Purists will scream “C!” (or at least in my fevered imaginings they do) or “C++!”.  The draw of C is that it is a simple and powerful language; a fundamental fact about the language that you would never be able to divine if your first experience was with the heavily-corrupted, bastard stepchild of that language known as C++.  Maybe someday I will rant coherently about C++ and lay out each argument in a methodical and socially-approved way, but it’s hard to do so when you’re foaming with wrath at the very thought.  Suffice it to say that C++ obfuscates your code too much–with template classes, structures that are classes and yet not, multiple inheritance, operators that sometimes do one thing in one context and something completely different in another context…!

I’d better stop.

But  have you ever tried to read the C++ code that someone else has written?  Or even code that you wrote last year and left behind when you moved to another project?  You must comment profusely when programming in C++; ideally with five lines of comment for every line of code in order to make sense of any of it.

So, C++ is out.

C is out because I have been spoiled by the object-oriented approach to programming.  Oh, sometimes it’s nice to let go of that approach for a while and return to my roots, but ultimately I end up programming object-oriented code again because it makes logical and structural sense.

C# is out because it has the taint of Microsoft on it (now, at any rate) that I can’t quite wash away; and, anyway, the implementations that I’ve seen require you to package the runtime libraries with it or hope that the end-user has these libraries already installed.  Don’t get me wrong–the Mono project does what it does well, but it doesn’t suit my tastes in programming. I want an object-oriented approach to programming that produces consistent results for multiple architectures from simple and elegant code.

Objective C is ideal. I like the syntax, which combines the simplicity of C with the elegance of a message-based architecture. Classes are easy to implement and extend and the small trade-off in speed that results from binding method calls at run time is negligible on modern processors. Unfortunately, it is poorly documented unless you want to target MacOS exclusively. The gcc implementation is good but only on Linux platforms; if you want to target Windows you must choose either the MingW implementation, which is an older one that leaves out language features such as exceptions; or the Cygwin implementation which is newer but suffers from the limitations of Cygwin itself.

Nevertheless, the elegant simplicity of Objective C makes the limitations of Cygwin easier to accept; and the fact that it is a cross-platform language is important when you want your game to run on as many types of systems as possible.  Objective C also has the advantage of being portable to the iPhone/iPod series of devices, where the games market is only beginning to be tapped.

For these reasons, Objective C will the language of choice as I develop this game, and any code snippets that I post here will be based in that language or in C itself.

While I would like to write the next World of Warcraft, it seems best to start with something a little more realistic.  To that end, I’ve resolved to write something that I hope will be half as successful as Braid, and in the same vein; by which I mean, it will be a two-dimensional side-scrolling game with beautiful artwork and (hopefully) compelling game play.

This blog will chronicle my efforts in that direction.  While perhaps you will find something useful here, in the code or ideas that are posted, I am also posting this for myself — it is my way of keeping track of the fact that yes, I am making progress even when it doesn’t feel like it.