I haven't waded into all this, and it's been years -- and years -- since my education touched upon systems that physically separate operating instructions from data memory.
But... sooner or later, it seems, we are going to go (back) there.
Instructions will become truly privileged, physically-controlled access. Data may go screwy -- or be screwed with -- but this will not directly affect the operating instructions.
Inconvenient? As development becomes more mature, instructions will become more debugged and "proven in the field". Stability and safety will outweigh ease and frequency of updates.
My 30+ year old microwave chugs along just fine. It doesn't have a turntable nor 1000 W, but I know exactly what it will do, how long to run it for various tasks, and how to rotate the food halfway through to provide even heating.
My 34 year old, pilot-light ignited furnace worked like a champ, aside from yet another blower motor going bad. I listened to the service tech when he strongly suggested replacing it before facing a more severe, "winter crisis" problem.
The new, micro-processor based model is better in theory (multi-stage speeds, and longer run times for more even heating). In practice, it's been a misery. The first, from-the-factory blower motor was defective. When that was replaced, the unit started making loud air-flow noises periodically.
Seeing the blower assembly removed, its constructed of sheet metal. The old furnace, by contrast, had a substantial metal construction that was not going to hum and vibrate if not positioned absolutely perfectly and with brand new, optimized duct work.
Past a point, reliability starts to -- far -- outweigh some other optimizations.
This is going to become true in our field, as well.
But... sooner or later, it seems, we are going to go (back) there.
Instructions will become truly privileged, physically-controlled access. Data may go screwy -- or be screwed with -- but this will not directly affect the operating instructions.
Inconvenient? As development becomes more mature, instructions will become more debugged and "proven in the field". Stability and safety will outweigh ease and frequency of updates.
My 30+ year old microwave chugs along just fine. It doesn't have a turntable nor 1000 W, but I know exactly what it will do, how long to run it for various tasks, and how to rotate the food halfway through to provide even heating.
My 34 year old, pilot-light ignited furnace worked like a champ, aside from yet another blower motor going bad. I listened to the service tech when he strongly suggested replacing it before facing a more severe, "winter crisis" problem.
The new, micro-processor based model is better in theory (multi-stage speeds, and longer run times for more even heating). In practice, it's been a misery. The first, from-the-factory blower motor was defective. When that was replaced, the unit started making loud air-flow noises periodically.
Seeing the blower assembly removed, its constructed of sheet metal. The old furnace, by contrast, had a substantial metal construction that was not going to hum and vibrate if not positioned absolutely perfectly and with brand new, optimized duct work.
Past a point, reliability starts to -- far -- outweigh some other optimizations.
This is going to become true in our field, as well.