There is still so much more to do that won't require gargantuan cpu/storage/io/screen/etc. resources, that I don't think we are in an "end of history" phase yet

There is still so much more to do that won't require gargantuan cpu/storage/io/screen/etc. resources, that I don't think we are in an "end of history" phase yet with the current conceptualization of these devices. Just a few personal itches off the top of my head that have yet to be scratched: * There is still not a good solution for group-transactional data over the devices as a first-class citizen feature in the system. Live, collaborative editing with versioning of just a list, for starters, is considered an app'able feature worth $50 USD per year per person. * Internationalization still takes a back seat. Calendar apps still don't pick up that if you set an appointment for someone in a different time zone than you, you are likely scheduling to a time convenient for them. There should be an option to auto-detect and display applicable time zones as you create the appointment. * Customization still takes a back seat. When I add custom phone types ("concall", "support", "after-hours support", etc.), it ends up UNSORTED in the list of custom phone types when I go to choose one. That's just the tip of the iceberg of enhancements. For address records of people in a specific organization, I want to be able to set only a certain list of custom phone types for anyone in that organization. * There still is no sense of tracking versions of information. I have thousands of contacts, and I keep track of old addresses for example; it helps me orient myself to the possible cultures someone has experienced. But only the here-and-now data is recognized. * Linking data between apps takes an explicit act of coordination between app developers; the Newton data soups were particularly good with breaking down this artificial barrier. * Deep-linking into your own data from within an app in general is not very well-recognized as a use case. Not to speak of deep-linking from an external laptop. I'm personally gravitating back towards making my information in my Emacs Org mode buffer the primary system of record, and the phone is just a thinly-sliced perspective of that.