if I can't pick a simple piece of software and use it in a basic way after 5 minutes, then I think it's poorly designed.
I've thought about this, and come to the conclusion that it's true if and only if the software is designed to be easy to learn. It's what Apple are good at, and something many software writers try to do, with varying degrees of success.
But it's not the whole story. There are plenty of pieces of software that aren't designed to be easy to learn, perhaps because they're designed to be highly efficient to use (eg. vi, or JAWS), and the trade-off for that is a learning curve. Or because they're designed to be versatile, and the trade-off for that is a complicated syntax (eg. awk). This doesn't make them poorly designed, it's just a different design objective.
And other software is complex, which means it takes time to learn. A piece of CAD software, for example, could have a lovely intuitive UI, but you're still going to have to take time learning how to use it to design things effectively. I'd put the eTrex firmware in that category. The UI itself isn't actually that complicated (it's just clicking around some menus and learning some terminology), but inherently you're learning how to use a complex navigation tool, and it's going to take time and experience in the same way that learning to navigate with a compass and map would (indeed, many of those skills are the same).
You could complain that an eTrex is much harder to use than a TomTom, but it's a bit like complaining that a car is harder to use than a bicycle.
A more legitimate complaint is that all the GPS navigation tools for bicycles are complicated eTrex-style swiss army knives, and why aren't manufacturers making simple ones that do one thing well. Market forces is probably the answer. There are plenty of simple to use devices for people who just want to know how far they've ridden (these generally just use a wheel sensor, rather than a GPS receiver). There are simple to use devices for finding the way to an address on the other side of town - you've probably got one in your pocket - they just aren't bicycle specific, because there's generally no need for them to be.
A simple to use GPS device that's optimised for audax is possible, but only likely to happen as an open source project; there's no money in it for commercial manufacturers. Hardware is expensive and difficult, so unless someone works out how to write custom firmware for a Garmin or similar, it's likely to be a smartphone app, with the limitations inherent with that hardware. More practically, open source projects tend to happen to fulfil the author's personal need for some tool. If J Random Programmer needs a GPS device for audaxing, they're likely to take the path of least resistance and learn to use a Garmin swiss army knife effectively, rather than embark on a complicated software project that's going to involve a certain amount of cat-herding vis standardisation of route data to be useful.