Red Books...

I've recently signed up again for a basic subscription to the online 'magazine' "Medium", whose tech feeds in particular I follow when I can. An article that caught my eye this week is by a contributor going by the handle TheOpinionatedDev: "Object-Oriented Programming is Dead - Here's What Killed it". Now, if you don't know what Object-Oriented Programming [OOP] is, I'm not going to try and explain it, as that's not the point of my mentioning it here: a quick WikiGoogle will fill you in on the basic concepts, and give you probably too much detail to absorb at one sitting if you don't have any knowledge of coding. The point is here that OOP has pretty much been the dominant architectural idea behind modern software coding for a good thirty years or so, and was the de facto dominant religion underlying all software development during that time [I use the term religion, because as in religion, dogma rules]. The key point is made in the first paragraph of the article, which exemplifies the simplifications and assumptions about the world that dogma always brings to the table: simple solutions to complex problems, always a dangerous mindset to get into.

OOP came out of an era when software was required to perform essentially monolithic tasks on known data sets: even when networked systems came to the ascendant, in the early days, the tasks assigned to software in dealing with data were pretty clearly defined and had manageable bounds within which to work. It seemed logical to move on from older, procedural programming models: "IF this, Do that, otherwise GO there": the straight line methods of the early days of programming, and to replace this methodology with something more compact in its thinking; breaking down tasks into broad classes of activity that could be re-used from data set to data set as needed. In the Medium article, that key paragraph uses the phrase: 'Software would be clean, modular, and "just like the real world."' Which sums up the problem in antithesis: the world [and people, and the data they generate] is anything but clean and modular: it's extremely messy and frankly fuzzy. Hence the 'death' of OOP in favour of the much more loosely structured approaches to code that are taking ascendancy currently.

My point is that computer programming - in its short existence, has demonstrated that orthodoxies and dogma are simply chapters in the history of whatever they emerge from and control: that there is a natural tendency for ideas and thinking to evolve over time as better ideas and thinking evolve to match their context. It was ever thus. In programming and software, the timeframe is measured by the decade; politics by the century, and religion by the millennium. What's needed now is for as radical a paradigm shift to happen in both politics [and by extension economics] and religion as happens regularly in the field of computing. This is the twenty-first century, and it's about time we caught up with ourselves in the most fundamental areas of our mutual association. There are no fixed or given points of focus: all shifts with us and before us: like quantum superpositions, everything changes depending on the viewpoint of the observer. I suppose, just don't let the grass grow under your feet, mes amis...