[fangthane]

I think that by far the most daunting thing about a project like this is the amount of balance some elements will need to receive, and the decision as to how far to take the game, relative to the webcomic.

Arkentools - leave them in, or take them out as overpowered destroyers of balance?
Tri-mancer linkups - what each might be capable of accomplishing is in question, as well as what each should be capable of doing vis-a-vis balance.
Unknown content - We know there's dance-fighting and we know a little about magic; there's more about magic that we don't know than what we do, and we're blissfully unaware of whether there're other forms of pseudo-combat (dozens, stand-up vs hecklers, freestyle rap, musical showdowns, etc)

Also, I should point out that for most of us, it's a serious issue to gather sufficient gamer-geeks for D&D, let alone to represent every Person of Interest in an Erf-style wargame. If you're going to make something anyone will ever play, it has to allow for one person to control a side wholesale.

Spoiler

I do have an idea as to how loyalty could be handled under these circumstances and remain a mystery, but it's a bit complex and involves 'loyalty challenges' - face-down loyalty cards, die rolls for the check and a new loyalty card after each check, and 'permanent' effects to add + or - loyalty modifiers to friends/enemies/torture victims. For example, pay the bazillion schmuckers for an iron maiden, rack and everything, and you get a better chance to affect your prisoner's loyalty (with a certain number of successful checks needed to fully turn a character)