In Search of Zora

Thanks for the Memories

While nobody knows quite how human memory works on a biological level, it's clear that our memory function is responsible for much more than just our ability to recognize people, places, and things.  For example:

1. Lists.  Of course our memory allows us to keep track of multiple things, but then, so does written language.  This basic type of memory function makes computers invaluable, but it's not unique to us.
2. Associations.  For most people a single word calls to mind dozens of specific memories (directly and indirectly related) that include specific places, people, events, sequences of events, etc.  A single word (in conjunction with its associations) can be all that's needed to get someone chatting for hours on end.
3. Prediction.  Without access to our uniquely adapted memory structure, we wouldn't be able to predict events beyond the present moment.  But it turns out, this ability to predict has intense consequences -- in a good way...

In fact, much like dreams, the act of prediction is really the combination of past memories with alterations to their original order or even their content. 

Enter the power of human imagination.  

Clearly, prediction allows us to plan the future.  But most importantly to my work, the ability to predict brings with it the power to attribute meaning to events as they change (or are changed) over time.  Again, I find myself amazed at the synergy between these observations and Clio's music analysis algorithms.  

Effective musical understanding begins and ends with the ability to remember and predict.

You Know What I Meant

Just ran across this elegant description of Google's context-awareness from Steven Levy:

"A rock is a rock. It's also a stone, and it could be a boulder. Spell it 'rokc' and it's still a rock. But put 'little' in front of it and it's the capital of Arkansas. Which is not an ark. Unless Noah is around."

Turns out, this a decent encapsulation of how Clio's algorithms operate during analysis. But abstract languages (like music) are simultaneously constructivist and self-referential in nature.  

With which does the bulk of identity and meaning lie?

And how does this play out on a large scale?  

Only time (and extensive experimentation with Clio) will tell...