?

Log in

No account? Create an account
2006-1: The Real Battle - The Cover Story
October 2013
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
 
 
Wed, Jan. 18th, 2006 09:32 am
2006-1: The Real Battle

Last Friday the first turn for my 2006-1 BSG was due. Since then I've spent about 24 hours modifying the fixing the command interpretter I built for it. To call this game a hurricane of information is to set a new bar in understatement. More importantly, as I'm finding, the game has an outstanding propensity for causing emergent behavior between its objects that I never saw coming (although in retrospect was obvious). The fact that FanPro released the rules set incomplete is actually helping me more than its hurting -- if its not specified, I can interpet it as I see fit, usually the way thats easiest to code. Unfortunately, the cascade of numbers and the lack of time to build a good GM interface means that what should be blatantly obvious errors are whooshing by me as I'm missing the forest for the collection of oddly colored branches an awful lot. I've already run one turn deadline, and I'm starting to think I'm in immeinent danger of running another if I don't make some breakthrough.

A coding project, at least in my experience, has consumed you when your application seems to take on a personality of its own. I know its human nature to anthropomorphize almost everything we interact with, but even so, as I add debugging output and error checks, and can't happen checks, and WTF checks, it seems as if the program is helping me fix itself, like a patient that is slowly learning medicine as its treated. Alternatively, I'm not getting enough creative time and my subconscious is rebelling.

Now, to figure out why the hell my Air and Ground ratings are getting swapped...

Tags:
Current Mood: busy
Current Music: Ben Newman - Beyond the Far Horizon

1CommentReplyShare

teenyweenyowen
Owen
Thu, Jan. 19th, 2006 07:38 pm (UTC)

Isn't the point of having debugging code mixed in with logic code? It's a bit more of a pain to write but it lets the code tell you when something goes wrong fast. Both in terms of time from mistake to realizing there's a mistake and line/call stack distance between where the mistake occurs and where it's effects are detected.


ReplyThread