Building Blocks

Intelligent Scoring is about discovering and sharing value.  The observation and interpretation of huge numbers of events behind the scenes in an active community allows the engine (the “Content Valuation Engine”, or CVE, that does the work of Intelligent Scoring) to make value judgements about the content and users in the community.  We call this process “valuation”. 

Valuation is how Intelligent Scoring works.  It is required.  The process of valuation results in values for a small set of key value indicators, we call “Building Blocks”.  There are three such building blocks — score, credibility and proficiency — each of which plays a role in indicating how valuable content and members in the community are.

Score is the easy example.  Let’s say you publish a blog entry, forum post, video or some other piece of content into the community.  You’ve categorized it, so it’s easier for people to find.  And a set of search indices have been updated behind the scenes, so searching will now include your new content item.  But what else happens?  Many times, not much.  Maybe the number of views or downloads is tracked, or people get to vote on how well they like what you’ve created.  But that’s about it.  Other than that, if you find it in a search, great.  Otherwise, you’re alone with your “My Library” content list.

Not so in the world of Intelligent Scoring.  Rather than just relying on a clever search to dig up your newly submitted item, the CVE monitors it.  Every activity relating to you or your item is tracked (even inactivity is observed), and each time something happens (or doesn’t happen), the engine interprets what’s going on and modifies the content item’s “score”.  Fast forward a week or two or three.  Someone in the community wants to know about all the content in a certain category (a question about red wines or a comment about a specific make and model of something-or-other).  Once that context is set, the community platform displays all the related content.  But in what order?  Well, if you assume that the valuation mechanism has been working properly behind the scenes, then just sort by score — descending.  Might sound too good to be true, but the fact is that a single numeric indicator of value is all you need.  If you trust how the engine arrived at the number, then all you need to know is…  the larger the number, the better.  Simple.

This is what we call “content item score”, or “score” for short … a single numeric indicator representing the value of a content item, given the process of Intelligent Scoring.  This value “contains” the history of all the observation and interpretation of actions in the community involving this content item.

But we don’t stop there.  Understanding the score of the content item is good.  But if you buy into the vision of community I proposed last week or if you believe that the social network created in online community is important, then you would agree that it’s also important to consider the value of the members of the community — not just what they’ve produced.  In the same way that the Intelligent Scoring engine observes and interprets activity to make value judgments about the content in the community, it makes similar decisions — leading to similar value indicators — about the community’s members.

First, the process defines a value indicator called “credibility”.  Credibility is a single numeric value – like score, only applied to a member not a content item — which represents the value of a member of the community.  Credibility is designed to represent aspects of the member’s character, such as their initiative, their integrity, and their commitment to the community.  Like an item’s score, this value contains the history of all the observation and interpretation of actions in the community involving this member.  In a sense, credibility is a member’s score.

Score and credibility get us pretty far down the path, but one more indicator is needed.  A person’s initiative and character are not the same as their knowledge of a particular topic.  So, in addition to credibility, a proficiency indicator is tracked by the CVE for every member of the community, with regard to each node in the knowledge taxonomy.  This value, also an historic set of observations and interactions, is called “proficiency”.  The more someone knows about a topic, the higher their proficiency indicator for that topic.

Score, credibility and proficiency form the three building blocks for intelligent scoring.  In nerd-speak, they are both the inputs and the outputs of every valuation equation … and will be central to our discussion as we carry it forward.

Technorati tags: valuation, score, content item score, credibility, proficiency

It’s not just what you know…

Community is a tricky thing.  If I’m just looking up some fact in a reference manual (and the manual actually has what I’m looking for), then I guess don’t really need it.  What I really need is a good manual, where “good” is defined as having the information I need, or a good author to generate the right manual in the first place.

Oh, wait!  Got me!  Maybe I do need more than the manual.  Why?  Because the “if” in the “if the manual has what I’m looking for” statement above is a pretty big “if”.  Isn’t that your experience?  It sure is mine.  Even if we lay aside the many other benefits of community and focus exclusively on knowledge management for a moment, there’s still a lot to be said for doing work in groups.  Learning, many times, is more than just a database or reference volume.

In other words, a lot of times I want to know who knows something … as much as, if not more than, just discovering the isolated fact they know.  Because, if they’re worth their salt, then they know far more than the one thing I gleaned from the one conversation I was in.  Instead, wouldn’t you rather be connected to their knowledge in general.  They probably have a lot more to teach you than just one thing, especially if you’ve identified that they’re knowledgable in the same things you’re interested in.

Well then, maybe their IQ and their phone number is all I care about.  That sounds good, right?  Not really.  I also want to know who they are.  If they’re not trustworthy … or not ambitious enough to share what they know … or not committed to the community … or don’t care if I need their help … or … or … or … Then being connected to them just isn’t as valuable to me, no matter what they know.

So, it’s more than what they know that I care about, it’s who they are.  Not just their proficiency, but their credibility as well.  Not just their level of knowledge, but their integrity, their motivation, their trustworthiness … lots of variables that have everything to do with them and their character, and very little to do with the the knowledge or shared purpose of the community.

Take TechLore for instance.  People that are always out there, always contributing, always reading something new about some facet of the consumer electronics world.  If I have a lot of questions about my gadgets or a lot of experiences to share, then I want to rub shoulders with these folks.  Yes, they have to know what they’re talking about.  But if they’ve demonstrated to me honesty and consistency and commitment to the community over time, then they’re the kind of person I want to be connected to — more than just being connected to the knowledge or experience they share in one blog entry, etc. Because life isn’t a reference manual.

Technorati tags: credibility, proficiency, character, techlore