habbo, Friendster 3rd wave: Focus on browser-based apps. Examples: MySpace, Facebook, YouTube, Flickr, Reddit 4th wave: Focus on mobile apps. Examples: Twitter, Instagram, Snapchat, TikTok 5th wave: Focus on decentralization. Examples: Mastodon, Bluesky, Matrix, Pixelfed, PeerTube 108 comments
Wave 4 of social media is already 17 years old. That was when Twitter initially came online. Interestingly, Twitter was established before the smartphone went mainstream. It was 140 characters because it had to be carried through SMS, and SMS had a limitation of 160 characters. And the service wasn’t easy to use at first. @atomicpoet The first 2-3 years of my Twitter feed (2007-9) take up only a few pages, as I logged on once or twice a week to see if anything was happening yet. 🙂 @atomicpoet And you really felt like if you said, “hello,” there would be an echo online. @atomicpoet my Twitter is is 2676. Back then, it was an SMS interface only. I think you sent a message to TWTTR to post  @atomicpoet To say the least. You had to type the letters RT to retweet, and other Twitter rules like: "If you just reply with @username the tweet will only go to people who follow both you AND that person If you add a dot before the username, it’ll go to all of your followers So, by default, when you reply to someone, most of your followers won’t see what you said to that person unless they also follow the person you replied to." Obviously, all waves of social media were accompanied by big leaps forward in Internet technology. Wave 1: Internet It’s my belief that the “appification” of client-server software combined with inexpensive hosting will be a game changer.  Self-hosting obviously has bigger ramifications beyond social media. Here's examples of self-hosted services that are downright revolutionary: 1. Nextcloud - File hosting Not only is this software you can host yourself, it is getting less expensive and easier to use! Beyond social media, what are the practical reasons people will want to self-host instead of pay for SaaS? Here's some reasons: 1. Most people own multiple devices. Client-server software makes managing content easier 2. Self-hosting is cheaper than SaaS -- especially when devices like the Raspberry Pi are involved. 3. Ownership of data. When you self-host, it's with the assurance that your data is truly your data. But what happens when you add decentralization into the mix? Mastodon is obviously the most famous example of ActivityPub integration. But what's just as exciting to me is that WordPress, Nextcloud and Gitea are working towards including ActivityPub into their apps. Certainly, both apps are already collaborative in nature. But once ActivityPub is turned on, they become decentralized *social media*. That's the real revolution right there. The possibilities for decentralized social media are endless. Imagine ActivityPub integration with self-hosted: 1. Calendar The Internet has now become exciting again! ``` This is so so true, feels like it could be good rather than exploitative. The move to decentralization and self-hosting is going to kill a lot of big companies. This always happens when there's a big shift in technology. For example, Kodak couldn't survive the shift to digital cameras. Sun Microsystems couldn't survive the shift to LAMP stacks. Nokia couldn't survive the shift to smartphones. The same will happen during this next technological shift. The first casualty will probably be Twitter -- but I suspect Meta won't be able to survive either.  @atomicpoet Future generations wont believe an algo can really be addictive and wonder why there are laws on the books preventing their use on children  @atomicpoet While Nokia didn’t survive, I’m pretty sure their phones did. Those things are the cockroaches of tech. They’d survive a nuke. 😂 I point out Kodak, Sun, and Nokia because they all have stuff in common: they were killed by a technology they pioneered. * Kodak invented digital cameras Each of these companies saw the future but they could not adapt. Now what I find fascinating about Wave 5 of social media (decentralization) is that Twitter saw it coming. Bluesky was spun out from Twitter in 2019. In fact, Bluesky's first (maybe current?) CTO was Parag Agrawal -- who later became Twitter's CEO prior to Elon Musk's acquisition. Nevertheless -- just like Kodak, Sun, and Nokia before it -- Twitter will not survive the next technology shift. Why is this? Right now, Twitter is making the same mistake that Kodak, Sun, and Nokia made. Elon Musk is chasing after what he believes is "high value customers" with what he believes is a superior product. This is why he's charging $42,000/month for access to Twitter's API. For now, Twitter might be a superior product to Mastodon. Maybe -- that's arguable. But the tech that underlies Mastodon has been iterated, and now it's beginning to go toe-to-toe with Twitter. @atomicpoet do you see Bluesky ever picking up? To me it feels like when a team who made a classic video game moves on to form a new studio, and people whole that their new game will fill the nostalgic void in their souls. But it’s never the same and others are trying to build similar games at the same time. @Jbquinn It could happen, but it's been slow to market, and now momentum is with ActivityPub. When it comes to Wave 5 of social media, here's the crux of Twitter (and Meta's) problem: 1 They must manage a big, unwieldy centralized infrastructure 2. They must keep current "high value customers" happy 3. By pivoting to decentralized social media, they must make the deliberate decision to kill their current cash cows Can they pivot? That's possible. Apple pivoted from iPod to iPhone -- killing their cash cow in the process. But it's extremely unlikely.   @atomicpoet Twitter is a carcass, Facebook is what people use to catch up on their cousins.  @atomicpoet Apple could afford to take that chance because the iPhone was an iPod… but better and still their own ecosystem. Meta/Twitter going decentralized provides none of that practical or corporate security. One thing I will say about decentralized social media is that we don't know where it will end up. For example, when the first smartphones came out, they came with physical QWERTY keyboards. Now they're input with touchscreens. Right now, decentralized social networks look like Twitter duplicated across multiple servers. But it will evolve into something substantially different in time. The problem with chasing after "high value customers" is that they know what they want. Concurrently, they don't know what hasn't been invented yet -- they're not considering future innovations. Back in the 19th century, if you asked someone what they wanted in terms of better transportation, most of them would have said "a faster horse". Elon Musk's focus is on people who are paying for Twitter. Who are Twitter's "high value customers"? Advertisers. What advertisers want is a better advertising system. They don't give a damn about decentralized social media because there's no centralized system that feeds people ads. Mastodon can't help them. Twitter advertisers are like folks in the 19th century who are demanding a faster horse. Guess what? The proverbial horse will soon be put to pasture. @atomicpoet that is a good analogy and correct nice thread @atomicpoet If you have to advertize your product it must not be very good. (This is also why I'm skeptical about Meta's move into the Fediverse. It could work if Meta is willing to kill Facebook, Instagram, and WhatsApp -- as they presently exist. I don't think they can.) One final thought. The conceit amongst social media incumbents is that the Fediverse has no customers. But this perception happens because the Fediverse's current customers are perceived as "low value". They're deemed so low value that people don't even acknowledge they exist 🙂  Great posts today 🙏 Time for the story of the Zen master and the little boy… #SocialMedia @atomicpoet iPod -> iPhone worked because iPhone IS an iPod (touch), just also with a phone function. It's literally just a feature add. I think the big companies can pull this off, but it'll have to be transparent to their big customers. @atomicpoet Strongly urge people to deactivate their birdsite accounts. I did and don’t miss it at all! @atomicpoet I don’t think that’s entirely accurate: mySpace pivoted and still exists, and that was popular decades ago. Nothing is stopping Twitter from pivoting, except maybe greedy shareholders. I'd take issue with the Sun example. You have to be very brave to step away from a successful existing business model.  Nokia might not have survived the shift to smartphones, but the Nokia 3310 will forever be indestructible!  @atomicpoet to your second point I think there is subscription fatigue and people want to pay once and own what they’re buying. The shift back to “shrink-wrapped” products and software is where we’re headed. @atomicpoet great points. WRT no. 2: “Self-hosting is cheaper than SaaS -- especially when devices like the Raspberry Pi are involved” The problem isn’t cost; it’s hassle, and resilience. Running services on a local RPi is cheap but then you lose years of data when a $10 MicroSD card dies.  @atomicpoet @Tedgarrison3 Some people are able to do it. Nevertheless, there are levels to this kind of thing. And the new technological shift will take awhile. Remember, at one point, you couldn't contact a remote server without incurring a heavy long distance bill -- but times have changed.  More things – not just social media – probably need to be offered as containerized solutions (be that Docker, flatpacks or whathaveyou). Anything that allows a casual to go to a hosting service and easily deploy something, preferably in a portable way. @atomicpoet Yes, except that RSS was part of the second Era, and had the potential to be a powerful force of decentralization back then. It did lack the identity portion of the fediverse, though. @atomicpoet and you know you are old when you’ve been around for all of them. [It’s me. I’m old.] @atomicpoet Only friendly addition I'd make is to add blogs and RSS to the second wave. @atomicpoet Damn... I've participated in all five waves. I'd probably put Unix `talk` ahead of finger, though. And mentioning IRC with no love for BITNET-Relay? @atomicpoet This is a great structure/breakdown/history of things. I'm still reading through the thread, but I'd say you're right on target. Now I have to check out #Spacehost and your other projects - looks quite impressive. @atomicpoet I like this framing, but feel like the 2nd wave currently misses the self-hosted blog/pingbacks/comments side of things alongside the centralized versions (LJ, blogger/blogspot, etc.)  @atomicpoet Funny thing is that as far as my personal use goes, the 1st wave has endured longer than all of the others. It'd be really nice if 5th wave acquired some of that protocol focus though, because making clients based on floating non-standard APIs isn't great. @atomicpoet I'd argue the focus is back on protocols. The point of protocols - heck, the point of ARPANet - is decentralization. Or put in another way, protocols and decentralization are two sides of the same coin. @atomicpoet @dmacphee You are making it too easy, probably because federated is on your agenda. Wave 4 is mobile based, but actually has 4 phases based on the media itself. Text based. |
Gregory's Server
@atomicpoet 1st wave: FIDO net