> If that's even a possibility it's not responsible to allow DIY work in this area.

At least in the US, in most states, you're entirely allowed to do your own electrical work on your primary residence, and it has to meet the same standards, same inspections, same permits, etc, as any other install.

As far as grid tie and rapid shutdown, the only thing that really matters there is the inverter. If they're a UL listed inverter, they'll shut down properly when the grid drops, at least on the backfeed ports (some have automatic transfer switches and such built in to continue powering the house grid down).

Also, can you cite evidence of a DIY install actually killing ("plenty of times") a lineworker? The techniques I've seen for line work involve "dead shorting the phases and neutral all together before doing anything," which makes the grid segment quite impossible to energize for invasive work, even if someone somewhere else screws up. And I've seen some reports of generator based backfeeding injuring or killing lineworkers (who generally were half a dozen safety protocol violations in at the time), but I've not seen any related to solar - because grid tie inverters simply don't work that way for any I've found.

I did a brief search and found a case of a line worker dying due to someone backfeeding a gas generators after a hurricane:(https://www.oshrc.gov/assets/1/6/06-0166.pdf)

Clearly this could happen with a dumb diy solar install, but also the case text shows that the company knew this was a possibility and enumerates several ways this accident could have been avoided.

I've seen that one before, and... as I mentioned, quite a few safety protocol violations in, bad things happened. They didn't force the line dead by grounding it on both ends, they didn't test to see if it was actually energized, and they didn't use the proper equipment or procedures to work on it as though it were energized. You can get away with that for a while, but at some point, that's going to bite you, and given power line voltages, bite hard.

But I've yet to see "plenty" of deaths attributed to DIY solar. A downed grid segment looks like a dead short to an inverter, and anything even faintly UL listed (even with a "UL Listed" sharpie marking on the box) shouldn't be feeding into a totally dead grid segment at all. Most of the grid tie inverters lack the ability to create a waveform in the first place - they can only feed into an existing waveform.

Does this basically mean that the only real piece of equipment that needs to be rock solid is the transfer switch, and not necessarily the whole system consisting of panels, batteries and inverters (for grid tied systems)?

You could mislabel lines that stay live after grid disconnect as though they were on the grid side of the cutoff switch.

Making truely idiot-proof high voltage wiring is difficult. However, that's what inspectors are for (in theory).

Also, solar panels are heavy and pointy and often stored on roofs where the wind can grab them. Again, inspectors should check this.

For safety, yes. For performance, no. Cheap Chinese inverters can create all sorts of nasty effects on the local distribution segment, but IMO performance requirements should be up to local distribution utilities when they allow the interconnect, not a matter of national concern.