Those people are all wrong
I write software for a living, but on very resource-constrained devices. My job involves getting a lot of functionality out of not a lot. And here I am holding this device which runs at 20x the clock speed of the stuff on my bench, with four main CPUs and an array of GPU cores, plus 100x the memory on a DDR3 bus, and there it sits, thinking about who-knows-what for 15 seconds, while I’m waiting for it to do something trivial like render a page of text. FFS. 15 seconds is enough CPU time to forecast next week’s weather and play a few games of chess during idle cycles.
I get it. A general-purpose device demands a lot of general-purpose software overhead. But the way I see it is this: if my fingers can move faster than your UI can update the display, then you, the UI designer, got something wrong somewhere.
I don’t think it’s just me with one underpowered device: I’ve had several that do the same thing. A bit like Windows, they tend to start off OK and then become increasingly unresponsive over a period of months, even with a junk-cleaning app.
Then there are the inexplicable reboots, the occasional failed shutdown (which drains the battery and ruins it), hit-and-miss WiFi connectivity, touchscreen malfunctions in the presence of minor EMI or damp, pointless updates to the OS that break compatibility … maybe I’m a crusty old technophobe, but seriously, just give me something with clicky buttons on it and a proper RTOS underneath.
For what it’s worth, I think it started off as a very solid concept. Then somewhere along the line, somebody decided to make it like Windows, except more painful to develop software for. They forgot what sort of hardware it was going to run on, and what it was going to be used for.