Maybe we can't? I have been wondering about that for quite some time now. We have been talking about ubiquitous computing for a long time, but instead of doing

Maybe we can't? I have been wondering about that for quite some time now. We have been talking about ubiquitous computing for a long time, but instead of doing that, it seems we have settled on using smartphones for everything, retiring specialized or underpowered devices in their favor, even if they are just 60% as good as their predecessors. When a company tries to build something in this niche, it seems we need as much power as possible to make it potentially as mass appealing as possible, which in turn means tradeoffs in many other areas (size, heat, battery life, software etc.) that make it unappealing to most but a small minority. Now, it would seem easy to create a specialized device by putting some constraints on it, right? Make a computer optimized for writers! Make one optimized for day planing! Make one that is not good for playing video or games but great for spreadsheets etc. These devices could excel at their given niche and give us back advantages like weeks long battery life or incredible small size etc. But it seems, no one is willing to work with these constraints. Of course, also no one seems to be capable to create custom software, which I guess these devices would also need to be good at what they were supposed to do. Sometimes you find these things in small, specialized markets, but even then they always seem to be not really much better or even worse than what we had in the 90s. I am not sure if it is because of market dynamics or lack of vision, but all we ever seem to get are compromises.