Inheritance versus ECS Pattern Concerns

"Should these data members become the base class that this class inherits from?"  If you've ever heard this phrase, or had any discussion involving multiple inheritance, read on!

If you've read anything about the Entity Component System design pattern (ECS), you might have dismissed it as being only relevant to game design.  But ECS short-circuits the need for inheritance - so it's worth trying to understand as a tool in your belt.

I started writing this with processing architectures in mind, because it helps to visualize how the ECS pattern seems to work.  So I will first enumerate computer processing architectures I have encountered.


From The Main Event Loop to Multi-Threaded OS

If you go back to before operating systems offered multi-tasking, code was written in a loop which waited for events, and every event was then processed.  There were several weaknesses - but we don't care for this article.

Later came multi-tasked operating systems.  They allowed events to be queued up in one (or several) tasks.  Other tasks could process event handling.  Then finally we have multi-threaded operating systems.  As more processors are available, tasks can be assigned to other processors.


But all we really need to understand is the main event loop, as it mimics the processing needed to make ECS work on a per-object basis.


The Entity Component System Pattern

Lets say you've got a bicycle class.  It has a gear data member.  And you later decide it needs to have a location too, so you create a location base class and have the bicycle inherit from location.  Seems reasonable doesn't it?  Unless you aren't always concerned with location of the bicycle.

ECS allows attaching concerns to an object dynamically.  It might help to first look at how ECS is implemented and processed in time.

In terms of time organization, the ECS pattern processes concerns in a loop, just like the old main event loop processed events.  An entity can have multiple class instances attached to it, where each instance represents a concern, and those instances can, but don't have to be, attached at run-time.

For this to work, the entity has to process through a list of all attached instances and run an update method (and maybe others).  Common methods can be called directly using polymorphism, but message sending to custom methods may require reflection, or a more sophisticated approach.

Because this architecture organizes by concerns that entity is responsible for, the overall code organization tends to avoid class inheritance.  That seems like a really big deal, and if you've never had exposure to using ECS, might expose assumptions about how code needs to be organized.

Thus a concern (the location of the bicycle) is not inherited by a bicycle class, even if bicycles have location.  Instead a bicycle will have a location concern possibly attached to it.

There are always trade-offs, and workflow for architectural design might be affected.  If you don't have to worry about inheritance as much, writing code can be more straightforward.  Also it seems easier to learn new code by visually seeing which classes are attached to entities.  But the penalty is in the ECS processing at run-time.

C# Reflection - It's What Allows The Entity Component System Pattern At Runtime.

Here is the C# Language Specification 5.0.

From the introduction:

C# is an object-oriented language, but C# further includes support for component-oriented programming. Contemporary software design increasingly relies on software components in the form of self-contained and self-describing packages of functionality. Key to such components is that they present a programming model with properties, methods, and events; they have attributes that provide declarative information about the component; and they incorporate their own documentation. C# provides language constructs to directly support these concepts, making C# a very natural language in which to create and use software components.

If you had never used components, you might walk away from this paragraph scratching your head "well what is component-oriented programming?"

Sadly, this is all the specification spells out specifically about component-oriented programming.

-------------------------------------------------------------------------------

Enter the Entity Component System design pattern.  Apparently there are multiple flavors, and many discussions and purists concerned with how to implement the ECS pattern, and one of those flavors is implemented in Unity, which allows components to be added to game objects in the editor, and at runtime:


// add a script foo to the gameObject
gameObject.AddComponent ("foo");


// add a collider to gameObject
gameObject.AddComponent("SphereCollider");



Unity must be taking advantage of C#'s Reflection feature to find out which components have been added to a gameObject.  I believe Reflection opened the world of component-oriented programming in C#, which is quite wonderful!  Yet searching the C# Language Specification for "reflection" only lightly discusses the reflection feature, and you might not grasp the significance by reading.

Where is the ECS pattern used?  So far I'm only aware of games.  I'm guessing Galactic Civilizations (ship editor), Fate, Reassembly, probably Skyrim, probably Trackmania, Robocraft, and WazHack among many others.

I say "probably Skyrim" and "probably Trackmania", because some features might still be implemented without ECS, but it's cleaner to add components rather than trying to anticipate every possible feature that may ever need to be added to an object.  Thus a modding feature might be best implemented with ECS, but there are other ways to allow a game to be moddable - they just won't be as future-proof.

For example, if you wanted a collapsible wall which responded to impacts, it could be coded without ECS, but it would be a nightmare.  With ECS, a collapsible wall is trivial to implement, as each section of the wall can independently respond to impacts and distribute damage to adjoining sections.

There is also some discussion that Unity "breaks" the ECS pattern by allowing code to reside in components.  Some people want the components to be data only with no methods.  My impression is they have overlooked other messaging features and Unity does it fine; it's a rabbit hole I'm not interested in digging into right now.  The way Unity implements ECS is to divide concerns into individual components.

The ECS pattern in Unity organizes the software in such a way that components can be attached to initial objects without much emphasis on inheritance.  I've written applications without ever deriving from anything other than monobehavior.

Also, if you want to understand Reflection by itself, watch C# Reflection to the Max, and thanks Jamie King for helping me grasp the concept!