Each year, folks ask me for predictions about what's going to happen in tech in the coming year. Generally, what they're really asking is what I hope/expect is going to happen to the five or six biggest companies in the tech industry. That matters, especially in the context of policy and social impact, but in reality, very few of the conversations about their futures are really about technology.
Instead of speculating about that (uh, regulate them more!) I'm going to focus on actual technologies that caught my eye in the past year, and which I think have reached a level of maturity which leaves them poised to break out into something notable in 2022. Of course I do want to note that most of these have been in gestation for years, and will continue to be in development for a long time to come, so I'm not pretending I'm the first to notice these, by any means.
Here then, in no particular order, are the technologies I'm watching most closely in 2022:
- CRDT: Conflict-free Replicated Data Types are data structures that promise to have a truly significant effect on how we make collaborative apps and experiences. (In marketing speak, it's going to revolutionize how we work together in real-time.) In practical terms, we're able to do lots of interesting real-time collaboration together in tools like Google Docs and Figma and Glitch, but most such platforms are relying on technologies like operational transformation (OT), which was behind pioneering experiences like Google Wave and SubEthaEdit and has matured greatly in the years since, but still has constraints about simultaneous editing of complex data, especially if one or more participants are offline. CRDT now feels a bit like when we first heard about Git after years of using earlier version-control tools like svn; once someone wraps a great user experience around it, it's going to be a leap forward for an entire seemingly-mature category of applications. So why does this matter now? Because we've just seen the first credible reference implementations of CRDT, after years of mostly having papers and most esoteric proofs of concept. Whoever implements an only-possible-through-CRDT distributed app first is going to stir up quite a buzz.
- Unreal 5: There's an interesting confluence of the usual this-engine-has-better-graphics evolution of gaming engines along with the broader cultural awareness of engines like Unreal being critical tools for disciplines as varied as filmmaking and architecture, in addition to of course being useful for gaming itself. This Digital Foundry deep-dive into the recent Unreal demo for The Matrix 4 ably describes the numerous discrete iterative changes that, together, all combine to push forward the engine to a fundamentally different set of assumptions about the level of fidelity that they're attempting to replicate.
- The move to ARM: This is no secret, but remains astounding for just how much impact it's having, and how rapidly things are advancing. The flagship of this moment in processor architecture shifts is Apple's move to their Apple Silicon platform, which after more than a year of the transition keeps astounding both in terms of raw performance as well as their adeptness in shifting the software ecosystem over to the new world incredibly rapidly. On the server, platforms like Graviton are earning similar levels of developer enthusiasm, and even though the focus there is on running a much more hetereogeneous software stack, the coder community is being surprisingly nimble in making the transition there as well. The point I'm watching for in 2022 is when major apps or technologies start to specify that they require these ARM architectures in order to enable major features, or perhaps to run at all. We'll know we're turning a corner then.
- Differential privacy: There's no more evocative, and culturally important, consideration around today's ubiquitous technologies than the privacy implications. Having the ability to mathematically assert certain levels of privacy around data is already having big impact for researchers and corporate policy makers, as they're able to better protect users, or to ensure that their own products or practices are following baseline standards of privacy protection. (Here's a good explanation of how differential privacy has been used to protect people during the 2020 U.S. Census.) But it's easy to imagine this evolving into a consumerized experience in the near feature, especially as more consumer products are using tools like encryption to enable things like fractional sharing of bandwidth or reporting back from IoT devices. Here's hoping 2022 brings consumer-ready tools that leverage differential privacy at a level that allows non-expert users to feel better about their privacy.
- WebVR: This technology is one I've gotten a front-row seat to watching mature, and it's been a delight to watch it thrive. Amidst endless amounts of metaverse hype and constant vaporware announcements of what's coming to VR, AR and XR, the WebVR community has kept quietly chugging along making cool shit, making it open, making it free, and making it fun. At this point, they now form the foundation for an entire ecosystem of open VR/AR development, and the latest work around making it easier to link between virtual creations promises to make the experiences even more immersive and interesting. Couple that with the fact that these open VR tools don't require costly hardware to participate, or proprietary tools to create, and it feels we're on the cusp of something genuinely generative and interesting.
So, that's it! Maybe something will come of these various technologies over the years to come. Let's check back here in a year or two and see how we did.