Knowledge Management: It’s All About Granularity

In my pursuit of the ideal collaboration platform, I’ve tested a few knowledge management systems lately: Knowledge Plaza, Seemy, a combination of del.icio.us and Twitter. And those tests were very interesting because they allowed me to spot the main common problem they all share.

How many times have I heard that the Google Wave presentation is too long, leading people to simply not watch it at all? How many times have my friends complained to me about the length of my own blog posts? The granularity of information on the web is simply too big. The web is all about resources, and there are billions and billions of these resources out there, and what makes it even harder to process and integrate them is that each resource mixes a lot of different information items.

And for me, THIS is the nightmare for my technology watch, and for knowledge management as a whole. You can comment on or share whole web pages through links, whole Youtube videos through embed codes, whole discussions through podcasts. But what if you want to extract what is to you the essential part of a blog post, the funniest moment in a video? Well, let’s say I don’t know any solution for that.

For my everyday technology watch, what I would really need is a knowledge management platform that allows me to select small chunks of information in text, video, audio or images, and then tag those chunks, comment on them, and store them somewhere in the cloud for sharing them with my friends or colleagues, or simply keep them for myself for later reference. All of that while keeping a link to the full original resource of course. That would be awesome!

Now of course because I love to solve problems, my next move is to think about a solution. I don’t know any existing system that does that, so if you do, please tell me about it. Now if it doesn’t exist, we have to invent it. And the way I see it, there are two main aspects to this system.

The first issue is how do we capture excerpts out of web resources. If we want to make it as simple as possible, we need to integrate deeply with a web browser in order to create a natural user experience based on drag-and-drop selection, keyboard shortcuts and so on. This is why I’ve tweeted about me looking for a Firefox extension developer to help me out: I’ve never developed any Firefox extension myself, and I could learn but (a) it would take much time and (b) I’m not fond of Javascript. So once again, if someone out there is a Firefox extension developer and would like to collaborate on this experiment, you are welcome. Let’s try first with text, we’ll see later for other kinds of multimedia content.

And the second issue is how do we store and present all this information in a highly usable and intuitive way, without being too disruptive, without inventing too many new concepts. This part I can handle. I already have a few ideas.

I think before the Internet, there were technology watch departments in companies, whose job consisted in cutting out paper pieces in newspaper, pasting them and composing press reviews with comments and writing reports about what competitors were doing. Nowadays, it’s as if we just gathered full articles or newspaper pages, videotapes, full interview transcripts and just put small post-it notes on them. It’s just too rough, not pre-chewed enough, not efficient enough. And as always, there’s gotta be a better way.

What do you think?