[kragen]

On the contrary, Java objects are very different from C++ objects, precisely because they lack a lot of the "primitive-like" features of C++ objects such as copying, embedding as fields, and embedding in arrays. (I'm tempted to mention operator overloading, but that's just syntactic sugar.)

[dpark]

Java differs from C++ in an endless number of ways.

What I’m saying is that in both C++ and Java, there are a set of primitive types that do not participate in the “object-orientedness”. C++ primitives do not have class definitions and cannot be the base of any class. This is very much like Java where primitives exist outside the object system.

If the C++ standard used the term “entities” instead of “objects” I don’t think this would even be a point of discussion.

[kragen]

It's not some minor point of terminology.

The entire design of C++ is built around eliminating all distinctions between primitive "entities" and user-defined "entities" in a way that Java just isn't. It's true that you can't inherit from integers, but that's one of very few differences. User-defined "entities" don't (necessarily) have vtables, don't have to be heap-allocated, can overload operators, can prevent subclassing, don't necessarily inherit from a common base class, etc.

C++'s strange definition of "object" is a natural result of this pervasive design objective, but changing the terminology to "entity" wouldn't change it.

[dpark]

> The entire design of C++ is built around eliminating all distinctions between primitive "entities" and user-defined "entities"

If the intent was to erase all distinction between built-in and user-defined entities then making the primitive types unable to participate in object hierarchies was a pretty big oversight.

But at this point I think we’re talking past each other. Yes, in Java objects are more distinct from primitives than in C++. But also yes, in C++ there is a special group of “objects” that are special and are notably distinct from the rest of the object system, very much like Java.

[kragen]

You can read Stroustrup's books and interviews, if the language design itself doesn't convey that message clearly enough; you don't have to guess what his intentions and motivations were. And, while I strongly disagree with you on how "special and notably distinct" primitive types are in C++, neither of us is claiming that C++ is less adherent to the principle that "everything is an object" than Java. You think it's a little more, and I think it's a lot more.

But we agree on the direction, and that direction is not "Java [did something] by deciding everything must be an object," but its opposite.

[dpark]

I don’t actually think it’s any more adherent to that notion. This is exactly why I tried to point out the discrepancies in definitions. You have to define what an “object” is or the discussion is meaningless.

If the definition of object is something like “an instance of a class that has state, operations, and identity” then C++ primitives are fundamentally not objects. They have no identity and they are not defined by a class. If “participates in a class hierarchy” is part of the definition, then C++ is way less OO than Java.

I don’t quite understand what your definition is, but you seem to be arguing that user-defined entities are more like primitives in C++, so it’s more object-oriented. So maybe “consistency across types == object orientedness”? Except C++ isn’t really more consistent. Yes, you can create a user-defined type without a vtable, but this is really a statement that user defined types a far more flexible than primitives. But also if “consistency across types” is what makes a language OO then C seems to be more OO than C++.

[kragen]

I don't think C++ is object-oriented, and it is certainly way less OO than Java in most ways. Its "classes" aren't the same kind of thing as classes in OO languages and its "objects" aren't OO objects.

In part this is because by default even C++ class instances don't have identity, or anyway they only have identity in the sense that ints do, that every (non-const) int has an address and mutable state. You have to define a destructor, a copy constructor, and an assignment operator to give identity to the instances of a class in C++.

With respect to "participates in a class hierarchy", that has not been part of the definition of OO since the Treaty of Orlando. But, in Java, all objects do participate in a class hierarchy, while no primitives do, while, in C++, you can also create class instances (or "class" "instances") that do not participate in a class hierarchy (without ever using inheritance, even implicitly). So, regardless of how OO or non-OO it may be, it's another distinction that Java draws between primitives and class instances ("objects" in Java) that C++ doesn't.