Charles Arthur flagged an interesting article by Richard Chirgwin on The Register about the weird worldview that the Google Nest Internet of Things house has about it’s occupants. Put simply, it really struggles to understand that there might be more than one person living in a house.

If you use online services and you are a member of a family you’ll possibly know this sensation already. Take, for example, Spotify. We share an account across the family, but it results in very weird suggestions coming up because my tastes (broad, eclectic), my wife’s (more poppy) and my kids (the soundtrack album to Despicable Me 2, again and again and again) lead to recommendations that are meaningless.

Similarly, Google’s Photo service which provides some remarkable functionality, not least text searching against your photo library using image recognition, singularly fails to be able to support the idea of a family album. I can share individual or groups of photos with my wife, but there is no way to allow us both to have equal access to everything. That fails to understand the small-group nature of family living.

As I ranted towards the end of last year, I reckon much of this is down to the individualistic nature of modern software design and development methods. And from a sociological point of view there is a big challenge ahead here.

Understanding the needs and actions and responses of an individual is difficult, but with enough data it’s predictable enough for services like Google Now to provide useful features based on big data number crunching. At the other end of the scale, mass groups are relatively straightforward to model and predict – that’s the world of market research.

But where it gets really tricky is with small groups of people with close relationships. How Mum, Dad and the kids interact on an ongoing basis. Or how a team in a work environment interact. These groups are small enough to behave like very, very complex individual units, but not big enough to aggregate into lumps where statistics come to the fore. And it’s this complexity of small groups that could be the thing that holds, particularly Internet of Things-type home automation, back for quite a bit longer than its acolytes might today believe.

One thought on “Groups think

  1. The sociological insight regarding a disconnect between “individualistic” design/development methods and the requirements of families and groups is very interesting. I recently tweeted another sociological factor in our social apps: Facebook’s automatically generated “Year in Review” function sounds great to a 20-something year old developer, but when it creates a giddy review saying that “2014 was a great year,” and its cover picture is that of a loved one who died (as happened to me), you can tell that people over 40 years old didn’t have much of a role in the developing that feature. As Facebook’s users are skewing older, this is an important point. Coincidently, the Washington Post picked up on a similar story, but it didn’t make the sociological connection. Products targeting a broad base of users benefit from having a broad-based development/management team.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.