> The impetus is on you to decide when things have started to look uncanny. “I wish that would be addressed by the industry — like, maybe if it reached a certai

> The impetus is on you to decide when things have started to look uncanny. “I wish that would be addressed by the industry — like, maybe if it reached a certain light-loss factor, it would just shut down, you know?” Nelson said. “Or if it shifted in color past a certain point, it went into failure mode.” This is complicated, and I don't think the consumer knows what they're asking for. From my understanding (and watching a lot of Big Clive lightbulb teardowns on YouTube), this would require an active sensor in the bulb. Most of the circuits in these bulbs are embarrassingly simple - 3-4 mostly passive components, and 1-2 silicon based chips or raw transistors. If you add an active sensor to that system, your cost balloons significantly. Then you have to calibrate the sensors. Then we get into the "printer cartridge" problem, where "my light bulb won't turn on because it's insufficiently cyan, but I only want a red light." We didn't know that we wanted things when we had incandescent bulbs. Now that we're being forced to switch away from incandescent bulbs and use a new technology, users are able to ask for things. I think that's partially exciting (users having preferences is good!), but it's also potentially complicated by companies not providing low-cost solutions that are as good as the old thing. So, overall, good and bad.