Email vs. Social Streams? Is that the right question to be asking?

Since the birth of social, or at least the proliferation of microblogging, there has been the ongoing arguments about e-mail vs social streams (some listed below). On the one hand you have the social media camp predicting the death of e-mail. On the other hand you have the e-mail camp stubbornly refusing to die :-) and arguing against the noisy social stream. Personally, I believe they are both correct in their views. [ way to sit on the fence Marie ]

I guess I see the problem slightly differently and don’t really think of it as E-mail vs. Social Streams. From a pure convenience perspective I would rather have a single communication tool that supports mobile and desktop where I can flexibly choose:

• Mode: Voice, Video, or Text.
• Latency: Instant (SMS or call) OR queued (e-mail or voice mail).
• Intimacy: 1:1 (individual), 1:N (group), or 1:? (public).

This integrated multi-channel approach is being widely adopted by folks like Google and I believe it can be both powerful and convenient. For example; it easily allows me to shift communication mid-stream. I start texting with a friend (is it a good time to talk?), I then move into video mode (what do you think of this dress?), then add more people to the video chat (who thinks I should buy the dress?), and finally go public (where is the cheapest place to buy this dress?).

However a single communication tool is only part of the solution. I most definitely don’t believe that just moving e-mail to the social stream is the right thing to do, or vice-versa. The folks that bemoan the level of noise in the social stream and its lack of actionability, are absolutely correct. As are the folks that bemoan the mountain of e-mail most of us are buried under. Both communication channels are noise generators so what I believe we should really be focusing on is How can I reduce the volume of communication that I send or receive?.

In 2005 as part of my involvement in the EU FP6 project Nepomuk, I led development of a prototype solution called Smart Assistant which was essentially a kind of bot that sat on your machine, analyzed what was going on and created a personal knowledge graph. Its purpose was to not only help to organize and prioritize your desktop but, more importantly, to action things. Over the last decade computers have “supposedly” gotten more intelligent, but as a knowledge worker I feel myself asking the question What have you done for me lately?.

Its definitely increased information overload, thank you very much. However when was the last time your computer actually answered an e-mail question on your behalf? Even though you’ve answered the same question a dozen times. When did it last proactively collect and prepare information you needed for a client meeting? Even though all the information it needs is in your calendar, e-mail, and social stream. Or alert you if something related happened that you need to know about before the meeting? Now thankfully there is light at the end of the tunnel and we are seeing these new types of analyses coming into the market, but more about that later.

Lets come back to our “reducing the volume of communication” question. I would estimate that my Inbox is divided between three categories of “stuff” all of which could be significantly reduced by some decent (not rocket science) analytics.

• Questions (40%)
  Content analytics allows you to capture Q&A exchanges and build a knowledge base so that when similar questions arrived in the future your computer should be able to proactively respond, at least in the first instance. What we did in 2005 was very simple, but just imagine who cool this could be with a Watson-esque system.
• FYIs (30%)
  A combination of content, semantic, and network analysis helps to organize and put into context these FYIs. So if you are working on a customer engagement, the FYIs related to that may be something you want to show up in your contextual social stream.
• TODOs (30%)
  Similarly to the FYIs, a variety of analysis helps to organize, prioritize, put into context, and even action (or at least prepare for action).

Google’s Social Card feature on Android is going in this direction and is a great example of how powerful some simple text analytics can be. I was recently in London for some meetings. I was on my way to my afternoon meetings and decided to stop for lunch on the way. I was sitting in the restaurant when a Google Card alert popped up. It had noticed from my calendar that I had a meeting. The invitation included the address. Google Maps knew where I was and because it was lunchtime and I wasn’t moving it guessed I was having lunch. So it was able to pop up and let me know that I needed to finish lunch in 10 minutes if I wanted to make my meeting. It had calculated the route I should take so all I had to do was pay my bill and get moving. Cool or what? Although I appreciate some folks may find this as bit invasive.

A snapshot of some interesting articles about E-mail and Social Streams (in chronological order):