With apologies to Thomas Paine’s, “Lead, follow, or get out of the way,” my headline for this post is a recitation of how most people filter the information and communication that comes to them online.

From a marketer’s perspective, as well as an information consumer’s, this is darned limiting in the oddest kinds of ways.

We can search for information we desire.  In this case, our keywords drive not only the search, but for marketers, they pre-screen the searcher to ensure there is reason to believe they may have interest.  After all, they’ve told us what they are searching for.  This curious intersection of declaration of desire and the use of a separate search engine to find the objects of desire is what has made Google such an incredibly valuable franchise.

The Social world has brought us the idea of “following” or “friending”.  We can identify some stream of information and subscribe to it.  This is another way of declaring desire.  Whether you follow by subscribing to an RSS feed, following on Twitter, or friending on Facebook, it’s all the same in that sense.  This is why marketers would buy ads on Twitter or Facebook.

“Getting it out of my way,” represents the last commonly used means of filtering information and communications.  By that, I mean the lowly spam filter that guards our email inbox.  People deride email as a time waster and Old School, but it serves a useful purpose because anyone can send an email to you.  If you are not actively searching (or actively finding is more like it), and you have not followed the right sources, how can you be reached?  It is valuable to be reachable, and this is why I wonder about important people who decide just not to use email.  Really?  Too much noise?  How do you ever take advantage of serendipity with no email and an unlisted phone number?

In some sense, this post is mostly about filtering the information sources that are used for serendipitous connections, e-mail being the best example.  How can I let anyone contact me via a channel, yet manage the channel so I mostly see what I am actually interested in seeing?  Sticking strictly to search, follow, or get out of my way is so limiting that I can’t help but wonder when we’ll be able to do better. 

What if I want to read emails sent from anybody who is a contact for anyone who is one of my contacts?  A friend of a friend is often someone you’d be happy to hear from, especially if your friend took the time to enter them into their own list of contacts.  This is the sort of thing that is pretty easy for mail providers to offer, but they don’t seem to think of it.  Maybe I am willing to go three levels out–a friend of a friend of a friend.  It’s rare indeed that I hear from someone interesting who is further removed than that.  Come to that, if I have all this friend info and I am searching, maybe I want to be able to re-rank my search results not by Page Rank, but by Friend Rank.  Which search results did my friends actually click-through to?  Maybe those should rank higher or be called out to me in some way so I know.  While a marketer wants to be able to tell me, “Your pal so and so bought that so you should too,” that’s way too invasive and privacy problem prone.  If all I know is, “Someone on your contacts list made that choice,” that might be more manageable, and something people would deal with on a privacy basis.

We’re going to have to do  more in this “affinity filtering” area.  Retailers like Amazon.com are already pretty good at it.  “People who bought this also bought that.”  But wouldn’t it be interesting if you knew more about those “People?”  Does it matter more to you to know that “Your friends who bought this also bought that” than just arbitrary “People?”

Systems will also need more memory despite the fact that statefullness is verboten in architectures.  The last time I searched for “trepanner”, I was interested in the metal-cutting tool, not the practice of boring holes in people’s heads.  That probably tells you that when I search for “reamer” I also want the metal-cutting tool, and not the surgical instruments or some other “reamer.”  If need be, you can disambiguate the way Wikipedia does.  It’s very helpful.

Getting back to our email, if my friends accepted that email without marking it as spam, I probably should too.  We do have affinity spam marking, and that’s good.  But, I want that contact-based capability.  If you are that busy VC, wouldn’t it matter to you if the person sending the mail was on the contact list of one of the other partners in your firm, but just not on yours yet?  Wouldn’t it also matter if they were on the contact list of some other VC at some other firm? 

Put all this next level filtering together, make it easy for me to apply it to my email, search results, and any RSS feed, and you will have made a tremendous difference in reducing information overload.