michael pucher

 Decentralized location based services

I’m sure you read about the dispora* project and the buzz it created. Backing on kickstarter is around 1800% as of May, 21st. diaspora* came along the right time the right place. It isn’t sure whether they will meet the high expectations or if they ever gain mainstream but lot’s of users are willing to take back control of what they share online. People are worried and do care about privacy and publicness.

Location based services like foursquare, gowalla and a lot of clones are considered (or have been at least) the next big thing in social networks. But those services have the same drawbacks like current mainstream social networks. They are operated centralized and users aren’t in control of their data. And all of those services build their own proprietary places databases.

I’ve been a user of foursquare and other services in Germany from the early days. Though user adoption isn’t that high in Germany yet, the biggest advantage of using LBS services is still connecting/meeting people. Meet colleagues or friends at the airport (business approach) or somewhere in town (private approach) is surely a plus. But we don’t need centralized authorities to do that. We don’t need a “man in the middle” who spies on us and sells our profile data to advertisers. We can connect on our own. As Eben Moglen said: “we should keep the logs” (NYU Speech “Freedom in the cloud” on youtube).

It’s time to create a decentralized lbs connection system for which I’m going to lay out a first conceptual draft now. Feel free to comment/get involved.

Places database

We need a central, royalty free places database. There has been discussion about a commercially created “Open Places Database” on TechChrunch but it seems nothing happened until then. Another yet better option is OpenStreetMap, an wikipedian style map of the world. The third option would be to use Google Maps and/or Google Places.

Private check in repository

Why check in using a service that keeps a repository of our data that we don’t own/control? I don’t want that. We will create repositories and check in services that store our records and we control what we share. The repository could be part of a diaspora* seed, a wordpress plugin or run on a google app engine account. The possibilities are endless. As I’ll describe in the discovery and connect section, that tools could be used freely by everyone (including businesses without anyone seeing what the employees are doing. No man in the middle or public visibility).

Checkin Clients

Clients to check in could be web based services, smartphone clients, html5 based web apps or could even be built using twitter or other scenarios.

I’ve created a mock up of a possible android client:

As you can see, the user is in full control of what is shared and who is able to view the data. Only friended seeds/users of your blogging community who are tagged “family” or “network” would be able to see your check in.

The system uses tags to grant access rights to your check in records. The tagging should be part of the underlying infrastructure such as diaspora*. Client check in data would be transmitted using Json or XML and would contain access tags.

Example:

There could also be your own tags to create special interest groups. And there could be a “public” group where posts would be visible without being friended.

**Sharing, discovering and connect **

By using tags to classify check ins it would be possible to create individual views on your check in data.

Examples:

Private stuff: You could always tag your check ins with “family” so they would always know where you are.

Work: Employees of a company could check in their work locations so their boss has a view where e.g. his sales force or service technicians are at the moment. But the boss wouldn’t see private stuff (like he would do with services like foursquare right now)

Network: going to a fair? Great. Tell your network that you’ll be there. At an airport? Tell your network and let’s see who else is there.

[Update 2010-05-29]

Emergency: someone I follow on twitter wrote that he is in search of a friend who has been brought to a hospital but he doesn’t know which one. The system described above could be used as an emergency information system as well. Imagine an “emergency” group with only your closest relatives, girl-/boyfriend and maybe your boss. In case of an accident, the hospital could send a broadcast to all nodes/seeds with encrypted name, birthday and other available information plus the hospital location and the node in charge could take the information and send a notice only visible to members of the “emergency” group.

Conclusion

I hope that more information becomes available on the diaspora* project to get started soon.

It can be done. We don’t need the man in the middle to connect. Let’s keep the logs and start to connect!

Feel free to comment.