Recently I heard a phrase that I now struggle to dislodge from my brain: “Obsolescence is opportunity.” On the surface, it is the polar opposite of how most of us view it. “Obsolescence” is for many a dirty word, a potentially catastrophic risk that we do our best to pseudo-manage, but in reality, we have little control over it. Attempting to address past failures, the embedded industry has been crowbarred into guaranteeing longevity of products, but with those containing hundreds of components from a multitude of vendors (often technology behemoths where your product is nothing more than an irrelevance on the global manufacturing scale), no one can ever truly guarantee a product’s availability in its exact current form.
That latter point is how vendors typically manage this. The longevity guarantee is not at the component level; it’s at the functional level, analogous to how retail motherboards are produced. This suits developers who implemented flexible operating systems using hardware abstraction layers, as their software doesn’t speak directly to the hardware.
For those utilizing more niche operating systems, arguably also definable as “embedded” in this context, where they’re communicating directly with hardware, such component changes can be hugely problematic. Consideration also must be given to the stringent approval requirements of specific industries under the embedded umbrella, where certification can be as specific as the exact components used within an embedded system, and changing any must trigger a costly re-approval loop. Finally, at the vendor level, although some drop-in replacements exist for components, that is far from typical. Components have their own NRE costs that they must sufficiently cover, so even risk of obsolescence drives up price somewhere.
So how can obsolescence be an opportunity? I’ve considered this concept and concluded that obsolescence drives us forward. Embedded vendors are businesses, and while their employees may be passionate about driving technology forward, at top level this can only be justified if it generates more profit; it’s never technology for technology’s sake. If company X is selling a decades-old product Y – generating higher profit than they perceive a replacement product could and with no development costs to consider – would they ever replace it? I suppose through this argument it’s obsolescence of old technology that drives innovation of new technology. It’s culling the weakest members at the rear of the herd that improves the genetic pool. Does that make those demonizing obsolescence analogous to the fallen prey? Perhaps that’s an analogy too far.
Nowadays I see embedded products announced as end of life well before any components contained within are obsolete. I genuinely do think our industry probably more than any other has an inherent desire to innovate and prove the value of embedded technology globally. With competition among embedded vendors greater than ever, no one can afford to rest on their laurels and risk sitting back purely collecting revenue from yesterday’s technology. I believe it is competition rather than obsolescence that drives innovation today, but obsolescence is that safety net, gathering up and quietly retiring anything that is left behind.