Chapter 15 - The Final Chapter
This is it. The last one. I don’t know what’s going to be in it. It’s supposed to give the reader a “where to go next” kind of outlook. I hope that one of the places will be here, but obviously there’s tons of places to go and tons of things for cocos2d developers to learn.
If you have a particular idea what should be included in the greater cocos2d developer outlook kind of sense, please let me know!
Summary of working on Chapter 14 - Game Center
Amazing! Simply amazing. I think I fell in love with Game Center in the process, I haven’t worked with a networking API that’s so smooth, straightforward and easy to comprehend. No thanks to Apple’s already excellent step-by-step documentation. Still, there were a couple pitfalls and things that one could forget, and I did, that I’ve obviously included in the chapter.
On the other side, with almost a week late for this chapter, I realized how time consuming network programming is. There’s a lot of testing going on, and especially if you’re testing on two devices the process is incredibly inefficient compared to a simple single-player game. You always have to deploy to two devices for every test, and every time the delay between request and response adds to the time spent on testing. Add to that common network errors such as a drop in connection or something blocking the line with a download, and you’re up for a fun ride.
But that alone didn’t account for the one week delay. In the past two weeks I’ve been helping to renovate someone’s apartment, I got a cold, I held a presentation at the Macoun Mac OS X conference (about cocos2d obviously) while still recovering from the cold. Then our pet cat Yoshi had to see the vet and almost died during anesthesia because of a pulmonary edema. On the brighter side I’ve also attended a wedding and wrote a GameKitHelper class for this chapter which contains more stuff than I could describe in the book, including storing achievements which failed transmission, as is recommended but not implemented by Apple’s Game Center Programming Guide.
That’s also why I haven’t been answering emails timely recently. Please be patient, I’ll get to yours soon!
What’s left?
Now that I’m almost finished writing the book, what’s left? Obviously I have to review technical and other edits done by Apress. Current chapter 13 is in technical review while chapter 6 is in editorial review, after which it’s ready for production. So I’ll still be busy reviewing and making changes and additions to chapters during October and probably even November.
Then there’s the issue of the example projects having used three different versions of cocos2d, starting with 0.99.3. That was actually an oversight on my part because at the time I had 0.99.4 available. Luckily those changes are really insignificant for the first few projects. The DoodleDrop game already uses 0.99.4 and so does most of the book’s code. But for the Game Center chapter, I had to migrate the Tilemap project from 0.99.5 beta 1 to beta 3 and that was a huge step. It was easier to simply create a new project from the latest cocos2d template, then re-adding all game source code and resources to the new project. Still, that’s doable.
The bigger issue I have here is the fact that I can’t change anything in the book anymore, so the code should reflect what’s in the book. If the book mentions CCLabel then the code should use CCLabel and not CCLabelTTF. What I think is probably going to be a good compromise is to update the important (final) versions of each chapter’s example projects to cocos2d v1.0 once that is released. Obviously the code supplied with the book will remain as it is described in the book, so the upgraded code would be for reference only and a separate download. Whether I wait for v1.0 depends on how progress towards v1.0 is coming along around the book’s release date some time in December 2010.
In hindsight, I really wished I had used my Xcode template project and used that throughout the book. Back then I decided against it because it was important for me to write the code like almost all cocos2d developers would do. Now I regret the decision because I could have changed the way cocos2d developers start new cocos2d projects for the better. The whole updating process for cocos2d is a major PITA and then some, so I think I need to bring this issue to the table more frequently, more actively. At least until the cocos2d template installation procedure is changed to not copy all of the source code into each new project and then leaving it up to the developer to deal with upgrades.
I’m back from Macoun, and it was nice meeting you all! Well, actually since that conference was just a 90 minute ride away, so I wasn’t really gone, except mentally. I spent a lot of time last week to prepare my presentation (in german), and fighting a cold. The subject was a game component system that I wrote on top of cocos2d, to encapsulate most of the cocos2d CCNode stuff and allowing me to focus on writing re-usable gameplay components. The system works but it’s not production tested. Still it’s proven very promising from my initial tests and opens some interesting opportunities, for example the re-use of components is a huge timesaver, and you can totally get rid of any class hierarchy and remain flexible throughout development. Due to writing the book and other things I hadn’t had the time to develop it as much as I wanted to. I’ll release it eventually but for the moment, I have a different focus.
For one, I’m late with the Game Center chapter due to the Macoun conference preparations, an upcoming wedding and an apartment renovation of the soon-to-be-wed as their wedding gift. So I’ll try to cram writing both the Game Center chapter and the final chapter in this week, before turning my attention to updating the Line-Drawing Starterkit and Xcode project. I did not intend to support non-beta versions of cocos2d, but given how long the recent betas last and how many developers actually use the latest beta (probably due to HD support) and how significant the breaking changes are this time, I’m going to update the starterkit and Xcode project with HD support while keeping the old project for those who would rather use the stable 0.99.4 version of cocos2d. Once that’s done, it’ll be mid of October and another project, only partly concerning cocos2d, kicks in full speed. No time to lose.
Speaking of the Starterkit, September went by and I made 10 sales, without advertising it or anything, so those numbers are at the lower end of the “potential sales” spectrum. That’s a sum (over $1,500) that I can live with. Well, actually, that’s a sum that I can live off of. On a related matter, I recently found a page detailling the traffic stats of the cocos2d-iphone.org website. If those stats are correct, the unique visits are over ten times that of my site. Now, frankly speaking, every time I get those little facts and stats from here and there over the past couple months, and then added up the numbers, compared them with mine, applied reasonably pessimistic estimations and caution, I do wonder: why the hell isn’t cocos2d run like a business?
If you ask me, with those traffic stats and a reasonable conversion rate of 0.1% per unique visitor (mine is over 0.2%), one could easily pay 3 people to develop cocos2d, test it, write documentation, moderate the forum, and in general adding more business value. Which in turn grows the business, speeds up the development of the engine, tools and by-products and will make everyone benefit from that. I don’t get it. Because at this point, it’s either going to be that, or a slow decline to a niche product over the next couple years due to the increasingly strong competition from other iOS engines. Especially those that offer cross-platform support, since that’s what contractors are asking for, or even demanding, more and more.
Anyhow, for my part, I decided that now would be a good time to start doing the Indie thing full-time, still accepting contract work but I’m able to be a lot pickier about it. I have a goal set out for myself, and it’s not a simple one. I’ve always enjoyed most to help my colleagues, to fix their problems, to support them and in general, to help them achieve excellence. I know, that sounds like something you’d find in EA’s job matrix - and in fact, you do. Helping others achieve excellence, throughout my professional career, that’s what I’ve been doing and enjoying the most. Now I find myself doing that for cocos2d developers, enjoying it and being able to support myself in the process. Well, I think I’ll have to thank you, my dear readers, for that. 
As a thank you back, I’ve started working on a bigger project about 3 months ago that will be useful and helpful to many indie game developers, regardless of the engine you may be using. Hence it deserves its own website and a cool name (darn, that is hard!). I’m looking forward to really getting into high gear with it after I’ve submitted the remaining book chapters. Stay tuned.
