As someone in the industry of supplying parts to keep older cars running, I view the increase of automotive software and electronic complexity as ensuring a future crisis of maintainability.
Availability of parts and service information has always been an issue for aftermarket repair/modification of vehicles. However, as long as there are enough vehicles and committed owners around to create a small market for repair parts and services, independent companies have grown to provide what the original manufacturer will not. This even applies for very niche vehicles where some devoted old fellow runs essentially a hobby business keeps the flame alive.
Even relatively recent vehicles with considerable electronic sophistication can be supported this way. I have worked with specialist companies that will modify, repair and re-engineer some of the more complex control units and electronic subsystems used on 2000s and 2010s vehicles (i.e. suspension control ECUs, digital dashboards etc).
However, the trend described in this article has the potential to upend the status quo described above, simply due to the escalating complexity involved. It's sort of a tradition that auto enthusiasts and aftermarket industry initially distrust new tech in automobiles- fuel injection, ABS, traction control systems, emissions controls such as EGR etc all got that reception initially. Expertise with all those systems was eventually absorbed throughout the industry and resistance decreased as the benefits were better understood. However, as complexity increases there is a gradual increase in costs (engineering, training, manufacturing, install/service labor) to deal with all these sophisticated systems. Without other unforeseeable changes, there are almost certainly various inflection points where increases in complexity will result aftermarket support collapsing for particular models (or specific subsystems). This is something that already happens, but mostly for relatively rare models as until recently automotive complexity increases were constrained by the slower pace of ICE and chassis development.
As these costs rise, fewer and fewer models of cars will have a healthy enough aftermarket to support investment by independent companies to analyze, repair and replace these complex systems. For sure, it will result in more models of cars becoming unmaintainable and fewer cars staying in operation beyond their warranty expiration dates. However, also I think this will result in a market space opening up not for repairs and replacements, but for various kinds of bypasses and defeat devices ("deletes" in industry terms) that will either remove complex subsystems entirely or allow replacement with more generic components. This is already occurring in some sectors of the automotive industry, particularly around diesel truck emissions control systems, where EPA has pursued aggressive enforcement actions against companies selling delete kits[1]. However I think where it will get really interesting is when we start getting widespread delete kits that aren't primarily mechanical in nature, but attempt to lock out or spoof entire software/electronic subsystems.
[1] https://www.epa.gov/enforcement/performance-diesel-inc-clean...