Why are some things easy to develop?

A question we received from a client recently gave me pause to reflect on why getting software to do some things was sometimes a lot harder than people think and sometimes a lot easier.

The questions was if we'd be able to get cadastral data (roads, places, boundaries) onto a map underneath other industry specific data. The answer is, yes, we just stick a Google (or Bing or OpenStreetmaps) layer in there. But if Google Maps didn't exist, or didn't support being used in this way we'd need to build that ourselves. In a very different world we might have to go and get paper based records and digitise them, then build our own mapping solution. So what is essentially a negligible coding effort and almost zero cost option would otherwise be a multi million dollar effort.

Why is it easy? Because someone has already built it, and is offering it to a huge base of customers. This massive economy of scale spreads the costs over millions of applications. This is true with plenty of other development tasks - there is often an existing software library or service to facilitate it. Even programming languages themselves are something we rely on to make what would otherwise be the difficult task of converting ideas and processes into to zeros and ones easy.

So when is it hard? If it doesn't already exist it's probably hard to do. When is it expensive? If it does exist but can't be sold to a lot of users to spread the cost. We've previously found a commercial software library that would help us acheive a task, but being a niche application the cost made it unviable. The library was far less costly than trying to build the same functionality from scratch, but still out of the project budget.

Nearly all software development stands on the shoulders of giants who have lifted us from binary code to a level that is much easier to work with, whilst providing the wheels we don't need to re-invent.