People Don't Understand OOP
blog.sigma-star.io
blog.sigma-star.io
Therefore, before we rush to say that other people don't understand OOP, perhaps we should consider the possibility that WE don't understand OOP if we insist on applying it where it doesn't apply.
What it is about the OOP paradigm that you find does not fit this client->relational database application? The idea behind OOP is that you send a message out into the world, and maybe another actor can make something of it. You don't really know or care about the other actors are out there. If an actor hides a relational database behind closed doors, so be it. Not your problem. That's outside of the concern of OOP.
On the flip side, we have active record, which imagines a way to try and marry relational databases to an object-based (not oriented) model. Here you will regularly find criticism about "impedance mismatch" and where things like constraints should be handled. This still reads like what you are talking about, but has little to do with OOP.
Conceivably you could use that pattern to implement the local retention of an actor. Is that where you find applicability to OOP? OOP doesn't speak to the underlying actor implementation, though. It purposefully doesn't speak to the implementation as not having knowledge of actors is kind of OOP's whole deal. If you need to know things about other actors, you can't just throw a message out there.
Because of the similar names, those who don't understand OOP oftentimes confuse it with object-based programming. Perhaps this is what you are trying to speak to, using a 'database layer' as an example of object-based programming as seen in the wild?
Admittedly, I still don't fully understand the link you are making between relational databases and OOP.
It's not me who finds that OOP is a mismatch with relational systems. It's me and many other people. Ted Neward talked about this 20 years ago. Jeff Atwood talked about it 18 years ago. Martin Fowler defended OOP from this criticism but at least he also talked about it, 12 years ago. There's a Wikipedia page on the subject and many related articles over the years. This is well-trod ground.
Now. Does this mean we critics are right? No. It doesn't mean that. It means that this is well-trod ground. It is the topic of a healthy debate that generates one line of criticism of OOP. Functional programming--which I haven't mentioned in this thread until now--is another source of OOP criticism. What's interesting to me at this particular moment is not the substance of these debates ("is the object-relation impedance mismatch real? is mutable state really bad?") or who's right. What's interesting to me is that author of Sigma's Blog doesn't address any of this. I claim that these two topics--the ostensible object-relational impedance mismatch and the mutable state bugaboo of functional programming--are of such primary importance in generating criticism of OOP that any author who sets out to defend OOP from its critics must address these topics in order to be taken seriously. And yet, all this author says is
It seems like a lot of people dislike Object Oriented Programming
They don't say who these people are, what their reasons are for disliking OOP, or why things seem this way to the author. They just accuse unnamed "people" of not understanding OOP and then dive into a bog-standard description of OOP which is well-understood by all the relevant programmers that I know.
None of that answers your questions on the topic of the object-relational mismatch, of course, but then I never promised to explain that topic to you. I might, but then again I might not. If I don't, there are plenty of resources online that you can study for yourself. If you form a killer argument that addresses all issues once and for all, perhaps you should write your own blog post. Surely such a groundbreaking achievement would be worthy of making it all the way to the front page.
Great, but what is the mismatch you speak of?
> Ted Neward talked about this 20 years ago. Jeff Atwood talked about it 18 years ago.
Are you referring to the "The Vietnam of Computer Science"? It is about active record and object/relational mapping.
OOP, despite the poorly chosen name (which Kay regrets), isn't about objects at all. As is defined, OOP is about "messaging, local retention and protection and hiding of state-process, and extreme late-binding of all things." But perhaps there is some other work by these people that I'm not familiar with that speaks to OOP?
> It seems like a lot of people dislike Object Oriented Programming
Sure. I expect all people dislike object-oriented programming in practice, which is why you never truly see it. Sending a message and maybe getting a response (i.e. extreme late-binding) is cool on a theoretical level, but not exactly practical when it comes to real applications that want to have certain guarantees. The rising love for statically typed languages highlights that people want such guarantees.
Ironically, OOP probably only shines for things like client->relational database systems that cannot guarantee the state of the database at compile time. At least the paradigm gets you thinking about what to do when an actor doesn't respond to your message.
Go do some independent study and then maybe we'll talk.
You keep pointing to the object-relational mismatch, which, while a thing, has no more to do with OOP than it does functional programming. It pertains to object-based programming. That is something else entirely.
Understandably, those who don't understand OOP often confuse OOP with object-based programming. Is that the problem here? Just that you don't understand OOP?
Fair that fluidity of language means that anyone can define OOP as they wish, and if they want to define OOP to be equivalent to OBP, good for them. They wouldn't be the first. And if we were discussing this topic in a vacuum we could set out a proclamation that defines OOP to be equal to OBP. However, OOP was already and explicitly defined by the article that setup the context for the following discussion, and the definition used is Kay's – the one that has nothing to do with objects. Other definitions for OOP would be clearly off-topic.
Perhaps it is that you are purposefully trying to go off-topic? Indeed, this thread started because someone (not me!) questioned exactly that.
If it is simply that you don't know how to clarify what you mean, why not just say so?
I don't have to. That's already explained in the standard literature on the topics of the object-relational mismatch and on functional critique of mutable state.
> You keep pointing to the object-relational mismatch, which, while a thing, has no more to do with OOP than it does functional programming. It pertains to object-based programming. That is something else entirely.
And I say that the object-relational mismatch does have more to do with OOP than it does with functional programming. If you want to make unfounded claims, two can play at that game.
> However, OOP was already and explicitly defined by the article that setup the context for the following discussion, and the definition used is Kay's
Point to the place on the doll where the article explicitly defines OOP and uses Kay's definition to do it.
> Perhaps it is that you are purposefully trying to go off-topic? Indeed, this thread started because someone (not me!) questioned exactly that.
And yet that person never came back. You can't rule out the possibility that this person was satisfied by my subsequent clarifications.
> If it is simply that you don't know how to clarify what you mean, why not just say so?
You first.
Understood. You have asserted that many times now. But what you are trying to say has not been made clear.
Let's break this down. As defined by context, OOP means: messaging, local retention and protection and hiding of state-process, and extreme late-binding of all things. Indeed, there exist other OOP definitions, but that is the one that was defined here.
I think we can rule out "local retention and protection and hiding of state-process" from the discussion. Hiding means that even if there is some kind of relational database access happening behind the curtain, we cannot know about it. If we don't know about it, we cannot talk about it.
That leaves messaging and extreme late-binding. It is not clear to me what it is about messaging and/or extreme late-binding that is at odds with relational databases. This is where I seek clarification.
While you pointed to the work of Neward and Atwood, all I could find from them was discussion about the pitfalls of trying to model relational databases as a particular object model. They say nothing about messaging or extreme late-binding. Unfortunately, they were of no help here.
Point to the place on the doll where the article author defines OOP.
Given that you don't know how to clarify your original message, perhaps we can still salvage something here? Let's assume that OOP is by another definition. Your choice. Define it however you please. Then explain how relational databases fit into that definition.
No, this has been covered exactly once. I asked 4 hours ago in this sub-thread this question, but you didn't answer it. I also asked again 3 hours ago in this sub-thread. I also asked a related question in another sub-thread 3 hours ago. You answered it one time 2 hours ago with "First thing you read. That's where definitions go." in the other sub-thread, then 1 hour ago in this sub-thread to its second query you said This has been covered several times which it absolutely not has been. THIS, dear readers, is what we call "bad faith."
> Given that you don't know how to clarify your original message, perhaps we can still salvage something here? Let's assume that OOP is by another definition. Your choice. Define it however you please. Then explain how relational databases fit into that definition.
Absolutely not. You have acted in bad faith, have been argumentative, have been snide, have been condescending, and have made things personal throughout these debates. I will absolutely not be corresponding with you in HN comments on this topic or on any other any longer. I would do so "in-person" in a Zoom call if you had the stones for it, but I doubt you do.
I count three times in this thread (plus apparently again in another thread, although that has no relevance to this one). I respect that I did not convey it in an understandable way. Communication is hard. But it is not clear to me what was interpreted instead, in order to allow me to turn the ship in the right direction.
> I will absolutely not be corresponding with you in HN comments on this topic or on any other any longer.
Because we wouldn't be able to figure that if you just stopped corresponding...?
> I would do so "in-person" in a Zoom call
Sure, I'll listen in if you want to expand upon how relational databases pertain to your definition of OOP. If you seek a debate, as suggested above, no thanks. High school debate club is best left for high school students. I did not engage in debate with you here for good reason, and I will not engage in debate elsewhere either.
Regardless, the ask was "Perhaps you could clarify what you were trying to say with respect to OOP?", not "Have you used (active record|ActiveRecord)?" Even if you somehow managed to not see the space, and didn't notice the capitalization, the response still doesn't logically follow.
Did you not read the comment?
I said that ActiveRecord was popularized by Ruby on Rails in 2004. I feel I was quite clear on this subject.
> Regardless, the ask was "Perhaps you could clarify what you were trying to say with respect to OOP?"
Sorry. I wrote my comment before you had revised yours. If you'd like to let me know when you're done with your revisions, I'm happy to wait.
Thanks for the random history lesson and all, but why are you bringing up ActiveRecord in the first place? The topic is OOP, not a library that is bundled with a web framework created many years ago.
Did you not even read your own comment? For a brief moment you seemed to recognize that the subject is about OOP.
Why? I enjoy and benefit from the fact that you think a software library and a theoretical concept are the same thing. Never change.
> I'm talking about OOP and not active record.
Yes. We established that in my very first comment. However, I stated that it reads like you were talking about active record, and that is why it was asked if you could clarify how it pertains to OOP, given that we know you are actually talking about OOP.
If you don't know how to clarify your message, why not just say so?
One of the most poisonous things about the "class" whether that's in Java or C++ is that it looks as though this procedure called "chirp" and this state called "loudness" are snuggled up together in my Bird class, after all in my source code they are literally on adjacent lines!
Of course they're not, the actual mechanics are the same as in say Rust, where "chirp" lives in an impl block, while "loudness" is part of the actual data structure, or a language with full blown Uniform Method Call Syntax where the relevance of Bird to a "chirp" procedure is just spelled out wherever "chirp" is defined, not in the data structure of Bird itself.
I think the heart of OOP is the opacity, the fact that we (for some definition of we) aren't allowed to poke around in the state directly, and this article talks about that idea at length, critiquing people who build a whole bunch of getter and setter mechanics for what is, in practice, just Plain Old Data and not really an object at all. So I think I'd start there, not by adding procedures.
So, I think a Doodad which may have internal state but we can't touch it, is an Object, even if there seemingly aren't any procedures for Doodads at all. But a Thingy, which has six named integers and a string inside it, that's not an object in the OOP sense, that's just some data, even if I have a whole tonne of procedures for Thingys.
This also makes sense when I don't have total information. If you give me a Thingy, I can do things with that, regardless of whether you provide procedures, but if you give me a Doodad, I can't do anything with that unless you provide procedures. Maybe you have procedures, but I can't necessarily use them.
At the heart of OOP is being able to throw a message out into the world and maybe find another actor willing to respond to it. It is not so much that you are not allowed to poke around in the state of other actors, but that you fundamentally cannot know what other state exists without breaking the paradigm.
> So I think I'd start there, not by adding procedures.
Agreed. A thin veneer over internal state does not fix a break in the paradigm.
> if you give me a Doodad, I can't do anything with that unless you provide procedures.
You aren't even given Doodad, per se. Under OOP, you just send the message out there. If Doodad chooses to respond, then so be it. If Doohickey responds instead, such is life. Who responds is not your concern.
This is one of the issues I've experienced with overly eager "everything is a class" style of programming. The plain old data gets wrapped up and partially hidden, complected together with procedures.
The other aspect is that plain old functions get wrapped up too, and can only be accessed by instantiating the class, and complected together with state.
This style of code can make the underlying data and algorithms harder to work with, less flexible, too specific to the exact architecture designed by the author - especially as hierarchy of classes, which extend the "complectedness" in more dimensions where they're tied up with each other.
I just want typed and structured "thingies" as plain old data, that lets me reach in, access and modify everything as my own. Then separately, I want stateless plain old functions that accept those and other thingies of the same shape, wherever they come from, however they're created. That way I can use them by themselves, without being tied to any associated state, for my own purposes.
Sometimes I think the problem is too much "architecture", where all the parts are fused and can only be used together; whereas I prefer a "toolset" or "workshop", a coherent collection of parts that are independently useful and usable.
I see it as an important watershed because it seems to me the reliability and particularly performance reliability went downhill around the time this technology went mainstream. I remember plain old X Windows was rock solid but once KDE and Gnome came around that were based on this technology the Linux desktop became about as reliable as the Windows desktop.
“Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it.” ~Dijkstra probably?
I get that many people either don't see, or don't place much importance on the invariants of their data and so they are happy writing functions and even get annoyed by people and their over use of objects. Maybe these people are just way smarter than me; I definitely need objects/classes to help me organize my code.
Take Rust's core::net::Ipv4Addr, that's obviously just a 32-bit value inside - it has no invariants to be maintained, but it's opaque, even though I know it's a 32-bit value, I am actually not able to just directly treat it as the value or prod that value, I have to use certain procedures, which include predicates (to ask if this is, for example, the broadcast address, or a loopback address) and some useful bit operations,
Inside of some class, you can then implement whatever you want inside of that class, ie:struct of arrays, or w/e.
I don't view "inheritance" as a necessity of OOP. Composition fits great. I don't view "array of structs" as a necessity of OOP, nor over usage of heap allocations and pointer chasing. People associate OOP with this, but it's not fundamental, it's a poor, out dated mutation.
It seems like when people say "OOP is bad" they really mean "over abstractions are bad", and yes, they are bad, but "data oriented design" and "procedural code" isn't some silver bullet that's gonna save you from writing trash code.
Ultimately, I consider ideal code to be "verifiable on sight". Ie; I can read it easily and mentally verify that it is correct. Bad code is code that is very hard to verify correctness by reading. There's an art to this and it doesnt always fit into these semantic buckets of "OOP" or "Data Oriented Design".
Things like functional programming, OOP, declarative programming, etc, might matter to language designers. They need these concepts to work out if their language is convenient/internally consistent/featureful and maybe these banner concepts help. Once the language has been built, I don't think programmers actually care. The abstraction isn't at the sweet spot where a developer gets the most power. The components of OOP (encapsulation, message passing, interfaces) are important and should be used liberally - but grouping them together under one heading is detrimental, because there will also be opportunities for other techniques from other schools of thought.
And as an aside inheritance seems to have been a mistake. I haven't seen a lot of places where inheritance was the best idea. Interfaces yes, inheritance no.
I like to use interfaces to expose the public side of things, but I find inheritance can be really powerful in terms of code-reuse for implementing those interfaces.
Of course if you got a language that makes encapsulation very smooth then inheritance might not be needed, but otherwise I find it can really cut down on code re-use and boilerplate to have a base class for certain interfaces.
Of course, since the interface is the public bit, if the base class is not a good fit for a certain implementation there's no requirement to use the common base class. So you use it when it makes sense.
Erlang is probably the only OOP language in existence. I'm not sure it was ever the tool of choice for GUIs.
Bingo
Smalltalk most definitely influenced how we thought about GUIs, but never really solidified itself as the place to develop GUIs. Class-based, non-OOP, languages seemed to be where most GUI development happened. And in large part were suited to the task. There was little need to introduce OOP.
There was COM and the like, which perhaps can reasonably be considered object-oriented, but only at a higher level of abstraction. The programming was not object-oriented. OO, but not OOP.
Has already been defined where exactly? Yes, evidently you believe it has been defined in the article, but where exactly in the article did that occur. Please be precise.
Says who? You? Did you get a confirmation from the article author that this was their intention? After all, the first thing you read in the article is a quote, but then there's a second quote later from a different person, which doesn't specifically address the first quote. Is that also a definition?
No, in reality, you don't know the author's definition. Neither of us does, which is why we have to make an educated guess (an "inference"). You've guessed evidently that the author's definition is identical to Kay's. Well, let's consider your confidence around that guess. When the author discusses other topics not included in that definition, like inheritance and polymorphism, does that tend to increase or decrease your confidence in your guess?
Finally, just as I'm not obliged to use your definition and you're not obliged to use mine, neither of us is obliged to use the blog article author's definition either. I certainly didn't agree to that, and there's good reason not to, which is that there are other people involved: the OOP critics.
The article author accuses OOP critics of not understanding OOP. Personally, I think that gives us a reason to use THEIR definition of OOP, not the author's. That is, unless the author's argument "they don't understand OOP" is essentially "they're using the wrong definition of OOP." But then, since there is no right or wrong definition of OOP, that could only mean, "they're using a different definition of OOP from the one I'm using", which they're free to do.
To sum up, your argument essentially is, "you critics of OOP are not using the definition that I use." OK. Well, who asked you?
Increases. After all, it is ultimately about how people do not understand OOP. The proof is in the pudding, as they say.
> To sum up, your argument essentially is, "you critics of OOP are not using the definition that I use." OK. Well, who asked you?
Says who? You? Did you get a confirmation from the author that this was their intention?
The author says this about "subtyping" and "inheritance", terms that do not appear in Kay's definition at the top:
Another term that – while not technically necessary – is often associated with OOP is inheritance.
Subtyping is not exclusive to OOP
Evidently, subtyping and inheritance are not mutually exclusive of OOP and in fact often co-occur with OOP. The author certainly doesn't say that they in and of themselves--or "polymorphism" for that matter--causes people to misunderstand OOP (though they might misunderstand the terms themselves, which could cause confusion). You and I don't know for sure what the author's definition of OOP is, and yet you're more confident that it doesn't include subtyping, inheritance, and polymorphism because the author mentioned them? Is that your final answer?
> Says who? You? Did you get a confirmation from the author that this was their intention?
What is it you often say? Oh right: "Did you even read" what was written? In this case I didn't say:
To sum up, the author's argument essentially is, "you critics of OOP are not using the definition that I use." OK. Well, who asked you?
Instead I said
To sum up, your argument essentially is, "you critics of OOP are not using the definition that I use." OK. Well, who asked you?
The careful reader will note that I was asking about YOUR argument, not THE AUTHOR'S ARGUMENT. So no, I didn't get confirmation from the author that this was their intention. I was seeking confirmation from you that this was your intention. Well? Is it?
And that's fine. We don't need to know the author's definition. That was deferred to Kay, which stands to reason as Kay defined the term. The author is focused on letting it be known that people, which includes the author himself, don't understand OOP.
In fairness, Kay, who appears to be a person, has made it clear that not even he understands OOP. The only thing he knows for sure is that C++ is not it. Inheritance and polymorphism are core concepts in C++, so given that they were brought up, we have proof positive that the article was written from the perspective of not understanding OOP (in case the title wasn't already clear enough; the author could not have been much more explicit).
> Instead I said – To sum up, your argument essentially is, "you critics of OOP are not using the definition that I use." OK. Well, who asked you?
To which I asked: "Says who? You? Did you get a confirmation from the author that this was their intention?"
As I am the author, I already know the answer, but I wanted to hear the other side of the story. Is there some reason for your strange diversion to avoid answering the question? Are you afraid of what we might learn if you were to become earnest with us? Or did you just not bother reading again?
We don't know that it was deferred to Kay. You just assumed that it was.
> To which I asked: "Says who? You? Did you get a confirmation from the author that this was their intention?"
I apologize. How foolish of me to assume you would notice that I assiduously avoided doing anything other than to refer to you by "you" and to refer to the blog article's author as "the author." Let me rephrase.
As a critic of OOP, my definition of OOP includes inheritance, polymorphism, and classes. An example of a programming language that fits this definition is Java. While my criticisms don't depend on those properties, nevertheless that is the definition that I personally use. Now, based on your comments, I believe your argument is:
as a critic of OOP you are not using the defintion of OOP that I use
That's what I believe your argument is, but I'm interested in your beliefs about your own argument. What do you believe your argument is?
I did notice, indeed. But when I called myself "the author", I was referring to myself.
Is it that you want me to call myself "you" instead? I don't think that works. It is not clear what you are trying to convey here.
Obviously I was the author. Were you under the impression someone else authored the work in question? If so, who? If so, how?
> As a critic of OOP
Was "bingo" meant to a criticism? While this is a fun story that you have randomly decided to share, I fail to see the pertinence.
> What do you believe your argument is?
I have no beliefs on the matter. Logically, once a believe is formed, there is no longer value in discussing it. What more can be learned once you are certain of something? Only before beliefs are formed can you gain insights by talking to other people. Beliefs are built on discussion. Discussion always has to come first.
But maybe you have some pet definition for "belief" that contradicts the common usage recorded in the dictionary?
Alan Kay: an object should be just as powerful as the entire computer, ie an object is a recursive structure.
The rest is implementation details. The implementation details chosen over the past many years, particularly those favored by OOP, generally involve making objects way less powerful than the entire computer.
(The article was a popular link when it was posted to HN back in 2020: https://news.ycombinator.com/item?id=30293622)
The performance comparison is also ridiculous.
The counter points are some random comments by the author’s friends rather than the plethora of excellent criticism out there.
This article is a lot of words without any substance.