There Is No Geolocation Social Paradox

I think that MG Seigler is dead on whe he observes that geolocation services are exploding right now, but he goes off the rails when he suggests there is a ‘location social paradox’:

A few weeks ago, our own Jason Kincaid wrote a post about how Facebook is poised to take over the geolocation space. In it, he makes a number of good points, but there’s one that’s particularly interesting to me. “At most, there are probably a few dozen people who you’d like to share your location with,” he writes. Overall, that’s likely true to a varying degree depending on who you are, but it points to a larger problem I’m starting to notice with these location services: The more people you follow on them, the less useful the service is. This is location’s social paradox.

I’ve written before that location is the missing link between social networks and the real world, and I absolutely believe that’s true. But the way most of these location-based services are built right now, they are becoming an unmanageable mush of finding the location of the people you actually care about. Perhaps the most popular of these networks right now, Foursquare, is a perfect example of this. When I was following 20 people on the service, it was very useful. When I was following 50, it was still useful, but there was some clutter. Now, at around 250 people, I find myself scrolling through my stream just to find certain people that I actually want to know their location. I cannot even imagine what Scoble does with the 1,700+ people he follows.

Of course, this problem is entirely my own fault. If I don’t want to know where someone is, I shouldn’t follow them. But there are two problems with this. First, there are some people that I would like to follow some of the time, like if we’re all in a different city at a conference together. Or maybe if I’m just bored and looking for something to do on a particular night. Second, and more importantly, today’s social networks carry a social pressure to accept many people that request your friendship. Again, of course you don’t have to, but not doing so can often be misconstrued as a slight on that person. Bigger picture: Today’s social networks are predicated on the idea of “more.” The more friends you have on these networks, the more social you are, the better you are at the service, the bigger ego you get, more, more, more. None of that is true, but the perception (as it is with most things in the world) is that more is better.

With these location-based social networks, more is actually worse, and that’s awkward.

via MG Seigler www.techcrunch.com

This is assertion has an oddly retro flavor to it. We are awash in yet another wave of essays, suggesting that information overload is sucking the juice from our minds (like Nick Carr’s Hypermultitasking, which I have yet to counter). Here we have MG Seigler worried about geolocation overload.

First of all, the notion that you’d like to limit your location to about a dozen people is simply false. The issue is one of scale, as in scale on the map. In general, even the most risk averse person would be willing to let the wide world know what continent they are on, and perhaps they would be open to broadcasting country and city. (Except for the people paranoid about their houses being burgled.) What Kincaid (cited by Seigler) probably meant was that ‘At most, there are probably a few dozen people who you’d like to share your precise location with,’ which is a very different proposition.

So, the first challenge posed by geoloco publishing is granularity: people what to limit fine granularity to a few, and then match less fine granularity with less close friendship. I will state this as a rule:

First Rule of Social Geolocation: The possible closeness of map location presented to a contact should match the degree of closeness in the relationship. 

This is bound to become an axiom of social geolocation. The trick, or magic, is going to be making the ‘closeness/closeness’ mapping simple. Note the weasel wording of ‘possible’. Manually deciding who should see your current location every time you update is way too hard, and it is equally difficult to manually assign some level of closeness each time you add a new friend.

Likewise, the complementary problem presents itself to us as recipients of others’ geoloco information: based on where we are and what we are up to, completely different subsets of your geo info are relevant.

This turns out to be a transposition the privacy problem (see Facebook Is Dumbing Down Privacy, But Really We Need ‘Eddies’ In The Stream), and I will offer a similar general approach. Some will suggest the use of defined groups to handle this mess — put friends into groups, and allow different groups access to your geo info, or use those groups to filter others’ geo info for you. This is a possible solution, but a better, and more bottom-up and elegant one, is using filter tags (like hashtags, only with a ‘+’ instead of ‘#’) as a way to limit visibility to incoming (others’) and outgoing (your own) geo location. Note: filter tags are not yet implemented by Twitter or other services: this is conjectural and conceptual at present.

Imagine that you can apply labels to people, for example (using Twitter style input), and that applications could listen to that, including perhaps Twitter:

@briansolis +sf +wine +pals

and imagine that I might do the same with other contacts. At some point I might post my location (again using Twitter type ‘tweet’, but this could be done through other UX), and I could direct the updates to the corresponding social circles:

Enjoying a nice Infinitus at /Mexico DF, San Francisco CA/ +wine +sf

The intention of the filter tags in that post would be to limit the geoloco info embedded in the geoslash (‘/’) — that I am currently at Mexico DF (a great place, by the way) — to only those people I have denoted as +sf or +wine. If I had wanted it limited to only people that both are wine oriented AND live in San Francisco, I could have separated the filter tags with a semicolon, like ‘+wine; +sf’. The default is OR.

Of course, I could simply post the tweet with no filter, which would have it transmitted to anyone following me, but is is the work of a moment to add a tag or two, to filter your location to only those you’d like to see it.

Note also that similar filters can be used to determine what you’d like to have flow to you, so that while in San Francisco you may opt to only see the geoloco data of +pals OR +sf contacts. (Note I am studiously using the logical OR and not the ‘and’ of everyday conversation.)

This exposition points out the issues inherent in building geolocation into a service like Twitter. I suspect the great majority of people will turn that off, without reasonable privacy protections. What is needed is an intentional, fine-grained control, like those that I am proposing with user tags and filter tags. This would also seem to satisfy Seigler’s concerns about overload, so I don’t think there is a social paradox around geolocation, just an inadequate model of filtering.