Choice of year schronization:
seasons: I've heard people say that the equinox is a more stable constant so it definitely has that going for it. The solstice was chosen because it is the darkest point (but only in the northern hemisphere). The new moon is at the darkest point and so is the day. I'm not completely convinced that the terran computational calendar should break with that standard, but maybe, the equinox would definitely be a more neutral location. But if we are staying on the side of neutrality then which equinox?
Well, I would argue for the spring equinox. It has long been used as the start of the year. The Romans used it as such (which is why September, October, November, and December are numbered so. They are the 7th to 10th month reckoning from a March start). It was used by the Celts, the Babylonians, Mayans, Germanic tribes, and a host of others. Stonehenge, Woodhenge, various "medicine wheels", even the Polynesians and Australian Aboriginal People had stone circles to kept track of the Equinoxes and the path the Sun made through the constellations.
By the way, the reason that the equinox is easier to determine than the solstice is this: On the equinox days, the sunrise is exactly due East and sunset is exactly due West. For the solstice, you have to measure the apparent altitude of the Sun, which varies by only a 60th of its apparent diameter from one day to the next near solstice time, and watch for the maximum (or minimum) altitude.
As far as starting at the darkest point, that's also not entirely true. I've already covered years, though I will mention that the Mayans used the winter solstice as their starting point. Lunar calendars generally start with either a full moon, or with the earliest visible crescent (consider the flags of the nations where lunar calendars predominate). This is because the new moon is nearly impossible to see, due to the Sun. As far as days, it has for the longest time been considered to be two step cycle, a twelve "hour" day and a twelve "hour" night (the hours were not of equal length). The day started at sunrise, and ended at sunset, though some cultures started the day at sunset rather than sunrise. In the early days, Roman timekeeping also started at sunrise. Time was kept using sundials and water clocks. Due to a quirk of Roman law, petitions before the courts needed to be made before midday (ante meredium), As the Roman empire spread across Europe and the near east, two things happened. Clocks became better, and noon became more important than sunrise. By the 4th century BC, Roman timekeeping had evolved to the current 12 hour day/12 hour night cycle with noon and midnight being 12:00.
january 1: If you're going to create a whole new calendar, I feel like keeping with a January 1st start date would be very confusing because you might expect the date to be a UTC date when it's totally not at all the same. But there'd be lots of confusion in ANY case. I know that TAI/UTC/UNIX uses January 1st, but besides that, do you know of any good reason to use January 1st as a start date other than convention?
Other than widespread convention, supported by numerous standards bodies and government decrees... well, no. However, I will throw this suggestion to you: If you're going to create a whole new calendar, and if keeping a Jan 1 start date is going to cause confusion anyways, then why not capitalize on that and use the spring equinox as your start point. I'm confident it will cause a lot less questions.
