Task 9: Network Reflections

This was a neat exercise and I found a couple interesting tidbits of data that caught my eye.

One of the first was how much of a “cloud” it created – meaning that there were no selections that were left out completely.  Each track had at least one person choose it.  In fact, 26 of the 27 tracks had more than one person choose it.  There were no tracks from the original record that created consensus about them *not* being on the record.  The closest one to this would be the ‘Men’s House Song’ that was only chosen by one of the curators (as you can see, off on its own on the left):

What is interesting about this, is that (from this data), we cannot tell WHY 26 of us chose to leave this track out.  Could it be because it’s from New Guinea and we didn’t feel that culture was pervasive enough to be included on a ‘worldwide’ record?  Could it be that we felt men’s voices were already very predominantly represented on the record and it was an obvious way to limit more of that because “mens” was in the title?  It’s impossible to know from the data we are given.

Another interesting combination I noticed comes from looking at Group 7….which ended up being a group of only two.  What was interesting about this group is that they chose 7 of the same songs for inclusion out of the 27 – the only differed on 3….closer than any other group.  When I saw this, I did some very basic research (ie. looked at their profile pages on the People page of Canvas….thought about trying FB, but didn’t want to get too creepy!) and made some very rudimentary judging (don’t be mad!) that the two curators may come from a similar demographic background in more than one way.  I wondered if these similar aspects would have contributed to picking such similar choices.   Here is their graph:

Unfortunately, this theory didn’t hold up to well, because another group I found very interesting was Group 4.  This group included three people and their choices were almost as congruous as Group 7.  Interestingly, they matched up on 6.6666 of their choices with each other – very high for a group of three, I thought.  So I did the same ‘research’ on their profile pages and found that my theory did not really hold up because their demographics did not appear to be as similar at first glance.  Their graph is also shown here:

One other thing of interest with this group was that I noticed that the picks they chose were some of the most ‘western’ tracks on the golden record – so I would infer that these are tracks that they related to on a more personal level based on past experience.

What I’m learning from this experience is that it is VERY difficult to capture the reasoning behind choices with this kind of data gathering.  I think in order to even come close to being able to make any statements about the ‘null’ choices that someone makes, there would need to be a LOT of other data collected about that person and their other previous choices.  But even doing that would not bring you to a 100% definitive answer.

This exercise is really showing me just how the information that large tech companies (ie. Google, Facebook, Microsoft, etc) is being used to create snapshots of who we are as people.  And it’s REALLY helping me to see WHY they are so driven to collect as much data as they possibly can –> because that is the only way they can start to make real inferences about who we are and what our choices/preferences would be in and situation.