After a flight that killed my ears (I have a terrible cold, and cabin pressure doesn’t help), Elmine and I ended up at BarCamp Vienna. It turned out someone had already put my name up on the roster to do a session. That’s why at the end of the afternoon I found myself presenting, without hearing much. I felt unfit, and talked way too fast I guess.

Basically I told the same story Boris Mann and I did a session about at last weeks BarCamp Brussels. And at the end asked for thoughts, ideas and pointers that might help create social software tools that are better at using human relationships for navigation, aggregation and views/slices.
Two things were put forward at the end (people were tired at the end of the day, and I really did talk too fast) before discussion died down.

Sebastian Fiedler put forward that 3D environments might be a way of presenting information/networks in a different way. This might be a route to break out of the more usual notions on how to present information(patterns).
I don’t see much of that at this point, but that doesn’t mean much. What I do see is how e.g. in Second Life attempts are made to connect SL to the information landscape of the internet, similar to how at the same time efforts are made to connect the information landscape to the physical landscape more. A set of parallel processes that might cross-fertilize.

Another participant mentioned a Linux tool called Dashboard, that by its description sounded a bit like 43people.com, a way to build portfolios around individuals. That is interesting as a first step, but I would like to be able to do the same for groups and communities. In a metafor I used in Barbara Kieslingers blog, and which I mentioned today as well: I do not need to know which trees are in a forest to know if the forest is on fire. Only when I really want to know which specific trees are burning, do I need to zoom in. Communities are my forests in that regard, and I only need a notion of what is important to a community at some point.

Sunday BarCamp Brussels took place in the beautiful venue SAP Lounge. A good number of participants, most staying on after the Drupal Conference in the days before. I’m not a coder, open source developer or such, and these meetings are very much about that, but I like to go to these events to talk about what I expect and need my tools to do, and how they might be improved.

Boris Mann and I finally had the more extended conversation we missed out on during the Amsterdam BarCamp last year, and decided to turn it into an impromptu session in the so-called ‘passionate users’ stream. I certainly qualify for that! In 20 minutes we whipped up a couple of sheets and examples, announced the session, and sat down with a group of about 15 people to discuss.

We talked how my strategy to deal with information abundance, and preventing it from becoming information overload, is to look at the human relationships behind that information.
It seemed not for all present it was immediately clear that with relationships I did not mean relationships between information objects :).

By shaping your social networks and context as one of two information filters (the other being your own passions, objectives and tasks. Representing both inside out and outside in perspectives). For really doing that well, you need to be able to evaluate relationships more thoroughly, preventing echo chambers in your social network, and adding enough variety of perspective and diversity in background. You need to be able to track the available feedback circles (they are important for patternbuilding) and keep them from being too much one-sided (turn into echo-chambers). Our tools therefore need to put people much more at the center.

Basically we tried to bring across the notion that most social software, although keeping the human visible, still put the information very much at the center. Which is understandable since it functions as the object of sociality, but a lot would be gained if you would make the aggregation, different views, and navigation a lot more people centered. Then it would allow us to put the information abundance to good use even more, by letting us really do social information filtering.

Take RSS readers as an example, where I have different feeds tracking different pieces of context around a person. (I don’t track information in my feed reader, I keep in touch with people by tracking the various information traces they leave on the net) Why not be able to create a portfolio like structure as they have at 43people (which both the people themselves as well as you can add stuff to)? But then take it a step further. Build network pictures and use it as a navigational device, add in semantic analysis that brings the flow of conversation in to view. (last two examples taken from the work Anjo Anjewierden is doing)

So in short, I’d like:

  • tools that put people at the center (make social software even more social)
  • tools that let me do social network analysis and navigate based on that (as I already called for at GOR 2006)
  • tools that use the principles of community building as principles of tool design (an idea I had writing my contribution to BlogTalk Reloaded)
  • tools that look at relationships in terms of social distance (far, close, layers in between) and not in terms of communication channels (broadcasting, 1 to 1, and many to many)
  • tools that allow me to shield or disclose information based on the depth of a relationship, relative to the current content (like Vox and Flickr attempt to do rather clumsily by having fixed layers of contacts, friends and family)
  • tools that let me flow easily from one to another, because the tools are the channels of communication. Human relationships don’t stick to channels, they flow through multiple ones simultaneously and they change channels over time.

Now this is all complex. Not surprisingly, as human behaviour tends to be that way.
The challenge is in finding a set of heuristics (not unlike the community design principles perhaps) that allows us to keep the tools essentially simple. And then let the people using the tools build their own unpredictable complex and complicated patterns with them.

One of the new insights that I took away from the session was that this puts the whole identity on-line discussion in a new light for me. An angle that is well worth exploring, and I am glad that both Boris Mann and Gunnar Langemark showed themselves willing to give that angle some thought. My gut feeling here is that we do not need any form of objective identity in the passport like sense, but something that is embedded with the context where it is used. Which makes it simpler I think. Similar to what Jimmy Wales said at Reboot 7: I don’t need to know who you are exactly, as long as I am able to know you in Wikipedia. Meaning that a persons behaviour pattern is enough identity if it is consistent within the context relevant to you at that point.

Big kudos to Peter Forret and team for organizing it all, and also to Skynet who made the excellent venue and excellent catering possible (thanks Clo!)

On to BarCamp Vienna next Saturday to bring the same message. Or perhaps to talk about the need to punch more holes from the information landscape to the physical landscape and vice-versa (which is the ultimate barrier for flowing in and out of tools/channels while interacting with others).