We Keep Coding

Interface with default methods vs abstract class, and what's the motivation?

Context
I've recently came across this C# proposal default interface methods
I've read both the specification and more importantly the motivation. Possibly I missed something, but the motivation is a bit stinks me.
The only practical difference between interface and a fully abstract class was that a future class can implement (so be [IS A]) multiple interfaces, but can inherit (so be [IS A]) from only one abstract class, (and all the consequences)
What is not clear for me what is the exact difference between abstract classes and interfaces with default methods now, except that we can bring multiple (implementation) inheritance into the picture with default methods, which is not possible with abstract classes. (I do not want open the question/discussion is it good or bad, this is not the topic here)
However the motivation talks about completely different, three points:
"...API author to add methods to an interface in future versions without breaking source...". Well an "API" author can add methods to an abstract class also in a future version if he implements them without breaking anything.
"...enables C# to interoperate with APIs targeting Android (Java) and iOS (Swift),...". I think a language design decision, especially about abstractions and OOP patterns like multiple inheritance is a bit higher level than interoperate with Swift. I also think, this is only the 0.0x percent of interoperation issues, and also can be solved with other ways.
"...As it turns out, adding default interface implementations provides the elements of the "traits" language feature...". This is a very shallow statement, especially it refers to wikipedia "traits". By definition, traits allow adding methods without multiple inheritance (having [IS A] relationship with a super). However interfaces definitely are about [IS A]... Not talking about the fact, that traits are at least arguable good practice
Question
My question is what is the real difference (or motivation), or what am I missing?

They add this feature in Java 8. So you can add the Java tag and ask to Java developers what they can do with it. It also exists on Haskell and Scala apparently.
Multi-Inheritance
What come to my mind first is Multi-Inheritance. As a class can implement multiple interfaces, you can for example to solve the diamond problem.
In Java this is how they do:
public interface InterfaceA {
public default void foo() {
System.out.println("A -> foo()");
}
}
public interface InterfaceB {
public default void foo() {
System.out.println("B -> foo()");
}
}
private class Test implements InterfaceA, InterfaceB {
// Compilation error : "class Test inherits unrelated defaults for foo() from types InterfaceA and InterfaceB"
}
So you have to either implement the methods (which override the default implementations) or call one of the super:
public class Test implements InterfaceA, InterfaceB {
public void foo() {
InterfaceB.super.foo();
}
}

After introducing Default Method, it seems that interfaces and abstract classes are the same. However, they are still a different concept in Java 8.
The abstract class can define constructors. They are more structured and can have a state associated with them. While in contrast, default method can be implemented only in the terms of invoking other interface methods, with no reference to a particular implementation's state. Hence, both are used for different purposes and choosing between two really depends on the scenario context.

Can I create a non-abstract method in an interface that will be called by every implementation? [duplicate]

I was asked a question, I wanted to get my answer reviewed here.
Q: In which scenario it is more appropriate to extend an abstract class rather than implementing the interface(s)?
A: If we are using template method design pattern.
Am I correct ?
I am sorry if I was not able to state the question clearly.
I know the basic difference between abstract class and interface.
1) use abstract class when the requirement is such that we need to implement the same functionality in every subclass for a specific operation (implement the method) and different functionality for some other operations (only method signatures)
2) use interface if you need to put the signature to be same (and implementation different) so that you can comply with interface implementation
3) we can extend max of one abstract class, but can implement more than one interface
Reiterating the question: Are there any other scenarios, besides those mentioned above, where specifically we require to use abstract class (one is see is template method design pattern is conceptually based on this only)?
Interface vs. Abstract class
Choosing between these two really depends on what you want to do, but luckily for us, Erich Gamma can help us a bit.
As always there is a trade-off, an interface gives you freedom with regard to the base class, an abstract class gives you the freedom to add new methods later. – Erich Gamma
You can’t go and change an Interface without having to change a lot of other things in your code, so the only way to avoid this would be to create a whole new Interface, which might not always be a good thing.
Abstract classes should primarily be used for objects that are closely related. Interfaces are better at providing common functionality for unrelated classes.

When To Use Interfaces
An interface allows somebody to start from scratch to implement your interface or implement your interface in some other code whose original or primary purpose was quite different from your interface. To them, your interface is only incidental, something that have to add on to the their code to be able to use your package. The disadvantage is every method in the interface must be public. You might not want to expose everything.
When To Use Abstract classes
An abstract class, in contrast, provides more structure. It usually defines some default implementations and provides some tools useful for a full implementation. The catch is, code using it must use your class as the base. That may be highly inconvenient if the other programmers wanting to use your package have already developed their own class hierarchy independently. In Java, a class can inherit from only one base class.
When to Use Both
You can offer the best of both worlds, an interface and an abstract class. Implementors can ignore your abstract class if they choose. The only drawback of doing that is calling methods via their interface name is slightly slower than calling them via their abstract class name.

reiterating the question: there is any other scenario besides these
mentioned above where specifically we require to use abstract class
(one is see is template method design pattern is conceptually based on
this only)
Yes, if you use JAXB. It does not like interfaces. You should either use abstract classes or work around this limitation with generics.
From a personal blog post:
Interface:
A class can implement multiple interfaces
An interface cannot provide any code at all
An interface can only define public static final constants
An interface cannot define instance variables
Adding a new method has ripple effects on implementing classes (design maintenance)
JAXB cannot deal with interfaces
An interface cannot extends or implement an abstract class
All interface methods are public
In general, interfaces should be used to define contracts (what is to be achieved, not how to achieve it).
Abstract Class:
A class can extend at most one abstract class
An abstract class can contain code
An abstract class can define both static and instance constants (final)
An abstract class can define instance variables
Modification of existing abstract class code has ripple effects on extending classes (implementation maintenance)
Adding a new method to an abstract class has no ripple effect on extending classes
An abstract class can implement an interface
Abstract classes can implement private and protected methods
Abstract classes should be used for (partial) implementation. They can be a mean to restrain the way API contracts should be implemented.

Interface is used when you have scenario that all classes has same structure but totally have different functionality.
Abstract class is used when you have scenario that all classes has same structure but some same and some different functionality.
Take a look the article : http://shoaibmk.blogspot.com/2011/09/abstract-class-is-class-which-cannot-be.html

There are a lot of great answers here, but I often find using BOTH interfaces and abstract classes is the best route. Consider this contrived example:
You're a software developer at an investment bank, and need to build a system that places orders into a market. Your interface captures the most general idea of what a trading system does,
1) Trading system places orders
2) Trading system receives acknowledgements
and can be captured in an interface, ITradeSystem
public interface ITradeSystem{
public void placeOrder(IOrder order);
public void ackOrder(IOrder order);
}
Now engineers working at the sales desk and along other business lines can start to interface with your system to add order placement functionality to their existing apps. And you haven't even started building yet! This is the power of interfaces.
So you go ahead and build the system for stock traders; they've heard that your system has a feature to find cheap stocks and are very eager to try it out! You capture this behavior in a method called findGoodDeals(), but also realize there's a lot of messy stuff that's involved in connecting to the markets. For example, you have to open a SocketChannel,
public class StockTradeSystem implements ITradeSystem{
#Override
public void placeOrder(IOrder order);
getMarket().place(order);
#Override
public void ackOrder(IOrder order);
System.out.println("Order received" + order);
private void connectToMarket();
SocketChannel sock = Socket.open();
sock.bind(marketAddress);
<LOTS MORE MESSY CODE>
}
public void findGoodDeals();
deals = <apply magic wizardry>
System.out.println("The best stocks to buy are: " + deals);
}
The concrete implementations are going to have lots of these messy methods like connectToMarket(), but findGoodDeals() is all the traders actually care about.
Now here's where abstract classes come into play. Your boss informs you that currency traders also want to use your system. And looking at currency markets, you see the plumbing is nearly identical to stock markets. In fact, connectToMarket() can be reused verbatim to connect to foreign exchange markets. However, findGoodDeals() is a much different concept in the currency arena. So before you pass off the codebase to the foreign exchange wiz kid across the ocean, you first refactor into an abstract class, leaving findGoodDeals() unimplmented
public abstract class ABCTradeSystem implements ITradeSystem{
public abstract void findGoodDeals();
#Override
public void placeOrder(IOrder order);
getMarket().place(order);
#Override
public void ackOrder(IOrder order);
System.out.println("Order received" + order);
private void connectToMarket();
SocketChannel sock = Socket.open();
sock.bind(marketAddress);
<LOTS MORE MESSY CODE>
}
Your stock trading system implements findGoodDeals() as you've already defined,
public class StockTradeSystem extends ABCTradeSystem{
public void findGoodDeals();
deals = <apply magic wizardry>
System.out.println("The best stocks to buy are: " + deals);
}
but now the FX whiz kid can build her system by simply providing an implementation of findGoodDeals() for currencies; she doesn't have to reimplement socket connections or even the interface methods!
public class CurrencyTradeSystem extends ABCTradeSystem{
public void findGoodDeals();
ccys = <Genius stuff to find undervalued currencies>
System.out.println("The best FX spot rates are: " + ccys);
}
Programming to an interface is powerful, but similar applications often re-implement methods in nearly identical ways. Using an abstract class avoids reimplmentations, while preserving the power of the interface.
Note: one may wonder why findGreatDeals() isn't part of the interface. Remember, the interface defines the most general components of a trading system. Another engineer may develop a COMPLETELY DIFFERENT trading system, where they don't care about finding good deals. The interface guarantees that the sales desk can interface to their system as well, so it's preferable not to entangle your interface with application concepts like "great deals".

Which should you use, abstract classes or interfaces?
Consider using abstract classes if any of these statements apply to your use case:
You want to share code among several closely related classes.
You expect that classes that extend your abstract class have many common methods or fields, or require access modifiers other than public (such as protected and private).
You want to declare non-static or non-final fields. This enables you to define methods that can access and modify the state of the object to which they belong.
Consider using interfaces if any of these statements apply to your use case:
You expect that unrelated classes would implement your interface.
For example, the interfaces Comparable and Cloneable are implemented by many unrelated classes.
You want to specify the behavior of a particular data type, but not concerned about who implements its behavior.
You want to take advantage of multiple inheritance of type.
New methods added regularly to interface by providers, to avoid issues extend Abstract class instead of interface.
http://docs.oracle.com/javase/tutorial/java/IandI/abstract.html

Things have been changed a lot in last three years with addition of new capabilities to interface with Java 8 release.
From oracle documentation page on interface:
An interface is a reference type, similar to a class, that can contain only constants, method signatures, default methods, static methods, and nested types. Method bodies exist only for default methods and static methods.
As you quoted in your question, abstract class is best fit for template method pattern where you have to create skeleton. Interface cant be used here.
One more consideration to prefer abstract class over interface:
You don't have implementation in base class and only sub-classes have to define their own implementation. You need abstract class instead of interface since you want to share state with sub-classes.
Abstract class establishes "is a" relation between related classes and interface provides "has a" capability between unrelated classes.
Regarding second part of your question, which is valid for most of the programming languages including java prior to java-8 release
As always there is a trade-off, an interface gives you freedom with regard to the base class, an abstract class gives you the freedom to add new methods later. – Erich Gamma
You can’t go and change an Interface without having to change a lot of other things in your code
If you prefer abstract class to interface earlier with above two considerations, you have to re-think now as default methods have added powerful capabilities to interfaces.
Default methods enable you to add new functionality to the interfaces of your libraries and ensure binary compatibility with code written for older versions of those interfaces.
To select one of them between interface and abstract class, oracle documentation page quote that:
Abstract classes are similar to interfaces. You cannot instantiate them, and they may contain a mix of methods declared with or without an implementation. However, with abstract classes, you can declare fields that are not static and final, and define public, protected, and private concrete methods.
With interfaces, all fields are automatically public, static, and final, and all methods that you declare or define (as default methods) are public. In addition, you can extend only one class, whether or not it is abstract, whereas you can implement any number of interfaces.
Refer to these related questions fore more details:
Interface vs Abstract Class (general OO)
How should I have explained the difference between an Interface and an Abstract class?
In summary : The balance is tilting more towards interfaces now.
Are there any other scenarios, besides those mentioned above, where specifically we require to use abstract class (one is see is template method design pattern is conceptually based on this only)?
Some design patterns use abstract classes (over interfaces) apart from Template method pattern.
Creational patterns:
Abstract_factory_pattern
Structural patterns:
Decorator_pattern
Behavioral patterns:
Mediator_pattern

You are not correct. There are many scenarios. It just isn't possible to reduce it to a single 8-word rule.

The shortest answer is, extend abstract class when some of the functionalities uou seek are already implemented in it.
If you implement the interface you have to implement all the method. But for abstract class number of methods you need to implement might be fewer.
In template design pattern there must be a behavior defined. This behavior depends on other methods which are abstract. By making sub class and defining those methods you actually define the main behavior. The underlying behavior can not be in a interface as interface does not define anything, it just declares. So a template design pattern always comes with an abstract class. If you want to keep the flow of the behavior intact you must extend the abstract class but don't override the main behavior.

In my opinion, the basic difference is that an interface can't contain non-abstract methods while an abstract class can.
So if subclasses share a common behavior, this behavior can be implemented in the superclass and thus inherited in the subclasses
Also, I quoted the following from "software architecture design patterns in java" book
" In the Java programming language, there is no support for multiple inheritance.
That means a class can inherit only from one single class. Hence inheritance
should be used only when it is absolutely necessary. Whenever possible, methods
denoting the common behavior should be declared in the form of a Java interface to be implemented by different implementer classes. But interfaces suffer from the limitation that they cannot provide method implementations. This means that every implementer of an interface must explicitly implement all methods declared in an interface, even when some of these methods represent the invariable part of the functionality and have exactly the same implementation in all of the implementer classes. This leads to redundant code. The following example demonstrates how the Abstract Parent Class pattern can be used in such cases without requiring redundant method implementations."

Abstract classes are different from interfaces in two important aspects
they provide default implementation for chosen methods (that is covered by your answer)
abstract classes can have state (instance variables) - so this is one more situation you want to use them in place of interfaces

This is a good question The two of these are not similar but can be use for some of the same reason, like a rewrite. When creating it is best to use Interface. When it comes down to class, it is good for debugging.

This is my understanding, hope this helps
Abstract classes:
Can have member variables that are inherited (can’t be done in interfaces)
Can have constructors (interfaces can’t)
Its methods can have any visibility (ie: private, protected, etc - whereas all interface methods are public)
Can have defined methods (methods with an implementation)
Interfaces:
Can have variables, but they are all public static final variables
constant values that never change with a static scope
non static variables require an instance, and you can’t instantiate an interface
All methods are abstract (no code in abstract methods)
all code has to be actually written in the class that implements the particular interface

Usage of abstract and interface:
One has "Is-A-Relationship" and another one has "Has-A-Relationship"
The default properties has set in abstract and extra properties can be expressed through interface.
Example: --> In the human beings we have some default properties that are eating, sleeping etc. but if anyone has any other curricular activities like swimming, playing etc those could be expressed by Interface.

Abstract classes should be extended when you want to some common behavior to get extended. The Abstract super class will have the common behavior and will define abstract method/specific behavior which sub classes should implement.
Interfaces allows you to change the implementation anytime allowing the interface to be intact.

I think the answers here are missing the main point:
Java interfaces (the question is about Java but there are similar mechanisms in other languages) is a way to partially support multiple inheritance, i.e. method-only inheritance.
It is similar to PHP's traits or Python's duck typing.
Besides that, there is nothing additional that you truly need an interface for --and you cannot instantiate a Java interface.

Why make Abstract classes and Interfaces?

Well I was going to ask what the difference is but it's been answered before. But now I'm asking why did they make these differences? (I'm speaking about java here, I don't know if the same applies to other languages)
The two things seem very similar. Abstract classes can define a method body whilst interfaces can't, but multiple interfaces can be inherited. So why didn't they (by 'they' I mean Sun when they wrote Java) make one thing where you can write a method body and this type can be inherited more than once by a class.
Is there some advantage in not being able to write a method body, or extend multiple times that I'm not seeing?

Because allowing classes to inherit multiple implementations for the same method signature leads to the obvious question, which one should be used at runtime.
Java avoids this by supporting multiple inheritance only for interfaces. The signatures declared in each interface can be combined much more easily (Java basically uses the union of all methods)

Multiple inheritance in C++ leads to semantic ambiguities like the diamond inheritance problem. MI is quite powerful, but has complex consequences.
Making interfaces a special case also raises the visibility of the concept as a means of information hiding and reducing program complexity. In C++, defining pure abstract bases is a sign of a mature programmer. In Java, you encounter them at a much earlier stage in the evolution of a programmer.

Multiple inheritance is more difficult to implement in a language (compiler really) as it can lead to certain issues. These issues have been discussed here before: What is the exact problem with multiple inheritance.
I've always assumed this was a compromise in Java. Interfaces allow a class to fulfill multiple contracts without the headache of multiple inheritance.

Consider this example:
public abstract class Engine
{
public abstract void switchPowerOn();
public abstract void sprinkleSomeFuel();
public abstract void ignite();
public final void start()
{
switchPowerOn();
sprinkleSomeFuel();
ignite();
}
}
Abstract class can help you with having solid base methods which can or cannot be overriden, but in these methods it uses abstract methos to provide you an opportunity to do your specific thing. In my example different engines have different implementations of how they switch power on, sprinkling some fuel for the ignition, and doing the ignition, however the starting sequence of the engine stays always the same.
That pattern is called "Form Template Method" and is quite frankly the only sensible usage of abstract classes in Java for me.

Making them one thing is the route that the Scala guys took with Traits which is an interface that can have methods and supports multiple inheritance.
I think interfaces, for me, are clean in that they only specify requirements (design by contract) whereas abstract classes define common behaviour (implementation), so a different tool for a different job? Interfaces probably allow more efficient code generation during compile time as well?

The other approach you are describing is the approach used by C++ (mixins for example). The issues related to such "multiple inheritance" are quite complex, and has several critics in C++.

Inheritance means you inherit the nature (meaning) and responsibility (behaviour) of the parent class, while interface implementation means you fulfill a contract (e.g. Serializable), which may have nothing to do with the core nature or responsibility of the class.
Abstract class let you define a nature that you want to be generic and not directly instanciable, because it must be specialized. You know how to perform some high-level tasks (e.g. make a decision according to some parameters), but you don't know the details for some lower-level actions (e.g. compute some intermediary parameters), because it depends on implementation choices. An alternative for solving this problem is the Strategy design pattern. It is more flexible, allowing run-time strategy switching and Null behaviour, yet it is more complex (and runtime swtiching is not always necessary). Moreover, you might lose some meaning & typing facilities (polymorphism & type-checking becomes a bit harder because the Strategy is a component, not the object itself).
Abstract class = is-a, Strategy = has-a
Edit: as for multiple inheritance, see Pontus Gagge's answer.

Why are interfaces preferred to abstract classes?

I recently attended an interview and they asked me the question "Why Interfaces are preferred over Abstract classes?"
I tried giving a few answers like:
We can get only one Extends functionality
they are 100% Abstract
Implementation is not hard-coded
They asked me take any of the JDBC api that you use. "Why are they Interfaces?".
Can I get a better answer for this?

That interview question reflects a certain belief of the person asking the question. I believe that the person is wrong, and therefore you can go one of two directions.
Give them the answer they want.
Respectfully disagree.
The answer that they want, well, the other posters have highlighted those incredibly well.
Multiple interface inheritance, the inheritance forces the class to make implementation choices, interfaces can be changed easier.
However, if you create a compelling (and correct) argument in your disagreement, then the interviewer might take note.
First, highlight the positive things about interfaces, this is a MUST.
Secondly, I would say that interfaces are better in many scenarios, but they also lead to code duplication which is a negative thing. If you have a wide array of subclasses which will be doing largely the same implementation, plus extra functionality, then you might want an abstract class. It allows you to have many similar objects with fine grained detail, whereas with only interfaces, you must have many distinct objects with almost duplicate code.
Interfaces have many uses, and there is a compelling reason to believe they are 'better'. However you should always be using the correct tool for the job, and that means that you can't write off abstract classes.

In general, and this is by no means a "rule" that should be blindly followed, the most flexible arrangement is:
interface
abstract class
concrete class 1
concrete class 2
The interface is there for a couple of reasons:
an existing class that already extends something can implement the interface (assuming you have control over the code for the existing class)
an existing class can be subclasses and the subclass can implement the interface (assuming the existing class is subclassable)
This means that you can take pre-existing classes (or just classes that MUST extend from something else) and have them work with your code.
The abstract class is there to provide all of the common bits for the concrete classes. The abstract class is extended from when you are writing new classes or modifying classes that you want to extend it (assuming they extend from java.lang.Object).
You should always (unless you have a really good reason not to) declare variables (instance, class, local, and method parameters) as the interface.

You only get one shot at inheritance. If you make an abstract class rather than an interface, someone who inherits your class can't also inherit a different abstract class.

You can implement more than one interface, but you can only inherit from a single class

Abstract Classes
1.Cannot be instantiated independently from their derived classes. Abstract class constructors are called only by their derived classes.
2.Define abstract member signatures that base classes must implement.
3.Are more extensible than interfaces, without breaking any version compatibility. With abstract classes, it is possible to add additional nonabstract members that all derived classes can inherit.
4.Can include data stored in fields.
5.Allow for (virtual) members that have implementation and, therefore, provide a default implementation of a member to the deriving class.
6.Deriving from an abstract class uses up a subclass's one and only base class option.
Interface
1.Cannot be instantiated.
2.Implementation of all members of the interface occurs in the base class. It is not possible to implement only some members within the implementing class.
3.Extending interfaces with additional members breaks the version compatibility.
4.Cannot store any data. Fields can be specified only on the deriving classes. The workaround for this is to define properties, but without implementation.
5.All members are automatically virtual and cannot include any implementation.
6.Although no default implementation can appear, classes implementing interfaces can continue to derive from one another.

As devinb and others mention, it sounds like the interviewer shows their ignorance in not accepting your valid answers.
However, the mention of JDBC might be a hint. In that case, perhaps they are asking for the benefits of a client coding against an interface instead of a class.
So instead of perfectly valid answers such as "you only get one use of inheritance", which are relating to class design, they may be looking for an answer more like "decouples a client from a specific implementation".

Abstract classes have a number of potential pitfalls. For example, if you override a method, the super() method is not called unless you explicitly call it. This can cause problems for poorly-implemented overriding classes. Also, there are potential problems with equals() when you use inheritance.
Using interfaces can encourage use of composition when you want to share an implementation. Composition is very often a better way to reuse others objects, as it is less brittle. Inheritance is easily overused or used for the wrong purposes.
Defining an interface is a very safe way to define how an object is supposed to act, without risking the brittleness that can come with extending another class, abstract or not.
Also, as you mention, you can only extend one class at a time, but you can implement as many interfaces as you wish.

Abstract classes are used when you inherit implementation, interfaces are used when you inherit specification. The JDBC standards state that "A connection must do this". That's specification.

When you use abstract classes you create a coupling between the subclass and the base class. This coupling can sometimes make code really hard to change, especially as the number of subclasses increases. Interfaces do not have this problem.
You also only have one inheritance, so you should make sure you use it for the proper reasons.

"Why Interfaces are preferred over
Abstract classes?"
The other posts have done a great job of looking at the differences between interfaces and abstract classes, so I won't duplicate those thoughts.
But looking at the interview question, the better question is really "When should interfaces be preferred over abstract classes?" (and vice versa).
As with most programming constructs, they're available for a reason and absolute statements like the one in the interview question tend to miss that. It sort of reminds me of all the statement you used to read regarding the goto statement in C. "You should never use goto - it reveals poor coding skills." However, goto always had its appropriate uses.

Respectfully disagree with most of the above posters (sorry! mod me down if you want :-) )
First, the "only one super class" answer is lame. Anyone who gave me that answer in an interview would be quickly countered with "C++ existed before Java and C++ had multiple super classes. Why do you think James Gosling only allowed one superclass for Java?"
Understand the philosophy behind your answer otherwise you are toast (at least if I interview you.)
Second, interfaces have multiple advantages over abstract classes, especially when designing interfaces. The biggest one is not having a particular class structure imposed on the caller of a method. There is nothing worse than trying to use a method call that demands a particular class structure. It is painful and awkward. Using an interface anything can be passed to the method with a minimum of expectations.
Example:
public void foo(Hashtable bar);
vs.
public void foo(Map bar);
For the former, the caller will always be taking their existing data structure and slamming it into a new Hashtable.
Third, interfaces allow public methods in the concrete class implementers to be "private". If the method is not declared in the interface then the method cannot be used (or misused) by classes that have no business using the method. Which brings me to point 4....
Fourth, Interfaces represent a minimal contract between the implementing class and the caller. This minimal contract specifies exactly how the concrete implementer expects to be used and no more. The calling class is not allowed to use any other method not specified by the "contract" of the interface. The interface name in use also flavors the developer's expectation of how they should be using the object. If a developer is passed a
public interface FragmentVisitor {
public void visit(Node node);
}
The developer knows that the only method they can call is the visit method. They don't get distracted by the bright shiny methods in the concrete class that they shouldn't mess with.
Lastly, abstract classes have many methods that are really only present for the subclasses to be using. So abstract classes tend to look a little like a mess to the outside developer, there is no guidance on which methods are intended to be used by outside code.
Yes of course some such methods can be made protected. However, sadly protected methods are also visible to other classes in the same package. And if an abstract class' method implements an interface the method must be public.
However using interfaces all this innards that are hanging out when looking at the abstract super class or the concrete class are safely tucked away.
Yes I know that of course the developer may use some "special" knowledge to cast an object to another broader interface or the concrete class itself. But such a cast violates the expected contract, and the developer should be slapped with a salmon.

If they think that X is better than Y I wouldn't be worried about getting the job, I wouldn't like working for someone who forced me to one design over another because they were told interfaces are the best. Both are good depending on the situation, otherwise why did the language choose to add abstract classes? Surely, the language designers are smarter than me.

This is the issue of "Multiple Inheritance".
We can "extends" not more than one abstarct class at one time through another class but in Interfaces, we can "implement" multiple interfaces in single class.
So, though Java doesn't provide multiple inheritance in general but by using interfaces we can incorporate multiplt inheritance property in it.
Hope this helps!!!

interfaces are a cleaner way of writing a purely abstract class. You can tell that implementation has not sneaked in (of course you might want to do that at certain maintenance stages, which makes interfaces bad). That's about it. There is almost no difference discernible to client code.
JDBC is a really bad example. Ask anyone who has tried to implement the interfaces and maintain the code between JDK releases. JAX-WS is even worse, adding methods in update releases.
There are technical differences, such as the ability to multiply "inherit" interface. That tends to be the result of confused design. In rare cases it might be useful to have an implementation hierarchy that is different from the interface hierarchy.
On the downside for interfaces, the compiler is unable to pick up on some impossible casts/instanceofs.

There is one reason not mentioned by the above.
You can decorate any interface easily with java.lang.reflect.Proxy allowing you to add custom code at runtime to any method in the given interface. It is very powerful.
See http://tutorials.jenkov.com/java-reflection/dynamic-proxies.html for a tutorial.

interface is not substitute for abstract class.
Prefer
interface: To implement a contract by multiple unrelated objects
abstract class: To implement the same or different behaviour among multiple related objects
Refer to this related SE question for use cases of both interface and abstract class
Interface vs Abstract Class (general OO)
Use case:
If you have to use Template_method pattern, you can't achieve with interface. Abstract class should be chosen to achieve it.
If you have to implement a capability for many unrleated objects, abstract class does not serve the purpose and you have to chose interface.

You can implement multiple interfaces, but particularly with c# you can not have multiple inheritances

Because interfaces are not forcing you into some inheritance hierarchy.

You define interfaces when you only require that some object implement certain methods but you don't care about its pedigree. So someone can extend an existing class to implement an interface, without affecting the previously existing behavior of that class.
That's why JDBC is all interfaces; you don't really care what classes are used in a JDBC implementation, you only need any JDBC implementation to have the same expected behavior. Internally, the Oracle JDBC driver may be very different from the PostgreSQL driver, but that's irrelevant to you. One may have to inherit from some internal classes that the database developers already had, while another one may be completely developed from scratch, but that's not important to you as long as they both implement the same interfaces so that you can communicate with one or the other without knowing the internal workings of either.

Well, I'd suggest the question itself should be rephrased. Interfaces are mainly contracts that a class acquires, the implementation of that contract itself will vary. An abstract class will usually contain some default logic and its child classes will add some more logic.
I'd say that the answer to the questions relies on the diamond problem. Java prevents multiple inheritance to avoid it. ( http://en.wikipedia.org/wiki/Diamond_problem ).

They asked me take any of the JDBC api
that you use. "Why are they
Interfaces?".
My answer to this specific question is :
SUN doesnt know how to implement them or what to put in the implementation. Its up to the service providers/db vendors to put their logic into the implementation.
The JDBC design has relationship with the Bridge pattern, which says "Decouple an abstraction from its implementation so that the two can vary independently".
That means JDBC api's interfaces hierarchy can be evolved irrespective of the implementation hierarchy that a jdbc vendor provides or uses.

Abstract classes offer a way to define a template of behavior, where the user plugins in the details.
One good example is Java 6's SwingWorker. It defines a framework to do something in the background, requiring the user to define doInBackground() for the actual task.
I extended this class such that it automatically created a popup progress bar. I overrode done(), to control disposal of this pop-up, but then provided a new override point, allowing the user to optionally define what happens after the progress bar disappears.
public abstract class ProgressiveSwingWorker<T, V> extends SwingWorker<T, V> {
private JFrame progress;
public ProgressiveSwingWorker(final String title, final String label) {
SwingUtilities.invokeLater(new Runnable() {
#SuppressWarnings("serial")
#Override
public void run() {
progress = new JFrame() {{
setLayout(new MigLayout("","[grow]"));
setTitle(title);
add(new JLabel(label));
JProgressBar bar = new JProgressBar();
bar.setIndeterminate(true);
add(bar);
pack();
setLocationRelativeTo(null);
setVisible(true);
}};
}
});
}
/**
* This method has been marked final to secure disposing of the progress dialog. Any behavior
* intended for this should be put in afterProgressBarDisposed.
*/
#Override
protected final void done() {
progress.dispose();
try {
afterProgressBarDisposed(get());
} catch (InterruptedException e) {
e.printStackTrace();
} catch (ExecutionException e) {
e.printStackTrace();
}
}
protected void afterProgressBarDisposed(T results) {
}
}
The user still has the requirement of providing the implementation of doInBackground(). However, they can also have follow-up behavior, such as opening another window, displaying a JOptionPane with results, or simply do nothing.
To use it:
new ProgressiveSwingWorker<DataResultType, Object>("Editing some data", "Editing " + data.getSource()) {
#Override
protected DataResultType doInBackground() throws Exception {
return retrieve(data.getSource());
}
#Override
protected void afterProgressBarDisposed(DataResultType results) {
new DataEditor(results);
}
}.execute();
This shows how an abstract class can nicely provide a templated operation, orthogonal to the concept of interfaces defining an API contract.

Its depend on your requirement and power of implementation, which is much important.
You have got so many answer regarding this question.
What i think about this question is that abstract class is the evolution if API.
You can define your future function definition in abstract class but you don't need all function implementation in your main class but with interface you cant do this thing.

Is OOP & completely avoiding implementation inheritance possible?

I will choose Java as an example, most people know it, though every other OO language was working as well.
Java, like many other languages, has interface inheritance and implementation inheritance. E.g. a Java class can inherit from another one and every method that has an implementation there (assuming the parent is not abstract) is inherited, too. That means the interface is inherited and the implementation for this method as well. I can overwrite it, but I don't have to. If I don't overwrite it, I have inherited the implementation.
However, my class can also "inherit" (not in Java terms) just an interface, without implementation. Actually interfaces are really named that way in Java, they provide interface inheritance, but without inheriting any implementation, since all methods of an interface have no implementation.
Now there was this article, saying it's better to inherit interfaces than implementations, you may like to read it (at least the first half of the first page), it's pretty interesting. It avoids issues like the fragile base class problem. So far this makes all a lot of sense and many other things said in the article make a lot of sense to me.
What bugs me about this, is that implementation inheritance means code reuse, one of the most important properties of OO languages. Now if Java had no classes (like James Gosling, the godfather of Java has wished according to this article), it solves all problems of implementation inheritance, but how would you make code reuse possible then?
E.g. if I have a class Car and Car has a method move(), which makes the Car move. Now I can sub-class Car for different type of cars, that are all cars, but are all specialized versions of Car. Some may move in a different way, these need to overwrite move() anyway, but most would simply keep the inherited move, as they move alike just like the abstract parent Car. Now assume for a second that there are only interfaces in Java, only interfaces may inherit from each other, a class may implement interfaces, but all classes are always final, so no class can inherit from any other class.
How would you avoid that when you have an Interface Car and hundred Car classes, that you need to implement an identical move() method for each of them? What concepts for code reuse other than implementation inheritance exist in the the OO world?
Some languages have Mixins. Are Mixins the answer to my question? I read about them, but I cannot really imagine how Mixins would work in a Java world and if they can really solve the problem here.
Another idea was that there is a class that only implements the Car interface, let's call it AbstractCar, and implements the move() method. Now other cars implement the Car interface as well, internally they create an instance of AbstractCar and they implement their own move() method by calling move() on their internal abstract Car. But wouldn't this be wasting resources for nothing (a method calling just another method - okay, JIT could inline the code, but still) and using extra memory for keeping internal objects, you wouldn't even need with implementation inheritance? (after all every object needs more memory than just the sum of the encapsulated data) Also isn't it awkward for a programmer to write dummy methods like
public void move() {
abstractCarObject.move();
}
?
Anyone can imagine a better idea how to avoid implementation inheritance and still be able to re-use code in an easy fashion?

Short answer: Yes it is possible. But you have to do it on purpose and no by chance ( using final, abstract and design with inheritance in mind, etc. )
Long answer:
Well, inheritance is not actually for "code re-use", it is for class "specialization", I think this is a misinterpretation.
For instance is it a very bad idea to create a Stack from a Vector, just because they are alike. Or properties from HashTable just because they store values. See [Effective].
The "code reuse" was more a "business view" of the OO characteristics, meaning that you objects were easily distributable among nodes; and were portable and didn't not have the problems of previous programming languages generation. This has been proved half rigth. We now have libraries that can be easily distributed; for instance in java the jar files can be used in any project saving thousands of hours of development. OO still has some problems with portability and things like that, that is the reason now WebServices are so popular ( as before it was CORBA ) but that's another thread.
This is one aspect of "code reuse". The other is effectively, the one that has to do with programming. But in this case is not just to "save" lines of code and creating fragile monsters, but designing with inheritance in mind. This is the item 17 in the book previously mentioned; Item 17: Design and document for inheritance or else prohibit it. See [Effective]
Of course you may have a Car class and tons of subclasses. And yes, the approach you mention about Car interface, AbstractCar and CarImplementation is a correct way to go.
You define the "contract" the Car should adhere and say these are the methods I would expect to have when talking about cars. The abstract car that has the base functionality that every car but leaving and documenting the methods the subclasses are responsible to handle. In java you do this by marking the method as abstract.
When you proceed this way, there is not a problem with the "fragile" class ( or at least the designer is conscious or the threat ) and the subclasses do complete only those parts the designer allow them.
Inheritance is more to "specialize" the classes, in the same fashion a Truck is an specialized version of Car, and MosterTruck an specialized version of Truck.
It does not make sanse to create a "ComputerMouse" subclase from a Car just because it has a Wheel ( scroll wheel ) like a car, it moves, and has a wheel below just to save lines of code. It belongs to a different domain, and it will be used for other purposes.
The way to prevent "implementation" inheritance is in the programming language since the beginning, you should use the final keyword on the class declaration and this way you are prohibiting subclasses.
Subclassing is not evil if it's done on purpose. If it's done uncarefully it may become a nightmare. I would say that you should start as private and "final" as possible and if needed make things more public and extend-able. This is also widely explained in the presentation"How to design good API's and why it matters" See [Good API]
Keep reading articles and with time and practice ( and a lot of patience ) this thing will come clearer. Although sometime you just need to do the work and copy/paste some code :P . This is ok, as long you try to do it well first.
Here are the references both from Joshua Bloch ( formerly working in Sun at the core of java now working for Google )
[Effective]
Effective Java. Definitely the best java book a non beginner should learn, understand and practice. A must have.
Effective Java
[Good API]Presentation that talks on API's design, reusability and related topics.
It is a little lengthy but it worth every minute.
How To Design A Good API and Why it Matters
Regards.
Update: Take a look at minute 42 of the video link I sent you. It talks about this topic:
"When you have two classes in a public API and you think to make one a subclass of another, like Foo is a subclass of Bar, ask your self , is Every Foo a Bar?... "
And in the minute previous it talks about "code reuse" while talking about TimeTask.

The problem with most example against inheritance are examples where the person is using inheritance incorrectly, not a failure of inheritance to correctly abstract.
In the article you posted a link to, the author shows the "brokenness" of inheritance using Stack and ArrayList. The example is flawed because a Stack is not an ArrayList and therefore inheritance should not be used. The example is as flawed as String extending Character, or PointXY extending Number.
Before you extend class, you should always perform the "is_a" test. Since you can't say Every Stack is an ArrayList without being wrong in some way, then you should not inheirit.
The contract for Stack is different than the contract for ArrayList (or List) and stack should not be inheriting methods that is does not care about (like get(int i) and add()). In fact Stack should be an interface with methods such as:
interface Stack<T> {
public void push(T object);
public T pop();
public void clear();
public int size();
}
A class like ArrayListStack might implement the Stack interface, and in that case use composition (having an internal ArrayList) and not inheritance.
Inheritance is not bad, bad inheritance is bad.

You could also use composition and the strategy pattern.link text
public class Car
{
private ICar _car;
public void Move() {
_car.Move();
}
}
This is far more flexible than using inheritance based behaviour as it allows you to change at runtime, by substituting new Car types as required.

You can use composition. In your example, a Car object might contain another object called Drivetrain. The car's move() method could simply call the drive() method of it's drivetrain. The Drivetrain class could, in turn, contain objects like Engine, Transmission, Wheels, etc. If you structured your class hierarchy this way, you could easily create cars which move in different ways by composing them of different combinations of the simpler parts (i.e. reuse code).

To make mixins/composition easier, take a look at my Annotations and Annotation Processor:
http://code.google.com/p/javadude/wiki/Annotations
In particular, the mixins example:
http://code.google.com/p/javadude/wiki/AnnotationsMixinExample
Note that it doesn't currently work if the interfaces/types being delegated to have parameterized methods (or parameterized types on the methods). I'm working on that...

It's funny to answer my own question, but here's something I found that is pretty interesting: Sather.
It's a programming language with no implementation inheritance at all! It knows interfaces (called abstract classes with no implementation or encapsulated data), and interfaces can inherit of each other (actually they even support multiple inheritance!), but a class can only implement interfaces (abstract classes, as many as it likes), it can't inherit from another class. It can however "include" another class. This is rather a delegate concept. Included classes must be instantiated in the constructor of your class and are destroyed when your class is destroyed. Unless you overwrite the methods they have, your class inherits their interface as well, but not their code. Instead methods are created that just forward calls to your method to the equally named method of the included object. The difference between included objects and just encapsulated objects is that you don't have to create the delegation forwards yourself and they don't exist as independent objects that you can pass around, they are part of your object and live and die together with your object (or more technically spoken: The memory for your object and all included ones is created with a single alloc call, same memory block, you just need to init them in your constructor call, while when using real delegates, each of these objects causes an own alloc call, has an own memory block, and lives completely independently of your object).
The language is not so beautiful, but I love the idea behind it :-)

Inheritance is not necessary for an object oriented language.
Consider Javascript, which is even more object-oriented than Java, arguably. There are no classes, just objects. Code is reused by adding existing methods to an object. A Javascript object is essentially a map of names to functions (and data), where the initial contents of the map is established by a prototype, and new entries can be added to a given instance on the fly.

You should read Design Patterns. You will find that Interfaces are critical to many types of useful Design Patterns. For example abstracting different types of network protocols will have the same interface (to the software calling it) but little code reuse because of different behaviors of each type of protocol.
For some algorithms are eye opening in showing how to put together the myriad elements of a programming to do some useful task. Design Patterns do the same for objects.Shows you how to combine objects in a way to perform a useful task.
Design Patterns by the Gang of Four