I've also been in a relationship and "fallen in love" or something close to that. Sadly, nothing during my foray into the emotional world of relationships turned out all that great: I start paying child support in a week; Keira's mother - Keira being my progeny - doesn't talk to me; and the girl that I fell in love with wanted children while I prefer otherwise. So, for now, I'm back to focusing on my self and on the observation, understanding, and development of humanity.
Back to personal goals. One enjoyable way to look at the world, and humanity specifically, is in terms of decisions. Therefore, a central architectural pillar in the software I want to write revolves around how decisions are made, how to measure their effectiveness, and how to create a feedback loop that balances decisions while allowing humanity to retain the importance that we seem to place on conflicting goals such as autonomy/collectivism, reason/passion, quality of life/quantity of life. Imagine the end result as some odd human-computer hybrid system wherein the human piece is made up primarily of voluntary individual contributions and the computer part is what aggregates and processes the human input. Of course, because the computer part is originally derived from the human part, we could consider the combination to be entirely human, and as man and machine continue to merge at the physical level, that interpretation will likely gain further acceptance.
A good place to start is probably developing a data model for decisions, and a software service that's commercially useful and tracks decisions. So far, I like the open government model for creating open source, service-based software solutions; but funding, marketing, and developing such a product/service, and especially such an organization that guides the evolution of that product/service will be challenging. Openness will be very important, as will security, which means privacy for the users: openness of the creators and the system, privacy of the users - a tenuous balance at times when creator and user are at times mutually inclusive.
Including metrics for decisions will be interesting because we currently focus on data to drive decisions, but really we can also focus on data to analyze prior decisions and desire, in addition to data, to drive decisions. That is to say that opinion polling - in addition to decision tracking - should be analyzed over time as a mechanism to provide insight into how humanity evolves. And finally, for context, we probably want to include some basic objective measurements of our physical environment so that we are able to analyze how humanity evolves with and ties to that physical environment.
Create a collection of demographic, environmental, opinion polling, and decision making data. Design a fun and useful user interface to engage individuals in collecting this information. Provide meaningful analysis of the data so that we can learn and grow as a result. Target government decisions, the general public's opinion, and scientific measurements.
Drawing on others: Multivac (Asimov) = Watson (IBM) + Quora + US State Department + ...
