doug wick

When I heard of and experienced Twitter at South by Southwest Interactive 2007, I groaned. Wow, another way for people to waste time online, I thought.

Twitter, for the uninitiated, is an online messaging service that allows you to broadcast short, 140 character updates (termed “tweets”) to a waiting audience of other users who are also posting updates. The updates are meant to be “what you are doing” at that moment, but really contain all types of short-form content that are as diverse as the users who write them. On Twitter, though, you select the people whose updates you follow, building an aggregated “twitterstream” of the people who matter to you. Similarly people may or may not follow your updates.

I finally started using Twitter a couple of months ago when I saw that it was not, in fact, merely a diversion for internet geeks (in my defense, there are many out there). It has staying power. What I noticed, once I started to use it, is that hidden in Twitter’s simplicity lies a game-changing adjustment in social communications that could end up reinventing the way we do a lot of things in our personal lives and in business. That’s quite a bold statement, so bear with me.

Back in my days as a software developer I worked for a company who did and still does publish a suite of sophisticated tools for managing the business flow of a commodities trading business. Along this business flow we had organized the software into many parts that matched the real-world process. The contracts subsystem would allow users to enter buy/sell agreements, the distribution subsystem then allowed them to match those agreements to the required physical distribution, and so forth into invoicing and accounting.

One of the big issues we contended with early on was the communication between the subsystems. The activities of one always affected another, so when something changed in one domain the others had to be alerted in real time. This was a painful process to code. Which other subsystems needed to be alerted of what, and when? It was even more painful to maintain as the system grew and evolved new capabilities.

In response to this issue, our team rewrote the messaging architecture to what in programming circles is referred to as a publish:subscribe model. This model created a virtual broadcast system where every subsystem would publish its activities (um, “what it was doing”) to the entire system and the other subsystems could choose what messages to pay attention to and which ones to ignore.

This approach was far superior, for many reasons. Here are my top three:

Transparency is rewarded.

First, the higher number of status messages developers published to the broadcast system, the more they were rewarded by not having to worry about what other systems might need. This allowed for a high degree of transparency within the system about what every part was up to at any period of time. This became very helpful for troubleshooting and performance tuning (oversight) of the entire software package.

Autonomy = efficiency

Not having to worry about communicating to other specific subsystems allowed developers to have more autonomy and focus on the more important work of making their specific piece of the system work better and faster. It saved time, and money.

Spontaneous innovation happens.

Availability of more activity information than you think you need leads to creative thinking about how you might actually be able to use the extra stuff. Early into the publish:subscribe model we found developers using information in unexpected ways to make their particular domain more powerful and useful.

So you might be able to see where I’m going with this. Twitter is the publish:subscribe model applied to personal instant messaging. As a result, it is superior to the other existing tools (basically any other IM client), which are based on the de facto one-to-one messaging model – for the reasons I mentioned above. In the Twitter world each person is like a miniature subsystem, broadcasting information it thinks the world might be interested in. And others listen, selectively.

The important thing in understanding Twitter’s importance is to focus on certain contexts for Twitter, not on the platform itself. Without context the tool does seem frivolous. Unless you are a researcher you wouldn’t care about listening to the global twitterstream. It would be an overwhelming torrent of which you might find a miniscule percentage relevant to you.

It’s also important to note that the default approach – to subscribe to your friends – is actually one of the less useful applications of Twitter. I love my friends, but many of them just aren’t doing things minute-to-minute that are of any relevance to me. Personal details for the purpose of friendship don’t usually require instantaneous communication (unless you like to stalk your friends), which is why social networking websites and one-to-one IM work just fine for those purposes – not to mention old-fashioned and still far superior technologies like using a phone or (gasp) actually spending time together. Plus, real friendship is forged and bolstered by the slow-moving plate tectonics of our lives, over the long term – not by the daily humdrum of changing moods and insignificant occurrences.

The best application I see is actually at the workplace, or really any club or organization that has to get things done. In the same way that my old company’s system had parts, so organizations have departments, and departments have people. People at work benefit greatly from timely transparency with each other. The more autonomy every department/person can have, the more resources it/he/she can focus on the appointed task. And the creativity that leads to broader capabilities and innovation is fueled by an understanding of what others are thinking and doing.

So Twitter-like technologies might allow workplaces to function more efficiently and creatively, but why am I making the wild claim that it will reinvent the way we do things? Well, if you had an internal Twitter and a profile page to post longer documents, photos, and files at the office, would you even need email? How often does a lack of transparency, either intended or unintended, hinder your group accomplishment? Think about it.

Would you need a boss? Would your boss really need a boss? Would you ever need to have a meeting? Or would, through Twitter, the organization be able to function more democratically, instantaneously solving problems? Given a couple of “elected” leaders and judges, could organizations become almost completely flat and dispense with the immense overhead of traditional command-and-control management? Consider it.

What if you injected the twittering of your customers into the mental collective? Could your whole organization provide customer service? Could getting your product or service to market become extremely agile, taking input from your body of customers in near real-time?

How we do business in the future (Enterprise 2.0?) could be radically changed, in a way that makes us all more productive, and more satisfied, all because of a little website cutely named Twitter.

UPDATE: Strangely, Max Kalehoff published a good post with exactly the same title as mine, on the same day. Not implying any shenanigans, our posts are very different. Just saying it must mean that a lot of people are figuring out that Twitter “matters.”