What is an Application?

Applications are slippery things. Everyone thinks they know what they are, but rarely will you get absolute agreement.

Most people tend to think in terms of a batch of related software, usually with a smallish set of user interfaces and perhaps and API or two. There are many business users who might not be able to put it in those terms, but if you pushed it that’s what they would mean. This is true but not quite the point.

Technical people would add things like data sources, processing layers, algorithms, storage and so on. All true but not quite the point.

For me, we need to go to the original meaning of the word. An application is where technology is applied to a business requirement or problem to solve, accelerate, simplify or in some other way improve life for the business. It’s not a thing in itself. It’s the ‘application’ of a bunch of other things.

Applications are the meeting point where the computing power provided by Infrastructure, desktop and mobile hardware meets a business problem and addresses it with some fancy math. Sitting in the middle is the user community, the business unit(s) that derive real value from it all. And they present an unknown variable because they are human.

Application Management teams spend a lot of their time dealing with people who perceive an ‘error’ which is in fact a misunderstanding of what the software is supposed to do. As a result, a fair amount of time is spent in conversations about what we wish the software did, not what it does.

Often, it’s not strictly a misunderstanding. I have lost count of the number of times I have been told “This report isn’t right any more”. Well hang on, the report hasn’t changed. If it’s not right ‘any more’ (i.e. it was right previously) there are a few possible explanations, and none of them are about the report:

1.     Your colleagues (it’s not you, Mr CEO, obviously!) are putting the wrong data into the application

2.     The application is in some way manipulating that data incorrectly, perhaps because of variables like exchange rates not being kept up to date. (The eagle eyed might notice this is a more detailed variant of option 1).

3.     Your requirements have changed, and you are now asking a question that the report was not designed to answer.

In my experience, option 3 is by far the most likely.

I include all of this to demonstrate that the applications, more than data centres, networks or even personal equipment, are the source of emotional user-driven requirements that are very hard to measure empirically. Applications are the representation in data of the activity of the business, and as such they are the source of much emotional attachment, frustration and, when you get it right, joy.

Nick Ellis