OOP And AS 3.0, Something That Confuses Me ALOT!

Question

Hello everybody, there's something that's been on my mind for quiet a while now, and everytime I think about I get confused even more lol. Anyway, this might sound stupid to most of you but I donno, I just don't get it.
There are many flash developers out there who are building flash applications using OOP, some of them provide their scripts for download and that's good. The problem is that when I download those files, I have no idea how to incorporate them into my project. The files usually contain the fla, the swf, class files inside a package, and the document class. All of this looks nice but the confusing part is how to use it in my project! I have some questions and hopefully when they're answered I won't be confused anymore:
1. Suppose I downloaded two different applications, each has its own classes and its own document class. I customized them to my liking and everything and now it's time to use BOTH of them in my project. How am I supposed to use both document classes in one? Do I make my own document class, import the classes of both projects to it, then copy the codes of the document classes into mine? Or is there some other way of doing it?
2. Can I use OOP and timeline code in the same project? Like some big application could be used in the project using the OOP/Classes approach, while other simple stuff like navigation could be coded in the timeline. But usually isn't the document class a replacement for the main timeline of my flash file? Hmmm confused as you can see...
That's all I could think of for now... I would appreciate anyone that answers these questions for me because this confusion is really bothering me... Thanks in advance

Answer 1

Technical books by nature are disconnected in many ways from realities of every day tasks. It is especially true, in my view, in Flash/ActionScript. In many case authors treat Flash development the same way they would relate to an enterprise level application. Many authors come not from design but development background in other stronger languages (Java, C++, etc.) and computer science. This is, actually, great for it is about time Flash/ActionScript gets a status of a “serious” development tool.
Nevertheless, I see Flash development as a very special set of skills and, in some instances, as something that needs an artistic talent and imagination as prerequisite. In a lot of cases Java/C++, etc. developers, no matter how genius they are, just cannot hold entire picture together. Flash developer must be able to coordinate sometimes humongous amount of use cases and forks in order to implement the final vision. As a visual result oriented process, Flash development deviates toward immediate practical needs more than in the direction of satisfying abstract programming concepts and may legitimately go against conventional structure because structure is getting in the way.
One of the important aspects of products developed in Flash that is often overlooked is that they must be very lean. Even with advances in bandwidth application loading speed and financial implications are still tangible. I saw many awesome, smartly and correctly architected cool looking Flash applications that resulted in huge multi-MB blobs that run their appeal to the ground for a single reason of slow loading.
Another thing that often makes little sense is design patterns. Don’t take me wrong – I got very useful ideas how to structure my applications from design pattern concepts. But what really annoys me is when design patterns become the goal – not a tool. I see this propaganda for DP went as far as making novices in AS to attempt learning patterns and custom frameworks before getting a clear idea of how to add event listener to a button. This is totally detrimental to learning and creativity IMHO. I hope you will not fall into this pitfall. Just for the safety reasons, though, you just should learn how to say “Design patterns are invaluable” with a wide smile because odds are huge you will encounter a militant interviewer who will kill your chances because you were not fast enough to scream “MVC!” This is a trade off of dealing with unimaginative programming dictators who made career by repackaging someone else’s ideas and convincing CTOs that JSON and JQERY are the best inventions after beer.
I am available, though, to admit that this may be my combative ignorance talking and in a year I might post a Mao Zedong quotations book-like proclamation that design patterns are the only way to go. But today I rather sit in the center of Pyongyang in a silent protest against Kim Jong Il’s drinking habits rather than deal with proponents of DP.
So, don’t listen to anyone including me. Just keep in mind that some humans like to drink warm vodka.
There is another brainwash that penetrates virgin minds. People say that if you use timeline – it is not OOP. I think it is a total BS. When one writes code on timeline – he/she declares and instantiates variables and classes, uses functions, etc. – all these things ARE objects. So, why is not it OOP? Who said that only writing classes is OOP?
I am saying that even though I practically don’t use timeline coding any longer - not for political but for purely practical reasons. And yes, a newly awaken geek in me loves cleaning and refactoring my code. We all loose our purity at some point.
OOP is grand and it has proven its usefulness and dominance. But even then one encounters periodically very smartly substantiated cases against OOP in defense of procedural programming. Again, remember diluted ethanol at room temperature preferences.
In other words, as Kennedy said, it is not what I can do for encapsulation but what encapsulation can do for me.
My point is that unless one sincerely internalizes an OOP concept and see its usefulness – one cannot utilize it correctly anyway?
Again, to me encapsulation, for one, is extremely useful and I love to do things in as abstract manner as possible for a single reason – I am lazy and I love to reuse things. But I also want to be in control and I want to have a very clear idea how this staff that I reuse works.
There are applications whose code is worth incorporating into your projects and others that are worth emulating by writing your own code after just looking at what it does as a result for it is easier that repainting Sistine Chapel to the specs. More often than not, reverse engineering takes longer than writing code from scratch. With very few exceptions someone else’s code is not sufficient at best. Let alone that the code is applied to a limited number of use cases where your vision is not represented.
For instance, TeewnLite/TweenMax is a very useful package. It is tightly focused and accomplishes common and often used functionality. The best thing is that you don’t need to tweak it – it is totally cooked. It can be used in many different environments and applications. Another positive example is Papervision3D. These two things are strong and sufficient to stand on their own two feet.
But if you get an FLA with packages attached – you probably need to relate to it as a suggestion that may help you to come with architecture quicker and allow for pasting code snippets but, at the end of the day, your thing is almost always is different.
This brings us to your questions (finally).
Document class is sort of a needle that pulls all the thread into the tapestry. One of its purposes is to initialize the application and make sure that all these things that one wrote are in order. As such – it is most probably manifests application’s architecture in a very tight manner. “Tight” is a keyword for once doc class is taken out of context it renders useless.
For instance, you have three applications that you just love to the point of you dreaming about them at siesta. These programs are slide show, video player and Mona Lisa rotating in 3D space in an insane coupling with Saddam Hussein’s photo taken while he was wistfully defecating in the hot winds of Babylon oasis.
Now you decide that your video player should have slide show’s thumbnail navigation and transitions like relationships between the beauty and the dictator. Combining document classes most probably will not accomplish anything and you will have to dig deep into the code in order to fuse these three breathtaking features. Also, even if there was a way to fuse these three doc classes – because they use their own packages the size of your application will be three times bigger than it is supposed to be for all of them import their own mucus and litter memory with the staff you don’t use.
Q2.
I hope that it is clearer now that if you use Flash IDE and its compiler – you are better off with a document class. Otherwise, you will need to find another way to start. In case of other compilers you have another way to initialize.
Q3.
If you      use a movie clip only once and code for this mc pertain to this mc only – there is no practical difference between a timeline code inside this mc or      writing a class and attach it to mc. But, more often than not, you will discover that several symbols in your library share a lot of      functionality. So, if you create a class that combines these commonalities and then extend it – you will end up with a slimmer code as well as with a      smaller swf. I like writing classes also because all my code is in one place – not in two - timeline and .as files.
You ARE using OOP approach if one believes anything I say. I suggest you delegate as little as possible to doc class. And yes, you can view doc class as a rightful representative of timeline.
You know what I mean?
Now I have to find this sauce pan my cleaning lady puts into different locations every time she tidies the kitchen (I feel like a bloodhound in training) and, hopefully, eat the necessity of which (despite the offensive nature of metabolism) has transpired from all this low glucose level / manic phase delirium I typed. And NO, I don’t like warm vodka.