Related
I'm thoroughly intrigued by Scheme, and have started with some toy programming examples, and am reading through Paul Graham's On Lisp.
One thing I haven't been able to find is a book or website intended to teach Scheme to "OO people", i.e. people like myself who've done 99 % of their coding in c++/Java/Python.
I see that closures are sort of object-y, in the sense that they have local state, and offer one or more functions that have access to that state. But I don't want to learn Scheme only to port my existing habits on to it. This is why I'm learning Scheme rather than Common Lisp at the moment; I fear that CLOS might just serve as a crutch to my existing OO habits.
What would be ideal is a book or website that offers case studies of problems solved in both an OO language, and also in Scheme in a Schemey way. I suppose I would most appreciate scientific computing and/or computer graphics problems, but anything would do.
Any pedagogical leads would be much appreciated.
I doubt CLOS would serve as a crutch for old habits, I found it to be pretty different from the OO style in C++/Java/Python, and very interesting. I don't understand all the details, but I would recommend Peter Seibel's Practical Common Lisp. If you are reading On Lisp without much trouble, you should be able to dive into the chapters introducing CLOS in PCL. Also, I'd recommend his Google Tech Talk comparing Java and Common Lisp.
Here's a few more recommendations to make this a more full-fledged answer:
The classic text Structure and Interpretation of Computer Programs covers quite a few examples in chapter 3 of building modular systems using closures (and addresses issues with introducing state and mutability). Chapter 2 includes some generic and data/type-directed programming which could be helpful for motivating study of CLOS. This book really needs no introduction though, it's a towering work, and I've only been reading it slowly since the spring. Highly recommended if you are interested in Scheme.
While SICP is a great book, it's not without its flaws: A really interesting look at these is the essay "The Structure and Interpretation of the Computer Science Curriculum" which elaborates on a few criticism of SICP, and is written by the authors of How to Design Programs (I haven't read HTDP but I hear it's very good). While this essay won't teach you specifically what you are looking for - comparing functional and OO programming - it is really interesting anyway. Their freshman undergraduate course starts with a first semester introduction to functional programming using Scheme (I think, PLT/Racket) and is followed by a semester of OO programming with C++ or Java... at least that's the course they describe in the essay.
These slides from Peter Norvig address some of the design patterns common in OO programming and show why they are missing or unnecessary in dynamic, functional languages like Scheme and Lisp: http://norvig.com/design-patterns/
I cautiously recommend the book by the same authors as the Little Schemer books: A Little Java, A Few Patterns. I can't say for sure if this is a really a good book or not, it was incredibly strange and there are some really bad typesetting decisions (italic, serif, variable-width, superscript doesn't belong in a text on programming), but it might be interesting to take a look at. You can probably find it cheap, anyway. Don't take this recommendation that seriously. I think it would be better to stick to the Scheme texts.
p.s. I have to disagree with one comment stating that functional programming is not as complicated at OO programming, I think that's grossly misstating it. Functional programming in all its breadth is truly mind-boggling. When you go beyond map/filter/reduce and first-class functions, and take a look at other things in the functional realm like lazy evaluation, avoiding side effects and mutation, and the strong, static-typed languages, it gets pretty interesting, and is certainly just as complicated as traditional OO programming. I've only just scratched the surface myself but have discovered a great deal of new ideas. Programming is complicated business, whether OO or functional.
Congrat you, my friend ! Love cs, love functional programming.
If you are python developer it takes 3-4 days to think in scheme
Here is the best simple tutorial I have ever met http://www.shido.info/lisp/idx_scm_e.html
I found this course http://cs.gettysburg.edu/~tneller/cs341/scheme-intro/index.html and it may be useful for you
One beginner's resource that is very helpful and geared very much toward the casual reader is "The Adventures of a Pythonista in Schemeland". It's written (obviously) from the point of view of a Python programmer taking first steps with Scheme. One especially nice thing about it is that it includes an overview of the current implementations and compatibility issues between each scheme implementation, which, unfortunately, can cause some headaches when you're just starting out.
With regards to object systems, these two documents (linked from here) give nice examples of very simple toy implementations using closures that I found helpful in understanding their use in capturing state.
If you are starting off with Scheme, have a look at How to Design Programs. This book presents the "Schemey" approach to problem solving. I don't think there is a book that compares OO and functional solutions to the same programming problems. But there is a nice presentation that shows how dynamic languages like Scheme could provide simple solutions to problems that demand complex design patterns in statically typed OOP languages.
On another programming related website, I saw this line in someone's signature. This is NOT the first time I've seen such sentiments, although this is the harshest:
"People who work in VB or any variant
thereof are not programmers, they are
circus chimps throwing feces into an
IDE..."
VBA is my bread and butter and I can automate quite a bit of stuff with it. Yes, I know it lacks polish and some functionality, but why so much negativity toward it? On the flip side, what do other languages have that VB doesn't?
VB6, VBScript, and VBA have the reputation because they just aren't industrial strength languages. Notably:
No OOP. Sure, you have classes and modules, but no inheritance. VB isn't a low-level language, it needs real objects.
No first-class functions, so you can't even simulate OOP or polymorphism.
Lack of a well-developed class library. VB6 has a small library of built-in functions, and almost all other functionality is delegated to Windows calls or (usually pricey) third-party components.
Lousy error-handling. ON ERROR RESUME NEXT is a pox on the planet.
Although its not the fault of the language, VBA earned a bad reputation by association with MSAccess.
Of course, VB wasn't really intended to be an industrial strength language, so maybe nothing mentioned above is really proper criticism of the language at all. Fortunately VB.NET and the latest versions of VBA fix everything above, so VB.NET is on par with any other "serious" language in the marketplace.
[anecdote]
In defense of VB, I find most people criticize the language just to go along with the status quo, not because they've actually used it.
A few years ago, in a chatroom, I ran across young neophyte railing against a VB6 developer for using such a crappy language. I innocently asked "what's wrong with VB".
The first thing he said was "Because its a WINDOWS language!" So I pointed out that Borland Delphi is a Windows only language*, but I've never heard anyone malign it for that reason. (* There was a product called Kylix which cross-compiles to Linux, but its expensive, buggy, and discontinued. Its been a while since I've used Delphi, but last I'd heard, its still not ready for Linux.)
So, he said "It has a HORRIBLE SYNTAX!" Is that really the reason people hate this language? I'd say Perl, Lisp, and C++ are worse on the eyes than VB.
Next, he says "Its too easy to learn!" Well, I'd consider that a point in favor of the language. I'll never write a GUI by hand if I have a drag-and-drop designer at my disposal. What else you got?
So finally, grasping at straws, he comments "It has... no string manipulation functions". Left, Right, Mid, Replace, InStr and Trim. QED noob.
Interestingly, VB has features found some "hacker" languages, namely variant datatypes and duck typing. Compiled code performed reasonably well, interop between COM and native windows DLLs was easy, and the GUI editor basically set the bar for all future RAD development.
[/anecdote]
Read some of Joel Spolsky's articles and you'll feel better about yourself. From his article Working on CityDesk, Part Three:
Visual Basic is an extremely productive way to write code, especially GUI code. Want bold text on a dialog box? It's one click in VB. Now try doing it in MFC. You have to create a subclassed control, it's a big mess, you have to know all about LOGFONTS and Windows window subclassing and a bunch of other things and you need about three lines of code once you have the magic class.
But many VB programs are spaghetti, either because they're done as quick and dirty one-offs, or because they're written by hack programmers without training in object oriented programming, or even structured programming.
What I wondered was, what happens if you take top-notch C++ programmers who dream in pointers, and let them code in VB. What I discovered at Fog Creek was that they become super-efficient coding machines. The code looks pretty good, it's object-oriented and robust, but you don't waste time using tools that are at a level lower than you need. I've spent years writing code for C++/MFC and years writing code in Visual Basic, and let me tell you, VB is just much, much more productive.
This simplicity attracts a lot of new programmers. Saying there are a lot of bad programmers using Visual Basic does not mean Visual Basic is a bad language; it simply means that Visual Basic is accessible to bad programmers (AKA new programmers).
I work in a place where all the code is C#, not VB .NET. One developer wrote most of the code. You know how he achieved this feat? Easy: He copied-and-pasted all over the place. A given method might have anywhere from a few to hundreds of copies throughout the system.
Good developers can be good in any language. Crappy developers can be crappy in any language.
Also just to note that VB, VBA, and VB.NET are all three different languages even though they might share some similar syntax. There's no real difference between VB.NET and C# (besides the keywords/syntax), so we shouldn't lump VB (6 and before) and VBA in with VB.NET.
The real problem that many programmers have with "VB" (just say all 3 of the languages) is really more about the people using it. Most of the time "VB" programmers have less formal education and write sloppier code. That's not true for all "VB" programmers (and that doesn't mean there's not sloppy code written in C++, Java, C#, etc.). It's just the typical expectation that someone who doesn't use VB has when they hear about VB programs.
Meh, these are just religious bigots.
There is no one true language, and most experienced folks not only know that, but instantly recognize these statements as a glaring sign of inexperience.
Average developer quality seems to be inversely proportional to popularity of language * ease of use of language. VB is very easy, and is/was widely used.
This is because
A) there's a demand for coders in popular languages, so every employer has to either lower their standards, raise their pay or go without developers.
B) people without a clue can still appear moderately productive in easy to use languages. There are enough libraries and GUI tools that they can slap together something that looks useful, even if it's complete garbage under the hood.
There's nothing inherently wrong with VB when used in the domains it was intended for, by people who know what they're doing. The same is true for almost any tool/language.
I dislike the language, but that's mostly because I worked with a vb-like language which stripped out absolutely anything that might be considered an advantage and forced "best practices" that really didn't make sense.
The biggest problem I have with VB is that there is an almost direct track from clueless non-programmer -> part time Excel/Access scripter -> VBA "guru" -> VB "programmer" -> lead programmer on the most important project in the company.
Honestly I wouldn't have believed it if I didn't see someone follow that path right in front of my eyes. I even tried to mentor the guy so that he would be familiar with OOP, exception based error handling, etc. but he just dug his head in the sand and wrote everything procedurally because that had always worked for him.
I have had a chance to work with VB.Net and as long as I treated it like an object oriended .Net langauge first and VB second it wasn't so bad. It would never be my first choice for a new project, though.
I'm visiting SO in between writing VBScript code and that statement really rings true to me -- I am currently a circus chimp. If you don't know anything else, VB and its variants seem like great languages.
In my opinion, the reason for the negativity is one basic statement -- On Error Resume Next. This makes bad code a feature of the language. If it didn't have this, it wouldn't have near the bad publicity...
Most every developer I know has worked at one point or another with a VB developer, or a developer with a heavy VB background that just didn't have a clue. Unfortunately, as with most things, all we remember are the bad things about something. So we relate VB to bad programming.
It is certainly not true that all VB programmers are poor developers. But when everybody has stories about "This one old VB guy I used to work with." The stereotype is spread.
In my ongoing effort to quench my undying thirst for more programming knowledge I have come up with the idea of attempting to write a (at least for now) simple programming language that compiles into bytecode. The problem is I don't know the first thing about language design. Does anyone have any advice on a methodology to build a parser and what the basic features every language should have? What reading would you recommend for language design? How high level should I be shooting for? Is it unrealistic to hope to be able to include a feature to allow one to inline bytecode in a way similar to gcc allowing inline assembler? Seeing I primarily code in C and Java which would be better for compiler writing?
There are so many ways...
You could look into stack languages and Forth. It's not very useful when it comes to designing other languages, but it's something that can be done very quickly.
You could look into functional languages. Most of them are based on a few simple concepts, and have simple parsing. And, yet, they are very powerful.
And, then, the traditional languages. They are the hardest. You'll need to learn about lexical analysers, parsers, LALR grammars, LL grammars, EBNF and regular languages just to get past the parsing.
Targeting a bytecode is not just a good idea – doing otherwise is just insane, and mostly useless, in a learning exercise.
Do yourself a favour, and look up books and tutorials about compilers.
Either C or Java will do. Java probably has an advantage, as object orientation is a good match for this type of task. My personal recommendation is Scala. It's a good language to do this type of thing, and it will teach you interesting things about language design along the way.
You might want to read a book on compilers first.
For really understanding what's going on, you'll likely want to write your code in C.
Java wouldn't be a bad choice if you wanted to write an interpreted language, such as Jython. But since it sounds like you want to compile down to machine code, it might be easier in C.
I recommend reading the following books:
ANTLR
Language Design Patterns
This will give you tools and techniques for creating parsers, lexers, and compilers for custom languages.
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
I have been asked to begin teaching C# and OO concepts to a group of procedural programmers. I've searched for ideas on where to begin, but am looking for general consensus on topics to lead with in addition to topics to initially avoid.
Edit
I intend to present information in 30 minute installments weekly until it no longer makes sense to meet. These presentations are targeted at coworkers at a variety of skill levels from novice to expert.
The best thing you can do is: Have a ton of Q&A.
Wikipedia's procedural programming (PP) article really hits where you should start:
Whereas procedural programming uses
procedures to operate on data
structures, object-oriented
programming bundles the two together
so an "object" operates on its "own"
data structure.
Once this is understood, I think a lot will fall into place.
In general
OOP is one of those things that can take time to "get," and each person takes their own path to get there. When writing in C#, it's not like the code screams, "I am using OO principles!" in every line. It's more of a subtle thing, like a foreach loop, or string concatenation.
Design center
Always use something (repeatedly) before making it.
First, use an object, and demonstrate the basic differences from PP. Like:
static void Main(string[] args)
{
List<int> myList = new List<int>();
myList.Add(1);
myList.Add(7);
myList.Add(5);
myList.Sort();
for (int i = 0; i < myList.Count; i++)
{
Console.WriteLine(myList[i]);
}
}
Using objects (and other OO things) first -- before being forced to create their own -- leads people down the path of, "Ok, I'm making something like what I just used," rather than "WTF am I typing?"
Inheritance (it's a trap!)
I would NOT spend a lot of time on inheritance. I think it is a common pitfall for lessons to make a big deal about this (usually making a cliché animal hierarchy, as others pointed out). I think it's critical to know about inheritance, to understand how to use the .NET Framework, but its nuances aren't that big of a deal.
When I'm using .NET, I'm more likely to "run into inheritance" when I'm using the .NET Framework (i.e. "Does this control have a Content property?" or "I'll just call its ToString() method.") rather than when I'm creating my own class. Very (very (very)) rarely do I feel the need to make something mimicking the taxonomy structure of the animal kingdom.
Interfaces
Coding to an interface is a key mid-level concept. It's used everywhere, and OOP makes it easier. Examples of this are limitless. Building off the example I have above, one could demonstrate the IComparer<int> interface:
public int Compare(int x, int y)
{
return y.CompareTo(x);
}
Then, use it to change the sort order of the list, via myList.Sort(this). (After talking about this, of course.)
Best practices
Since there are some experienced developers in the group, one strategy in the mid-level classes would be to show how various best practices work in C#. Like, information hiding, the observer pattern, etc.
Have a ton of Q&A
Again, everyone learns slightly differently. I think the best thing you can do is have a ton of Q&A and encourage others in the group to have a discussion. People generally learn more when they're involved, and you have a good situation where that should be easier.
The leap from procedural to object oriented (even within a language - for four months I programmed procedural C++, and classes were uncomfortable for a while after) can be eased if you emphasize the more basic concepts that people don't emphasize.
For instance, when I first learned OOP, none of the books emphasized that each object has its own set of data members. I was trying to write classes for input validation and the like, not understanding that classes were to operate on data members, not input.
Get started with data structures right away. They make the OOP paradigm seem useful. People teach you how to make a "House" class, but since most beginning programmers want to do something useful right away, that seems like a useless detour.
Avoid polymorphism right away. Inheritance is alright, but teach when it is appropriate (instead of just adding to your base class).
Operator overloading is not essential when you are first learning, and the special ctors (default, dtor, copy ctor, and assignment operator all have their tricky aspects, and you might want to avoid that until they are grounded in basic class design).
Have them build a Stack or a Linked List. Don't do anything where traversal is tricky, like a binary tree.
Do it in stages.
High level concepts : Describe what an object is and relate it to real life.
Medium level concepts: Now that they got what object is, try compare and contrast. Show them why global variable is bad compared to an encapsulated value in a class. What advantage they might get from encapsulating. Start introducing the tennets of OOP (encapsulation, inheritance)
Low Level concepts: Go in further into polymorphism and abstraction. Show them how they can gain even better design through polymorphism and abstraction.
Advance concepts: SOLID, Interface programming, OO design patterns.
Perhaps you should consider a problem that is work related and start with a procedural implementation of it and then work through (session by session) how to make an OOP implementation of it. I find professionals often grasp concepts better if it is directly related to real examples from their own work place. The junk examples most textbooks use are often horrible for understanding because they leave the student wondering, why on earth would I ever want to do that. Give them a real life reason why they would want to do that and it makes more sense.
I would avoid the "a bicycle is a kind of veichle" approach and try to apply OO to an environment that is fairly specific and that they are already used to. Try to find a domain of problems that they all recognize.
Excercise the basics in that domain, but try to move towards some "wow!" or "aha!" experience relatively early; I had an experience like that while reading about "Replace Conditional with Polymorphism" in Fowlers Refactoring, that or similar books could be a good source of ideas. If I recall correctly, Michael Feathers Working effectively with legacy code contains a chapter about how to transform a procedural program into OO.
Teach Refactoring
Teach the basics, the bare minimum of OO principles, then teach Refactoring hands-on.
Traditional Way: Abstractions > Jargon Cloud > Trivial Implementation > Practical Use
(Can you spot the disconnect here? One of these transitions is harder than the others.)
In my experience most traditional education does not do a good job in getting programmers to actually grok OO principles. Instead they learn a bit of the syntax, some jargon they have a vague understanding of, and a couple canonical design examples that serve as templates for a lot of what they do. This is light years from the sort of thorough understanding of OO design and engineering one would desire competent students to obtain. The result tends to be that code gets broken down into large chunks in what might best be described as object-libraries, and the code is nominally attached to objects and classes but is very, very far from optimal. It's exceedingly common, for example, to see several hundred line methods, which is not very OO at all.
Provide Contrast To Sharpen The Focus on the Value of OO
Teach students by giving them the tools up front to improve the OO design of existing code, through refactoring. Take a big swath of procedural code, use extract method a bunch of times using meaningful method names, determine groups of methods that share a commonality and port them off to their own class. Replace switch/cases with polymorphism. Etc. The advantages of this are many. It gives students experience in reading and working with existing code, a key skill. It gives a more thorough understanding of the details and advantages of OO design. It's difficult to appreciate the merits of a particular OO design pattern in vacuo, but comparing it to a more procedural style or a clumsier OO design puts those merits in sharp contrast.
Build Knowledge Through Mental Models and Expressive Terminology
The language and terminology of refactoring help students in understanding OO design, how to judge the quality of OO designs and implementations through the idea of code smells. It also provides students a framework with which to discuss OO concepts with their peers. Without the models and terminology of, say, an automobile transmission, mechanics would have a difficult time communicating with each other and understanding automobiles. The same applies to OO design and software engineering. Refactoring provides abundant terminology and mental models (design patterns, code smells and corresponding favored specific refactorings, etc.) for the components and techniques of software engineering.
Build an Ethic of Craftsmanship
By teaching students that design is not set in stone you bolster students' confidence in their ability to experiment, learn, and discover. By getting their hands dirty they'll feel more empowered in tackling software engineering problems. This confidence and practical skill will allow them to truly own the design of their work (because they will always have the skills and experience to change that design, if they desire). This ownership will hopefully help foster a sense of responsibility, pride, and craftsmanship.
First, pick a language like C# or Java and have plenty of samples to demonstrate. Always show them the big picture or the big idea before getting into the finer details of OO concepts like abstraction or encapsulation. Be prepared to answer a lot of why questions with sufficient real world examples.
I'm kinda surprised there's any pure procedural programmers left ;-)
But, as someone who started coding back in the early 80s on procedural languages such as COBOL, C and FORTRAN, I remember the thing I had most difficulty with was instantiation. The concept of an object itself wasn't that hard as basically they are 'structures with attached methods' (looked at from a procedural perspective) but handling how and when I instantiated an object - and in those days without garbage collection - destroyed them caused me some trouble.
I think this arises because in some sense a procedural programmer can generally point to any variable in his code any say that's where that item of data is directly stored, whereas as soon as you instantiated an object and assign values to that then it's much less directly tangible (using pointers and memory allocation in C is of course similar, which may be a useful starting point also if your students have C experience). In essence I suppose it means that your procedural -> OOPS programmer has to learn to handle another level of abstraction in their code, and getting comfortable with this mental step is more difficult than it appears. By extension I'd therefore make sure that your students are completely comfortable with allocating and handling objects before looking at such potentially confusing concepts as static methods.
I'd recommend taking a look at Head First Design Patterns which has really nice and easy to understand examples of object oriented design which should really help. I wouldn't emphasize the 'patterns' aspect too much at this point though.
I'm a vb.net intermediate programmer, and I'm learning OOP. One of the things I find is the lecturing about the concepts over and over is unnerving. I think what would be perfect documentation would be a gradual transition from procedural programming to full blown OOP rather than trying to force them to understand the concepts then have them write exclusively OOP code using all the concepts. That way they can tinker with little projects like "hello world" without the intimidation of design.
For example (this is for VB.NET beginners not advanced procedural programmers).
I think the first chapters should always be about the general concepts, with just a few examples, but you should not force them to code strictly OOP right away, get them used to the language, so that it's natural for them. When I first started, I had to go back and read the manual over and over to remember HOW to write the code, but I had to wade through pages and pages of lecturing about concepts. Painful!
I just need to remember how to create a ReadOnly Property, or something. What would be real handy would be a section of the book that is a language reference so you can easily look in there to find out HOW to write the code.
Then you briefly explaining how forms, and all the objects are already objects, that have methods, and show how they behave, and example code.
Then show them how to create a class, and have them create a class that has properties, and methods, and the new construct. Then have them basically switch from them using procedural code in the form or modules, to writing methods for classes.
Then you just introduce more advance codes as you would any programming language.
Show them how inheritance works, etc. Just keep expanding, and let them use thier creativity to discover what can be done.
After they get used to writing and using classes, then show how thier classes could improve, introducing the concepts one by one in the code, modifying the existing projects and making them better. One good idea is to take an example project in procedural code, and transform it into a better application in OOP showing them all the limitations of OOP.
Now after that is the advanced part where you get into some really advanced OOP concepts, so that folks who are familar with OOP already get some value out of the book.
Define an object first, not using some silly animal, shape, vehicle example, but with something they already know. The C stdio library and the FILE structure. It's used as an opaque data structure with defined functions. Map that from a procedural use to an OO usage and go from there to encapsulation, polymorphism, etc.
If they are good procedural programmers and know what a structure and a pointer to a function are, the hardest part of the job is already done!
I think a low level lecture about how Object Oriented Programming can be implemented in procedural languages, or even assembler, could be cool. Then they will appreciate the amount of work that the compiler does for them; and maybe they will find coding patterns that they already knew and have used previously.
Then, you can talk about best practices in good Object Oriented design and introduce a bit of UML.
And a very important thing to keep in mind always is that they're not freshmen, don't spend much time with basic things because they'll get bored.
Show Design Patterns in Examples
There where some plenty good answers, alright. I also think, that you should use good languages, good, skillful examples, but I have an additional suggestion:
I have learned what OOP means, by studying Design Patterns. Of course, I have of course learned an OO-language before, but until I was working on Design Patterns, I did not understand the power of it all.
I also learned much from OO-Gurus like Robert C. Martin and his really great papers (to be found on his companies site).
Edit: I also advocate the use of UML (class diagrams) for teaching OO/Design-Pattern.
The thing that made it click for me was introducing Refactoring and Unit Testing. Most of my professional programming career has been in OO Languages, but I spent most of it writing procedural code. You call a function on an instance of class X, and it called a different method on an instance of class Y. I didn't see what the big deal about interfaces was, and thought that inheritance was simply a concept of convenience, and classes were by and large a way of helping us sort and categorize the massive code. If one was masochistic enough, they could have easily go through some of my old projects and inline everything until you get to one massive class. I'm still acutely embarrassed at how bad my code was, how naive my architecture was.
It half-clicked when we went through Martin Fowler's Refactoring book, and then fully clicked when started going through and writing Unit and Fitnesse tests for our code, forcing us to refactor. Start pushing refactoring, dependency injection, and separation of the code into distinct MVC models. Either it will sink in, or their heads will explode.
If someone truly doesn't get it, maybe they aren't cut out for working on OO, but I don't think anyone from our team got completely lost, so hopefully you'll have the same luck.
I'm an OO developer professionally, but have had had procedural developers on my development team (they were developing Matlab code, so it worked). One of the concepts that I like in OO programming is how objects can relate to your domain (http://en.wikipedia.org/wiki/Domain-driven_design - Eric Evans wrote a book on this, but it is not a beginner's book by any stretch).
With that said, I would start with showing OO concepts at a high level. Try to have them design a car for example. Most people would say a car has a body, engine, wheels, etc. Explain how those can relate to real world objects.
Once they seem to grasp that high level concept, then I would start in on the actual code part of it and concepts like inheritance vs aggregation, polymorphism, etc.
I learned about OOP during my post-secondary education. They did a fairly good job of explaining the concepts, but completely failed in explaining why and when. They way they taught OOP was that absolutely everything had to be an object and procedural programming was evil for some reason. The examples they were giving us seemed overkill to me, partly because objects didn't seem like the right solution to every problem, and partly because it seemed like a lot of unnecessary overhead. It made me despise OOP.
In the years since then, I've grown to like OOP in situations where it makes sense to me. The best example I can think of this is the most recent web app I wrote. Initially it ran off a single database of its own, but during development I decided to have it hook into another database to import information about new users so that I could have the application set them up automatically (enter employee ID, retrieves name and department). Each database had a collection of functions that retrieved data, and they depended on a database connection. Also, I wanted an obvious distinction which database a function belonged to. To me, it made sense to create an object for each database. The constructors did the preliminary work of setting up the connections.
Within each object, things are pretty much procedural. For example, each class has a function called getEmployeeName() which returns a string. At this point I don't see a need to create an Employee object and retrieve the name as a property. An object might make more sense if I needed to retrieve several pieces of data about an employee, but for the small amount of stuff I needed it didn't seem worth it.
Cost. Explain how when properly used the features of the language should allow software to be written and maintained for a lower cost. (e.g. Java's Foo.getBar() instead of the foo->bar so often seen in C/C++ code).Otherwise why are we doing it?
I found the book Concepts, Techniques, and Models of Computer Programming to be very helpful in understanding and giving me a vocabulary to discuss the differences in language paradigms. The book doesn't really cover Java or C# as 00-languages, but rather the concepts of different paradigms. If i was teaching OO i would start by showing the differences in the paradigms, then slowly the differences in the 00-languages, the practical stuff they can pickup by themselves doing coursework/projects.
When I moved from procedural to object oriented, the first thing I did was get familiarized with static scope.
Java is a good language to start doing OO in because it attempts to stay true to all the different OO paradigms.
A procedural programmer will look for things like program entry and exit points and once they can conceptualize that static scope on a throwaway class is the most familiar thing to them, the knowledge will blossom out from there.
I remember the lightbulb moment quite vividly. Help them understand the key terms abstract, instance, static, methods and you're probably going to give them the tools to learn better moving forward.
I'm looking for some good courses on object oriented programming. I've been programming for about 4 years so far, but I don't feel like I have a SOLID grasp on OO.
How did you learn?
Its all learning by doing. Courses and books are good, but the bottom line: Coding is all about writing programs; start from there. Another thing I personally think is very important: read other peoples code!
Try these: http://webcast.berkeley.edu/course_details_new.php?seriesid=2008-D-26263&semesterid=2008-D
Maybe you'll like the "Head First" approach.
The mechanics of OO (i.e. the syntax and semantics, terminology, what it is, what it does) are very easy. They "why" and "where" of it is the hard part. I'd suggest you just learn the mechanics quick and dirty on Wikipedia if you haven't already, and then pick up a book on design patterns. Head First is a good choice here. I don't believe that you can truly understand OOP (or really anything for that matter) unless you know where it's useful and what problems it's meant to solve.
After this, the next step would be to look at APIs you use every day in your favorite language from your new perspective. You should start seeing why using OO techniques in certain situations makes the API much more usable than if everything were just free/static functions and plain old data objects. You might also see areas where OOP is awkward and overused and free/static functions and plain old data objects might be better solutions.
Lastly, it's useful to know how some aspects of OO are implemented at a lower level. Understanding things like virtual function tables and how to do OOP in pure C really adds a level of depth to your understanding of what can and cannot be accomplished with it, and what the tradeoffs are inherent in it.
And obviously, try to integrate what you learn into the code you write.
Since you ask :) I learned with Object Oriented Software Construction, 2nd Edition by Bertrand Meyer, Prentice Hall, 1997.
Meyer is due to publish in a very short while a book which is focused on learning to program from the start: Touch of Class. This should be a good option for students.
I notice that people's tastes about learning materials vary quite a lot. Some like summaries, tutorials, how-to's... I learn complex subjects by taking my time to follow the complete dissertations.
A pure-object language is the best course in OO.
I can suggest Java, Ruby, or Smalltalk
Learn any of those (from scratch, not just as rewriting C or Perl) and you know OO.
I would go with ruby, it's more mainstream than smalltalk, and has a richer syntax than java. Perfect for the beginner
It is REALLY difficult to learn this by yourself. The best chance is to learn it alongside someone else at your workspace, who's going to get you through tasks and shows you things on the way.
I can recommend this book. http://squeakbyexample.org/ it's far from perfect, but it's free and comes with a special squeak version that is optimized so you find the code fast. The examples are kind of nice for starters, I think. At least I enjoyed reading it.