A car analogy! I like this one, too, because it lets me point out that most cars actually do change their performance characteristics when necessary, based on the operating conditions: when the engine is very cold, or getting too hot, or a different type of fuel is used, or a load is being towed. The idea is to keep the car moving as well as possible, given the circumstances.
If I punch the accelerator in the summer, my air conditioning cuts out to give me more power. And nothing pops up on my dashboard to tell me it's happening! "RAAAWR I MUST SUE TEH MANUFACTURAR/ I DID NOT CONSENT TO TEMPORARY WARM AIR!!!!111one"
The difference here is that, while the warm air is very noticeable, most iOS device "throttling" is not. As far as I've seen, any quantifiable impact on performance has been measured using a tool like GeekBench. The thing is, benchmarking tools are designed to peg the CPU to its maximum, which is where iOS begins limiting performance: when the CPU spikes and could cause an power cut. In real-world usage, you're not going to peg the processor that often, and you won't see a significant reduction in day-to-day performance. This is important because anyone seeking to win damages from Apple is going to need to prove that there was a significant impact to real-world performance to the point of harm or disability.
Regardless of the outcome, Apple, for their part, should expand the status in the "Battery" section of the Settings app to remove the ambiguity. They should inform the user how often the processor is being limited, including a trend indication of some kind, and translate that into plain messaging about how much that's actually likely to impact their usage. If the impact is significant, they should also let the user know that they can replace the battery to address the issue.
