I would go 4e, if that is what I thought I’d be playing more of in the future. What I am doing to begin with, is working on a “simple mode” for the tracker where all of the data is basically volatile and isn’t saved. You just click on a “new row” icon and then start entering values. Then you can reorder however you like. Once that is ironed out, all the basic functionality is done, and I can do the “advanced mode.”

So the plan is to introduce a simplified form of the tracker with few features and then progressively add features. In fact I think I might break out MS Project and create a timeline (though all the time would be “relative” since I’m not really on a schedule) of milestone releases where requisite features are added in the appropriate order.

Thanks for the input. I too have converted some traditional philosophy to apply to software development in Occam’s Razor: Pluralitas non est ponenda sine neccesitate or Entities should not be multiplied unnecessarily. But since this case deals with my hobby and not my career, I’m not too worried about biting off more than I chew.

In the end, I think it was the right decision. Swordgleam is right that you end up having to take into account crazy rules and situations that are extremely rare. I have a lot of respect for tools like PCGen that seem to work on even the craziest character combos.

Here are some quick pointers from one tabletop gaming programmer to another. As always, take them or leave them:

1) Create “synthetic identifiers” for your DB tables. Essentially, an integer ID column for each table, and use that as your primary key. Using things like encounter name will almost always get you into trouble, since people will want to change that stuff. Completely artificial IDs are the way to go.

2) I would suggest going either/or on 3.5 vs 4e, and my gut says 4e. I think 4e will be easier to tackle since there’s less of an emphasis on continuing effects and almost everything is either “to the end of combat” or “until your next turn” So, you could cover the vast majority of effects without a lot of custom coding. 3.5 combats are complicated!

3) Mercilessly cut features. Writing something like this in your own time is very difficult, and the worst possible outcome is to get so bogged down that you never release anything. So, concentrate on the smallest possible core you can and get that out. Get some feedback from users (and yourself) and then continue from there. This is related to what I’ve termed The Ontological Argument Applied to Programming: The worst piece of working software is still better than the best piece of vaporware.

Anyways, this comment seems to have suffered from feature creep already, so I should just shut up. Good luck with your tool, and make sure it can output combat results in HTML…for easy pasting into your favorite campaign management website ;)

As someone who’s been combining my love of roleplaying with my love of PHP, let me just say: enjoy all the hours scouring through rulebooks to figure out if you need to double the amount of code just to account for one exception that only happens when a certain class mixes with a certain creature type on an odd-numbered tuesday.

Hey, we wouldn’t do it if it weren’t fun. Right? ..Right?

I’ve also thought about doing a lite version in Flash that would be cross platform, I’d just skip data storage.

Looking into the Blackberry API…

Here’s a thought – take it or junk it. Code this up for use on a blackberry or iPhone. Since I don’t game with laptops or even books at the game table, an iPhone app or something similar would be no nuisance at all… and this app would be great to have on board.

jonathan’s last blog post..Existere – A Map of RPG Blogs & Gaming Cons