In the panels they said players can choose sides. So instead of stopping the siege, a player could choose to help the siege. If you extrapolate, obviously the conclusion ( if there is one ) could one of two things. So, it isn't completely linear, although at this point I think everything we are discussing is just that, ideas, which may or may not come to fruition. I doubt SoE will be able to accomplish all these things they are talking about right now. Especially, when it comes to the emergent ai with storybricks stuff and all the destructibility of the environment. There are so many issues with both, I am not sure they are going to be able to do what they are saying they will do and have it all work out in the game.
This doesn't really sound as exciting as what they indicated. These sorts of things shouldn't progress linearly from one event to the next. They should be branching events. It shouldn't be "starting a new chapter," it should be "writinga new chapter." If a server isn't able to fend off the siege, there should be consequences to that. Halas should be razed to the ground or the enemies should take over it until you clear them out.
This is sort of the same trap GW2 fell into. There were many events that would just hang around until someone finally completed them. That doesn't make the world feel alive. It makes it feel like "stages." Protecting a city while it's being built or while it is under sieged should feel like an emergency. It should feel like there are huge consequences to not fending off the enemies, which forces the server to work together. It shouldn't be, "Oh well. We can wait on that. It's not like the siege is going anywhere!" How much of an impact are we really having if there's a pre-written conclusion we are just working towards?