Roland Tanglao has an excellent presentation of RSS Remixing. Listen to the podcast, particularly to the questions at the end; there’s an application that wants to be born.
What we really want to do is improve the signal to noise ratio. As communication becomes cheaper and more efficient, the volume of data has exploded, and will continue to rise exponentially. In addition fragmentation has also skyrocketed as it becomes easier to express an opinion through blogs, tags and spam. In short, the meme-space is severly fragmented and in dire need of defragging. How are we going to do this?
Read Umair Haque’s Edge Competencies on bubblegeneration.com for a background insight on how this next piece works.
Each user is effectively sorting signal from noise on a daily basis. Most are duplicating effort without any coordination to reduce the size of the haystack. The people who are helping effectively reduce the size of the web haystack is the team-up of Google and the bloggers. The emergent semantics of folksonomies has promise, but so far has been largely untapped. When I mention collaborative effort, I’m not thinking of mechanical turk. I don’t want to farm out the task of separating signal from noise to someone collecting micropayments for that task. Why? Because I’ll likely end up with garbage. Who better to filter my content than me, right? So how do I leverage others if not explicitly?
Implicitly. I already sort my own content daily, just save the results to a tag file. What if I had a tool that helped my with this solo task? Here’s the gist. Currently I bookmark or tag only a tiny fraction of the links that I find interesting. Tagging is currently too distracting from my task at hand, reviewing content. I don’t want to leave my page, or be limited to three tags, or go to a shadows page. I want the basic implicit context of my current page to become an automatically generated tag for all sub-links on the page, and I want an easy way to add tags such as toBlog, toRead, toSave,
, for, etc. The option to add these tags is in a hover-menu when I pause over a link (the implicit tags are auto-logged and tagged when I view the page; example auto-tags include timestamp, parent-page, email author/group, Google search terms, etc.). Now I’ve got implicit and explicit annotated history of all links that I come across.
How to use this history? Remember the toBlog tag, when I go into blog mode, if I search my taglog for toBlog, all those links come up, and I can sort by date, project, or other tag. While composing my entry, if I highlight a phrase, and hover over the phrase, I can create a link to one of these taglog links.
Here’s the cool thing. This tag-log is mine. I can export it to del.icio.us or shadows.com if I like, or keep it private. Hmmm. Private is last millenium. I want to share my content. Remember, I’m not just in it for me. I want other people to be able to leverage my meaning-making, so I want them to be able to read my log. If I’m shy about something, perhaps I can make specific entries wholly or partially private. Maybe I can make my taglog available anonymously.
In my own little taglogged folksonomy, links become somewhat relevent as soon as I’ve been exposed to them. They gain relevency when I tag them, visit them, or blog them. Other people can see the links that occupy my attention at various levels. Now, imagine a tool that was harvesting these annotated attention streams in the same way that google is currently doing page-rank.
What we’re doing here is turning every web citizen into a Maxwellian Demon, separating signal from noise for his own benefit and then sharing that benefit with whoever would like to use it. This is creation of value in the form of meaningful context. Collectively, this value will be astronomical. There is a little problem though. Right now, the web isn’t nearly stratified enough. Think of wikipedia. There is one entry for Bill Gates. We need to microchunk this. Currently, each little demon is fighting with every other little demon to push the content around. This is because we are used to thinking in the metaphor of physical proximity, and a thing can only be in one place at a time. Tagging has broken that metaphor, but we still fall back on it by habit. Contextual proximity has no geographical bounds, and therefore can be multiple places all at once. I “Bill Gate’s” microchunk of information might be simultaneously “funny” “serious” “factual” “disputed” “interesting” and “worthless.” Much of the value of folksonomy tags is their subjectivity, not their objectivity. If I can find other users whose subjective experience of a micro-chunk matches my own, then those matches could possibly be harvested across other micro-chunks.
In this way, I might want to subscribe to an RSS feed of “funny” Bill Gates micro-chunks. Or maybe subscribe to a particular user’s “funny” tag for other micro-chunks. Or maybe to some heuristically determined subset of users who’ve tagged this micro-chunk as “funny.”
There are almost unlimited ways that I can control my feeds now, and if I make it so that my incoming stream is easily adjusted by turning the volume up on signal and down on noise, we’ve got our app for harvesting collective intelligence.
I’ve probably moved ahead too fast. Bring it down several notches and just think of your known peers (not FOAF). What if you could just share your taglogs with this peer group? I could see what you find interesting before you blog about it. We could see who in the group hasn’t seen bubblegeneration.com. The work that we do quickly moves from having individual value to having collective value.
Effectively, what we’re doing is laying down multi-variable pheremone trails (see Emergence, by Steven Johnson), while we’re searching for content crumbs. I can easily change my state so that certain pheremone trails become invisible and others become highlighted based on my current intent and internal state.
There are other issues that I’ll address in a future entry. These include trust+identity commons, and compensation+reputation. These concepts are seen as difficult primarily because of paradigmatic habits of viewing resources as scarce. If we can break down this habit, we can see that attention is one of the best forms of compensation. So much so, that it can be monetized, not by Amazon or Google, but by you and me. In the coming omniarchy of an abundance economy based on attention, the citizen becomes king, and the corporation becomes servant. Don’t take my word for it. Look at the data. The prevailing worldview is flipping on its axis as we type…
- David