Love the kobolds!
@Beoric : Your scenario sound very much like gang warfare, but possibly mixed with a murder mystery? The faction time-table is a difficult thing to pre-plan IMO. I think it's the difference between "I want to publish this, so I need to communicate all the branches" vs. a home brewed campaign in which you (as DM) have time between sessions to react or are fleet-footed enough to do on the fly. It's good to have a nominal mental map of how things might play out, but after each event I think it's more exciting to have random outcomes (possibly via PC actions, possibly not) that forces you-as-NPCs to re-evaluate and plot the next move.
Am I making any sense? There's no possibility of a railroad then, just factions with motives on the way to the next station. Also, it way easier to sketch out the static (current) situation than it is to construct a hypothetical timeline. Use the DM's brain and creativity to keep things dynamic (as oppose to trying to pre-programming it (if-then-else-if...).
I get what you are saying, and that is how I would run it in a home game; the question is, is it enough for a published module? Seven years ago I thought it needed more structure, but now I would really have to think it through.
The thing is, in this imaginary universe where I actually have time to publish modules, I would probably aim them at teaching Trad and Neotrad/OC players some of our tricks. So there are parts of this that are really classic, like an old school dungeon and an old school slaver's lair. And then there is the attempted assassination mystery, which also feeds into why everyone is so paranoid (people's dreams are being manipulated by the faction that sent the assassins). Or they could play peacemaker, or they could pick a side (but only one side out of four). But really, there is too much for them to do everything; basically there was enough time to find and explore one dungeon, and then only because characters heal faster in late edition D&D.
That is a lot, and I just thought it would be too much to layer into that B2 advise that essentially says, "there are all these crazy factions, ad lib what they do". Especially at the time, after decades of 2e, 3e and 4e modules, and before some of the looser 5e WotC modules had come out, most DMs had no experience with a module that didn't spoon feed them. If the individual elements required a lot of improv, I thought I needed a mechanic to manage the framework that all these elements ran in.
So the original mechanic was basically (a) randomly determine who raids whom; (b) determine type of raid by the faction's hostility level; (c) if PCs don't prevent raid (or successfully assist it), then certain events occur and target's hostility levels are increased; (d) if PCs do prevent raid, then different events occur and different things happen with hostility levels. So for example:
(b) Posing as a merchant in the Bloody Market, a Monk will spread rumors of being cheated by a Dragonmarked House or priest of one of the gods of the Sovereign Host. If successful, increase the Traders’ hostility towards the Expatriates to 2, and adjust rumor 11 accordingly. If exposed, increase the Expatriates’ hostility towards the Monks to 2, and adjust rumor 5.
"The PCs intervene" meant I needed to have a planned encounter for the various types of raids, distinct for each attacker and defender. That added up to 36 planned encounters of this type, of which only a handful are likely to actually be intervened in by the PCs unless they are really motivated to chase down rumors.
The fun part for me was that the players can't do everything, and every element they ignore means something bad happens with the others. If you don't solve the mystery, key witnesses disappear and key villains get away. If you don't rescue the slaves they drown. If you don't clear the dungeon, it floods and is inaccessible for a long time. If you don't deal with the rising tension levels, it becomes hard to get anywhere, and you have to deal with random street fights; your haven from the dungeon is no longer a haven.
There is probably an optimum strategy on replay. Like if you can prevent the sewer workers from going on strike, or get them back to work, the sewers don't back up, the dungeon doesn't flood and the slaves don't drown, so you can save them for later. But the odds of figuring that out the first time through are pretty slim.