Do you remember your mother telling you to look both ways before you cross the street? This was a right of passage that most of us learned before turning five and heading off to elementary school. Years later I worked part-time in London near John Street. The office was a dozen blocks away, and there were as many streets to cross. The thoughtful Brit's had stenciled on the streets to "look right" or "look left" on some one-way streets. This was presumably for those of us who learned on the American system of roads, where we drive on the "wrong side of the road", as my British colleagues were apt to point out.
If you looked both ways, you were safe. But over time, people have a tendency to simplify things to the usual, so our experience on the other side of the pond was to only look left, which may get you struck by a fast taxi in London. Hence the reminders.
There are two lessons here for our system and process designs. The first is that context matters. It matters where the audience for your system reside and how they look at systems and expect them to work. For example, a processing-heavy smartphone app may meet expectations in the US or the UK, but fail miserably in Sub-Saharan Africa. Why? In most African countries, once you are out of major cities, the connectivity is often a sometimes-connected 2G reality. Even in the cities, 3G connections may be the most you get. Also, context covers much more than geography and networks. Culture and language play a role as does the economic reality; what types of devices (and data plans) are affordable? A bit of self-examination is called for: what are we assuming about our users?
The second lesson is that as designers, we need to go back to our childhood and look right and look left before proceeding. Are we looking at our designs from both angles? In disaster preparedness, whether for the data center, our community or country, we develop A-plans and B-plans. Why? If our A-plans fail, our B-plans kick in. Having worked with IT systems for many years, I can tell you that it's not a question if your system will fail; it's when will it fail. And then a key question for your design is, will it fail gracefully? A and B-plans are like looking both ways and anticipating from where the crass may occur. |