OOP exists to add a quadrillion of layers to otherwise perfectly understandable business code in an attempt to obfuscate meaning and intent and guard against malicious extraction of valuable trade secrets.
Oh, it also helps me to pay my bills, because someone has to untangle the mess.
And it helps my therapist, because he has to keep the madness that grows inside of me in check.
Jokes aside: I was taught OOP at university.
I was also taught functional programming, answer set programming and other forms such as logical programming.
Focus was definitely on OOP though.
I mainly see OOP as a way to (dis) organizer code and a way for hardware manufacturers to sell more hardware.
OOP is sold on the promise that everything is modular. Since everything is neatly encapsulated with clearly defined API surfaces, you can easily swap out any part for any other. This should make developers interchangeable, because anyone can go in to the code and swap out parts as needed.
In reality this seems to make coupling even worse. Now you have to dig 8 classes deep just to find the business logic for a particular feature. Changing it is even harder because the rest of the program expects this part to behave in a very specific way.
Rigid class-based inheritance (especially multi-inheritance) is what sucks. This over-engineered complexity really only benefits things like window toolkits where you want super rigid modeling of widget types. But even that's a stretch.
Traits do OOP the right way. Complete flexibility.
Dog goes "woof"
Cat goes "meow"
Bird goes "tweet"
And mouse goes "squeak"
Yes. And structs are just bytes. Levels of abstractions don't do anything the underlying levels don't already do, they just give you means to express your intent in a clearer form.
> If you think you already know OOP, this article will change the way you think about programming
No, it won't. I'm especially sad that this, like so many OOP guides before it, did not go into how it interacts with data structures and bigger picture stuff. It had the perfect chance to do that as it could have made a Catalog class and then had a discussion about what belongs to that and what to the books. Instead it started to abstract on author, in the stupidest way possible (no, no one will ever look up a book using the author birth date).
OOP, like most stuff discussed in dev-related web, exists so that permaonline people can have another reason to create their own tribes and fight each other.
OOP is fine. Not using OOP is fine too if your architecture / design demands it.
What people forget -- much like the design patterns in the gang of four book -- is that languages and frameworks evolve.
A decorator pattern was a niche but useful abstraction in the 1990s. In Python today you can @decorate stuff just like that. It's evolution.
The same holds for OOP. Encapsulation and co-located methods with the encapsulating slots was an incredibly powerful upgrade over basic structs. Now most languages have first-class functions and lexical scoping so you can build your own encapsulation that way.
Some problems don't require OOP, but most large complex problems do, IMO.
Most complex codebases I've seen which were pure Functional Programming were spaghetti code; unmaintainable.
What I saw every single time was that the project was syncing a huge amount of state in a central place and then passing it through a large number of components and sub-components. The top level component basically had to have full awareness of everything going on inside the system in order to do its job and there were no separation of responsibilities because none of the components had sovereignty over the state they needed to do their job independently. It's just components micro-managing components all the way down.
React with Redux is probably the best FP implementation I've seen thus far but even it is kind of a mess. I have nightmares about Redux Saga. The Redux Saga logo is literally a stylized drawing of entangled spaghetti.
Had Redux + Saga been presented to people as part of React at the time it was introduced, React would not have become popular. Because people would have understood that it creates too much unnecessary baggage which isn't worth it.
> none of the components had sovereignty over the state they needed to do their job independently. It's just components micro-managing components all the way down.
Isn't one of the tenants of FP that nothing is mutable anywhere?
OOP hate is definitely a popular way to signal that you aren't just a "common grunt programmer". And there are personalities who have made it a big part of their appeal.
But I also spent some time with modern Spring Boot, and I get where the hate is coming from. There is a lot of complexity that just doesn't seem necessary at all.
> But I also spent some time with modern Spring Boot, and I get where the hate is coming from. There is a lot of complexity that just doesn't seem necessary at all.
I'm sure there is some (very high) level of inherent project complexity where the massive number of abstractions and tools that Spring Boot gives you (and forces on you) actually starts turning into a net positive... And I'm also fairly sure that most projects don't quite break even, and would be better off with something more lightweight.
the functionality is often necessary and useful the question is if they're required to be implemented in the way that they are or if there is a simpler approach.
also a lot of oop hate is 1) hatred of compile-time hierarchies attempting to match the domain model and 2) all the complexity in oop systems which isn't really inherent to OOP but made possible by it. oop doesn't require a PrototypeFactoryInjectorFactoryBean but it doesn't prevent you either :)
on the flip side I've seen functional codebases which are also full of complexity because of how fragmented everything is, like 20 functions all applied when it could have been a single imperative one making it hard to keep track of what is happening.
I think the hate typically comes from the mis-application of various tools; overly complex solutions for simple problems. There's nothing wrong with patterns per say, or frameworks that have complex building blocks for every conceivable permutation of relatively common issues. However, when people reach for unnecessarily complex tools out of laziness/habit, some hate is justified. I suspect this is very similar to what you're saying, and of course, it's not OOP specific. Functional languages have their patterns too, but instead of a famous book enumerating them, we have a famous blog post stating that functional languages are so awesome that they don't need patterns (or at least, this is how most people seem to interpret it).
I often have to teach basic OOP concepts to junior - mid level developers.
Many who've entered the career in the last decade or so seem to have missed what I 'd consider to be quite a fundamental grounding.
Not even advanced concepts from the gang of four book - though I have taught these too - but often just an un-awareness of OOP in general, even simple things like understanding encapsulation.
Frameworks and languages do evolve, but never being taught basic things like encapsulating state or exposing dependencies seems an issue. Very differnt from my own time as a junior when this stuff was constantly hammered home.
I used to be OOP hater, but realized that it has it's place and some problems are OOP shaped and OOP is a wide range of techniques. I do think that python is a sloppy language and generally don't like it.
Functional programming has it's place and I love it and is probably what my mind goes to the most, but often in Functional languages I find you can 'over fit' to a particular way of doing things, And with OOP your get more degrees of freedom with things you can do with a class and things can be modified easier, think sort of problems like a board game just as an example treating the pieces as an object can make it easier to modify any section of the logic of the game and add new pieces then the typical functional approach.
Well, what I don't understand is the SW engineering propensity towards cargo culting and doing things the harder way than mathematicians do.
I see it with OOP, XML, design patterns, and other things, now with LLMs. (LLMs - you really want to build complex systems in badly specified natural language, compiled or interpreted with an inscrutable algorithm and possibly indeterministic?)
Is it the excitement from a new analogy? Or are engineers naturally empiricists, while the rationalism/empiricism distinction doesn't work in computer science?
I think we would be better off if we just learned functional programming and few abstract concepts (categories, monads). Simpler than OOP.
But functional programming and OOP are intrinsically the same, but expressed using different primitives.
The thing that empowers first-class functions and closures (lexical binding) is the exact same method by which encapsulation works, even if the latter opts for heap vs stack. The fundamentals are the same.
Here's a simple one in Emacs Lisp:
(defmacro send (object method &rest args)
"Sends a METHOD to an OBJECT with ARGS."
`(funcall ,object ',method ,@args))
(defun make-animal (name)
;; 'hunger' is entirely private (encapsulated)
(let ((hunger 5))
;; here's our lambda using lexical binding
(lambda (method &rest args)
(cond
((eq method 'get-name) name)
((eq method 'feed)
(setq hunger (max 0 (1- hunger)))
"Yum!")
(t (error "Animal does not understand: %s" method))))))
(let ((good-boy (make-animal "Rex")))
;; "call" (via our macro) the "get-name" method; then feed. hunger does down 1.
(send good-boy get-name)
(send good-boy feed)
;; pretty-print the "object" good-boy
(pp good-boy))
;; this is the state of it after the two calls.
#[(method &rest args)
((cond ((eq method 'get-name) name)
((eq method 'feed) (setq hunger (max 0 (1- hunger))) "Yum!")
(t (error "Animal does not understand: %s" method))))
((hunger . 4) (name . "Rex"))]
Yes, you can express everything using lambda calculus. Kinda.. so why not use a thing that already exists? Why invent a new language (encapsulation) when previous (binding) suffices?
OOP is not about doing things harder, it‘s about doing them efficiently. The key elements of software engineering are abstraction and encapsulation: we, humans, cannot load the entire domain into our memory and reason about it at all levels of detail at once. We need to zoom in and zoom out, decompose problems into smaller pieces, then reassemble, treating those pieces at surface value. OOP isn‘t rocket science, it just adds one more piece — the hierarchy of concepts with inheritance and polymorphism. And it‘s just a technique: most programs are written in a mix of styles anyway. The foundational style is procedural, because that‘s how computers work. FP and OOP just build on top of it and translate to it.
This article describes the OOP approach that leads to object-relational mapping, boilerplate code, database schema duplicated in code, navigational data access and the impedance mismatch. It defines OOP around data modeling and taxonomy, rather than around responsibilities. Principles such as "Tell, don't ask" and "Single responsibility" are just ignored. What responsibility does a book have in a library management system? It doesn't, it's just a subject of recorded facts, and a better approach would be to identify the behavioural components of the solution space, construct those as classes/objects, and let facts be encapsulated in or communicated between objects.
"Object thinking" by David West was eye opening about this back in the day. The gist is to model the objects not along the line of real world entities, but along the line of "behaviour" like you mention.
Yes, OOP is hard to do right. It's not helped by examples and schools largely teaching from real world nouns. Most people in OOP languages are writing procedural code.
For all these reasons I just skip it. I still use OOP languages, but I leave the OOD at the door.
For what it's worth, you can get all that with plain old C. Declare structs up front. Allocate them all as pointers. The C file is the class, and has the actual definition of the structs.
Strangely enough, in a sense, OOP languages makes it harder to do this than in C. Not that C doesn't have other flaws.
And then you start adding some real world books that have some of the attributes missing or uncertain, have several authors, go by different titles or uses some kind of exotic notation to their titling, and this OOP structure kind of crumbles?
Nothing in this article is even OOP-specific. You can do all of this with structs and static dispatch. The code in a language like Go or Zig that supports the dot-notation would look pretty similar too.
The article feels a bit late since we have come full circle: OOP attempted to force natural language ontologies into deterministic code, while LLMs can now process natural language ontologies directly. Prose has become the source code humans write (or at least read), and model weights are the new compiler that extract objects (or functions) and relationships directly from a written specification to generate Python. OOP hasn't disappeared; it has finally been supplanted by the natural language it always tried to mimic.
Sometimes or even quite often considering objects as records is reasonable approach. It has set of fields. And maybe functions helpful to manipulate these fields or even use them. Even more complicated things are just records.
And then sometimes it is useful to extend these functionalities for specific special cases.
Maybe OOP is often taught going too abstract. And trying to model wrong things. But thinking of it as grouping fields and then adding functions to manipulate or use those and things related to them is useful.
> “That's because I thought it made sense to represent a book as a sequence of Unicode code points stored along with the cumulative lengths of the individual fields. But you don't have to worry about it, since the auxiliary functions book_xxx are the ones that represent the interface that users should interact with.”
I’m not saying you should ever fire a coworker out of a cannon. I’m just saying that I wouldn’t always hold it against you.
My pet hate is not any paradigm but python, I really wish it did not become as big as it did. I don't think the language does a single thing well just a lot of things 'Ok'.
Python people used to be irrationally anti-fp I guess that has changed, though the language is geared towards mutability and a fairly heavy weight OOP structure, that does not do justice to what OOP can be. fp in python used to be needlessly painful and lambda functions where discouraged and deemed 'unpythonic'.
40 comments
[ 0.30 ms ] story [ 18.6 ms ] threadOh, it also helps me to pay my bills, because someone has to untangle the mess.
And it helps my therapist, because he has to keep the madness that grows inside of me in check.
Jokes aside: I was taught OOP at university.
I was also taught functional programming, answer set programming and other forms such as logical programming.
Focus was definitely on OOP though.
I mainly see OOP as a way to (dis) organizer code and a way for hardware manufacturers to sell more hardware.
In reality this seems to make coupling even worse. Now you have to dig 8 classes deep just to find the business logic for a particular feature. Changing it is even harder because the rest of the program expects this part to behave in a very specific way.
I agree with the reality.
Rigid class-based inheritance (especially multi-inheritance) is what sucks. This over-engineered complexity really only benefits things like window toolkits where you want super rigid modeling of widget types. But even that's a stretch.
Traits do OOP the right way. Complete flexibility.
Does not need classes.The "behaviours" being modelled here were data access. Writing .name() instead of .name.
You can save yourself the time of manually packing these structs by writing out a constructor in full. Which the article called "automatic".
(You don't even need to write out the constructor for a struct in C99. Probably any other modern language too)
Yes. And structs are just bytes. Levels of abstractions don't do anything the underlying levels don't already do, they just give you means to express your intent in a clearer form.
No, it won't. I'm especially sad that this, like so many OOP guides before it, did not go into how it interacts with data structures and bigger picture stuff. It had the perfect chance to do that as it could have made a Catalog class and then had a discussion about what belongs to that and what to the books. Instead it started to abstract on author, in the stupidest way possible (no, no one will ever look up a book using the author birth date).
What people forget -- much like the design patterns in the gang of four book -- is that languages and frameworks evolve.
A decorator pattern was a niche but useful abstraction in the 1990s. In Python today you can @decorate stuff just like that. It's evolution.
The same holds for OOP. Encapsulation and co-located methods with the encapsulating slots was an incredibly powerful upgrade over basic structs. Now most languages have first-class functions and lexical scoping so you can build your own encapsulation that way.
It's all good. Just use whatever fits best.
Most complex codebases I've seen which were pure Functional Programming were spaghetti code; unmaintainable.
What I saw every single time was that the project was syncing a huge amount of state in a central place and then passing it through a large number of components and sub-components. The top level component basically had to have full awareness of everything going on inside the system in order to do its job and there were no separation of responsibilities because none of the components had sovereignty over the state they needed to do their job independently. It's just components micro-managing components all the way down.
React with Redux is probably the best FP implementation I've seen thus far but even it is kind of a mess. I have nightmares about Redux Saga. The Redux Saga logo is literally a stylized drawing of entangled spaghetti.
Had Redux + Saga been presented to people as part of React at the time it was introduced, React would not have become popular. Because people would have understood that it creates too much unnecessary baggage which isn't worth it.
Isn't one of the tenants of FP that nothing is mutable anywhere?
But I also spent some time with modern Spring Boot, and I get where the hate is coming from. There is a lot of complexity that just doesn't seem necessary at all.
I'm sure there is some (very high) level of inherent project complexity where the massive number of abstractions and tools that Spring Boot gives you (and forces on you) actually starts turning into a net positive... And I'm also fairly sure that most projects don't quite break even, and would be better off with something more lightweight.
also a lot of oop hate is 1) hatred of compile-time hierarchies attempting to match the domain model and 2) all the complexity in oop systems which isn't really inherent to OOP but made possible by it. oop doesn't require a PrototypeFactoryInjectorFactoryBean but it doesn't prevent you either :)
on the flip side I've seen functional codebases which are also full of complexity because of how fragmented everything is, like 20 functions all applied when it could have been a single imperative one making it hard to keep track of what is happening.
and of course everything in between
I think the hate typically comes from the mis-application of various tools; overly complex solutions for simple problems. There's nothing wrong with patterns per say, or frameworks that have complex building blocks for every conceivable permutation of relatively common issues. However, when people reach for unnecessarily complex tools out of laziness/habit, some hate is justified. I suspect this is very similar to what you're saying, and of course, it's not OOP specific. Functional languages have their patterns too, but instead of a famous book enumerating them, we have a famous blog post stating that functional languages are so awesome that they don't need patterns (or at least, this is how most people seem to interpret it).
Many who've entered the career in the last decade or so seem to have missed what I 'd consider to be quite a fundamental grounding.
Not even advanced concepts from the gang of four book - though I have taught these too - but often just an un-awareness of OOP in general, even simple things like understanding encapsulation.
Frameworks and languages do evolve, but never being taught basic things like encapsulating state or exposing dependencies seems an issue. Very differnt from my own time as a junior when this stuff was constantly hammered home.
I used to be OOP hater, but realized that it has it's place and some problems are OOP shaped and OOP is a wide range of techniques. I do think that python is a sloppy language and generally don't like it.
Functional programming has it's place and I love it and is probably what my mind goes to the most, but often in Functional languages I find you can 'over fit' to a particular way of doing things, And with OOP your get more degrees of freedom with things you can do with a class and things can be modified easier, think sort of problems like a board game just as an example treating the pieces as an object can make it easier to modify any section of the logic of the game and add new pieces then the typical functional approach.
I see it with OOP, XML, design patterns, and other things, now with LLMs. (LLMs - you really want to build complex systems in badly specified natural language, compiled or interpreted with an inscrutable algorithm and possibly indeterministic?)
Is it the excitement from a new analogy? Or are engineers naturally empiricists, while the rationalism/empiricism distinction doesn't work in computer science?
I think we would be better off if we just learned functional programming and few abstract concepts (categories, monads). Simpler than OOP.
The thing that empowers first-class functions and closures (lexical binding) is the exact same method by which encapsulation works, even if the latter opts for heap vs stack. The fundamentals are the same.
Here's a simple one in Emacs Lisp:
For all these reasons I just skip it. I still use OOP languages, but I leave the OOD at the door.
For what it's worth, you can get all that with plain old C. Declare structs up front. Allocate them all as pointers. The C file is the class, and has the actual definition of the structs.
Strangely enough, in a sense, OOP languages makes it harder to do this than in C. Not that C doesn't have other flaws.
And then sometimes it is useful to extend these functionalities for specific special cases.
Maybe OOP is often taught going too abstract. And trying to model wrong things. But thinking of it as grouping fields and then adding functions to manipulate or use those and things related to them is useful.
I’m not saying you should ever fire a coworker out of a cannon. I’m just saying that I wouldn’t always hold it against you.
Python people used to be irrationally anti-fp I guess that has changed, though the language is geared towards mutability and a fairly heavy weight OOP structure, that does not do justice to what OOP can be. fp in python used to be needlessly painful and lambda functions where discouraged and deemed 'unpythonic'.