I woke up this morning and felt like blogging. It’s been a long time :)

We’ve had Reboot, FOO Camp, and a bunch of other opportunities for rewarding conversations and so it feels like I could write posts for a week about all the new or further developed thoughts and ideas that are crowding my head. I’ve touched on some of those in recent talks (here’s video&slides from one), but I haven’t gotten around to blogging about them yet.

I’ll start by jotting down a few notes on questions that might be useful to ask when evaluating the potential of something to be turned into an online social object. As a disclaimer I guess I should say this may not make much sense unless you’re familiar with the previous posts on the subject (a number of people have suggested I revamp this blog to make it easier to navigate the material – I’ll eventually get around to doing that).

• How well does the potential object yield itself to breaking it down to structured data? For instance trips can be pretty easily structured, as on Dopplr. Dopplr trips have only three key data points: a start date, end date, and a destination. Each one is expressed as a discrete. It got too complex with free text entry for destinations, so they decided to use cities with over 80,000 inhabitants as a proxy (so when I travel to San Sebastian in spain, I need to pick Bilbao on dopplr).
• What data points to pick? You want to pick the data points that are sensible definitions of the object and give you the most interesting handles for generating sociographs. This is tricky because the more data points you introduce, the more fine-grained sociographs you can generate, but the more complext the system becomes. Events (as on Upcoming) are already more complicated than trips because you want a title, start and end date & time, location, and some kind of invitation policy. It’s a bigger usability hurdle but the tradeoff is reasonable if the assumption is it appeals to a broader population of users. Like on Upcoming vs. Dopplr, more of us go to events even if we don’t travel that much; events are more interesting than just who’s traveling where.
• How often do people generate new instances of the object? This question should replace the "who’s your target customer?" question because your main target are probably the same people who generate lots of instances of the object. If a lot of people generate new objects often, ads+subscriptions probably make sense. If they don’t use it that often but the social networking adds a lot of value (as when looking for a book, car, real estate), then you need higher-value ads and/or transactions. If it’s the sort of object that few people create, but those who do do it lots, you’re probably talking about a hobbyist or professional audience (e.g. Dogster for petlovers) and might be able to tap into its special channels to figure out a business model
• How much social gravitational pull does the object have? Complex social objects offer a lot of handles for discussion. A movie, for instance, has a cast, a plot, special effects, and plenty of other conversation points that people can talk about. Simple social objects like microblog posts don’t have as many such handles. On microblogs like Jaiku it’s typically someone just asserting something, to which you might or might not be inclined to reply. Big social objects have more social gravity. Movies attract viewers and conversation like stars and big planets attract matter from space. Tiny social objects are more like a meteor shower; each one has very little gravitational pull as such, but when you add up all the tiny particles in space, they embody more total matter than the big constellations.