Four stories logged on this desk in recent days have little in common on their surface: a design suite on iPad, a VR headset losing OS updates, a studio closure, and a delayed smart home robot. The thread running through them is that the constraint in consumer technology has moved from hardware to software and sustained software support. Companies can increasingly build the device; what they cannot reliably do is finish, maintain, and staff the software around it on the schedule the hardware implies.
The Robot That Waits for Its Software
The clearest statement of the pattern comes from Apple. According to The Verge, citing The Information, Apple has reportedly delayed its rumored smart home robot until next year at the earliest, possibly until 2028. The report says the robot's hardware "has been finished for around two years," but the software is not ready, and the device's software team has reportedly pushed to delay the launch.
That detail is the whole argument in miniature. A finished piece of hardware sitting idle for roughly two years is not a manufacturing failure. It is a software failure, or at least a software schedule failure, and Apple is treating it as a reason to hold the product back rather than ship it incomplete. For US consumers, the practical effect is that a category Apple has been rumored to enter keeps receding. For Apple, it means the company's reputation for finished-feeling products now carries a cost measured in years of delayed revenue.
Support Windows Are Now the Real Product
Meta's announcement, as reported by The Verge, points at the other end of the same problem. Meta plans to release the last OS update for the Quest 2 and Quest Pro by March 2027, with v213 as the final version, after which it will continue security patches and critical bug fixes through the end of that period.
The Quest 2 in particular was the headset that put mainstream US consumers into VR. A published end date for OS updates tells those owners that the software life of their device is now bounded, even as the hardware continues to work. That has become a familiar rhythm in consumer electronics, but it lands harder in a category where the device is a platform. Every month a headset goes without new OS features is a month its owner is less likely to buy software for it, and less likely to stay in the ecosystem when the next headset arrives. Meta is choosing to spend its engineering capacity on newer devices rather than on the install base it already has. That is a rational allocation and a real cost to the people who bought in early.
Building the Software Is Also the Cheap Part, Until It Isn't
Affinity's arrival on iPad, as The Verge reported, shows the pattern from the other direction. After relaunching the Affinity creative suite as an all-in-one desktop app last year, Canva has introduced Affinity for iPad, offering the same pro-grade tools for vector graphics, illustration, and photo editing, and it does not have to cost users anything.
This is the good version of the thread: a company taking software it already built and extending it to a new platform, spending engineering effort rather than inventing a new device. It is also a reminder of what makes the other stories sting. Porting a mature application to a new platform is a bounded, well-understood job. Keeping a platform's operating system current across a fleet of devices for years, or finishing the software for a robot whose hardware is already done, is not. The free iPad app is the kind of software work that ships on schedule. The delayed robot and the expiring headset are the kinds that do not.
