2024: Predictions for the Year Ahead
Published on 28 December 2023 by Andrew Owen (6 minutes)
This is my second go at making predictions for the year ahead. As with last year, I’m not going to make any predictions on geopolitics or the climate crisis and will confine myself to commenting on technology. Although with all the general elections taking place around the world, it will certainly be a year of change. But first, let’s take a look and see how I did on my predictions for 2023.
The learning curve for Kubernetes won’t improve. People will continue to look for alternatives to Kubernetes. Most of these alternatives will still have Kubernetes as a dependency of some kind.
I don’t think this was a particularly controversial prediction. Container as a Service (CaaS) platforms like AWS Fargate, Azure Container Instances and Google Cloud Run probably still have Kubernetes under the hood, even if they abstract it away from the user.
Platform as a Service (PaaS) solutions like Openshift and Rancher are still based on Kubernetes. Lightweight orchestrators like Docker Swarm and Nomad don’t seem to have gained much traction. And Amazon, Google and Microsoft are still offering managed Kubernetes services.
Data will continue its migration from on-premises and at-home storage to remote servers. I’ll still be uncomfortable referring to remote servers as “the cloud”
Another uncontroversial prediction. The cost benefit of not having to maintain your own servers and software installations is pretty self-explanatory. The wide availability of fast, reliable broadband and the relative infrequency of service outages makes this an easy choice for most IT departments.
Despite storing vast amounts of their personal identifiable information online with social networks and computer retailers, people will continue to suck at picking good passwords. And IT department password policy will continue to be at odds with best practices.
IBM’s 2023 Cloud Threat Landscape Report identified valid credentials as “the most common initial access vector in cloud security breaches, occurring in 36 percent of all cases that the X-Force IR team responded over a 13-month period.”
Humans remain the weakest link in cybersecurity, being both predictable and susceptible to phishing attacks. And human-readable passwords are still inherently weak.
More and more outlandish claims will be made for artificial intelligence. The perception of what it can do will massively outstrip the reality. Vast data sets of human created data will remain a requirement.
I’ll call this one a qualified win. AI is not magic. But the early-development phase that ran from 1956 to 2012 is long past. The last decade saw great advances in deep learning and natural language processing. We’re now at the point where the technology is mature, but we’ve barely started to think of ways to apply it.
Lua will remain an obscure language outside of video games development. Parents will continue to introduce their kids to programming with Python. People will write even more desktop apps in JavaScript using Electron.
According to IEEE Spectrum’s tenth annual ranking of programming languages, Lua comes in 26th, behind Objective-C and Visual Basic, with 2.25% of the users of Python; the number one language. Not a hard one to call.
There will be a lot more talk about Web 3.0 without anyone really understanding what it is or why they would want it.
I’m going to say I got this one wrong. There is confusion about what it is because Web3 (decentralization and blockchain) is different from Web 3.0 (semantics). But hardly anyone is talking about it, because it’s been completely overshadowed by AI and global events.
Virtual reality will remain a niche product.
In February, Sony launched the PlayStation VR2. But Microsoft and Nintendo don’t have equivalent hardware. And Meta’s Facebook Metaverse has been less successful than Second Life. In June, Apple launched the Vision Pro mixed-reality headset, but at a price of $3,500 it wasn’t aimed at consumers.
Technical writing will begin to catch up with trends in software development from a decade ago. Docs as code will become more widely adopted.
This year, docs as code has come up more often in conversations I’ve been involved in about tech writing. However, it’s not a representative sample.
The tech crunch will end. “TechCrunch”, the online magazine, will continue.
Tech layoffs in 2023 will likely top 250,000 jobs, a third more than the 2022 figures. We’ll have to wait to see the 2024 numbers, but I suspect the worst of it is over. But I expect hiring to be flat until at least 2025.
Twitter will remain the “free-for-all hellscape” that it has always been, but will stay in business.
By some estimates, the value of Twitter (which almost no-one calls X) is down two-thirds since the change of ownership. I would argue that it was overvalued in the fist place. But thus far, none of the alternatives have supplanted it. And implementing a share button for federated social networks is a drag.
There will be consolidation in video streaming. Only the most well paid sci-fi nerds can justify subscribing to Amazon Prime, Apple TV, Disney+, HBO Max, Netflix and Paramount+. This will likely come in the form of packages from traditional service providers.
5G coverage will reach 50% of 4G coverage globally. This is a fairly conservative prediction given that, according to TeleGeography, in March this year there were 703 4G LTE networks and 259 5G networks with the number of 5G networks expected to reach 390 by the end of 2023.
ARM will continue to cut into Intel’s PC market share with Qualcomm-based laptops from traditional vendors like Acer, Asus, Dell, HP, Lenovo and Samsung running Windows 11.
The chip shortage will finally end. But this will come about mainly because industry has migrated to newer chips, and the increased cost will be passed on to the consumer.
The Apple Watch series 10 will deliver non-invasive blood glucose monitoring. AI will be the key to solving the problem of reduced optical sensor accuracy for people with tattoos or darker skin.
Chinese automaker BYD (Build Your Dreams) will overtake Tesla in pure BEV (battery electric vehicle) sales. BYD started out as a battery manufacturer and, unlike Tesla, its home market has enthusiastically embraced electric vehicles.
The hybrid working backlash. From those who were perfectly happy being fully remote. And those who don’t have a suitable home work environment. But also from managers who want to keep a closer eye on their staff.
Driven by home automation, the internet of things will become more of a daily reality. It will be harder to buy a major kitchen appliance that doesn’t have WiFi.
Developer Relations will evolve to survive. In October, Mary Thengvall published the final DevRel weekly update citing the end of free Twitter API access as one reason. DevRel is important, but making the case for it will remain a challenge.
AI everywhere all at once. This year, AI crossed over into the mainstream. In 2024, it will become pervasive in ways that it’s too soon to predict.