Archive for February, 2016

Virtual Assistants as Parallel Human Processors

Posted on February 17, 2016. Filed under: Uncategorized |

I’ve been thinking quite a bit about virtual assistants lately…and the other day i had a notion that the rise of virtual assistants feels in part a response to the growing demands on human beings around information processing. When you are always connected, and there is an ever growing amount of data to be processed, but a limited capacity to do so, physically, something has got to give. In computing, when a server can no longer efficiently process the requested volume of data, what do we do? we spin up another machine and split the work. Of course there are optimizations on that single server, to make it more efficient in handling increased loads, but at some point, it makes more sense to run processing in parallel. I think there is something similar happening in human information processing. The discovery tools have reached some limit of optimization (i.e. the feed, search, etc), and even the size of the informational units has reached a plaeau in processing optimzation (tweets vs blogposts, headlines vs articles, images vs text, things can’t get much smaller/denser, etc)…and there just really isn’t much more room along those axis to help us handle the ever growing body of information that needs to be processed…in the last 10 years we were able to lessen the load by spreading our processing over much more time than was previously available…mobile gave us much more connected time to distribute the information processing over, but now that we are on our phones all of our waking day, we’re running out of processing time to add…hence a general sentiment of “information overload” and “phone addiction” that is becoming mainstream affliction in today’s day and age…so what do we do, when we physically can’t process any more, and we don’t have more time to add, and we’ve done all the mechanical optimizations we can to make the time we’ve got more efficient? We start to think about parallel processing…about spinning up a second machine, or in this case, a second human, to split the load…granted many of these virtual assistants are focussed on task oriented information processing, but in general…we are adding parallel humans to interact with the same mechanical tools we ourselves have access to, in cases where we can no longer eek out gains in a “single server” architecture (i.e. doing it ourselves)…i think this is a trend that will continue, and i’m interested in seeing how higher order information processing, that’s focussed less on completing tasks, and more on augmenting your knowledge and informational awareness via a second human brain, will come into focus over the next 24 months…I think people will increasingly value being able to parallelize their processing….and it will become more common for people to offload even non-intuitive processing tasks if their interface and ux to do so becomes more frictionless. I can, for instance, see a world where people ask a 3rd party to read and monitor their facebook for them…as odd as that sounds…I use that just an example of the broader theme of parallel human processing as an architecture of living in a world of growing available information with limited physical and mechanical processing tools…