This is the first of a series of articles where we will try to address current software design problems in a minimalist way.
In university and industry we are taught the following definition about software:
Software is a set of instructions that with certain input data generates output data.
If we look at Wikipedia, we will understand that this is precisely the definition of computer program:
A computer program is a collection of instructions that can be executed by a computer to perform a specific task.
But many decades ago we understood that software is much more than a
program and we still don’t have good definitions of what software really is.
Again according to Wikipedia:
Computer software, or simply software, is a collection of data or computer instructions that tell the computer how to work. This is in contrast to physical hardware, from which the system is built and actually performs the work.
In other words, the definition is by the complement (everything that is
not hardware) or by physical omission (everything that cannot be kicked
and can only be cursed).
These definitions take us away from the only purpose we have whenever we build software that is:
Mimic something that happens in a possible reality.
When we do this, we are forgetting about the origins of modern programming languages like Simula. (the first object-oriented programming language and the first to include classification).
In that language, it was clear from the name that the goal of software construction was precisely to build a simulator. This is the case of most of the computer software applications today.
Therefore and returning to the origins we can define the software as the construction of a:
(M)odel: (A)bstract (P)artial and (P)rogrammable (E)xplainig (R)eality.
The model is the outcome of observing a certain aspect of reality under a
certain lens and a particular perspective with a certain paradigm.
It is not the truth revealed.
It is the best possible conception at a certain moment with the present knowledge.
Since Plato’s time, human beings have tried to build good models of reality.
Photo by
Max LaRochelle
on
Unsplash
We can only understand it and observe it with instruments like a black box based on its behavior before our stimuli. Like Schrödinger's cat, we cannot know what state it is in without disturbing it with our measuring instruments.
The use cases technique is an excellent tool to describe a model in a declarative way.
If you are not a declarative programmer, you might lose your job very soon.
Because it has to run in some type of simulator that reproduces the desired conditions. It can be a Turing model (current commercial computers), a quantum computer or any type of simulator that can accompany the evolution of that model.
Like any scientific model, we will make a partial cut of the reality that interests us for the problem we are modeling, taking certain simplifications on aspects that do not interest us in order to isolate the problem.
A 1: 1 scale map of the earth would not make sense.
The model must be declarative enough to let us observe its evolution, allowing us to reason and predict behaviors in the reality we are modeling.
It should reproduce conditions that occur in an observable environment.
The ultimate goal is to forecast the real world, like any simulation
Once the software is defined, we can begin to infer good modeling and design practices that we will publish and add to this article.
From the axiomatic definition presented in this article, we will derive principles, heuristics and rules to build excellent software models.
Part of the objective of this series of articles is to raise debate and discussion spaces on the problem of software design.
We look forward to the comments and suggestions on this article.
Previously published at https://medium.com/@mcsee/what-is-software-9a78c1172cf9#9065-c