This is a problem that has been bothering me for a week now.
A pointcrawl map is an excelent solution to dealing with navigation in two situations. When you want to go from one known point to another known point and there are only a few possible routes that make sense, and when there is only a limited number of possible paths you can take from your current location. Which is the majority of overland travel. However, while working on the first segment of my next sandbox campaign, I noticed that it’s really difficult to use this method for searching a hidden ruin in a swamp. You don’t know the destination and there are no preexisting paths. The original pointcrawl concept explicitly mentions that real wilderness travel is almosy never blindly going in one direction but always either following a path or heading for a landmark. And I agree with that, so that I have started to believe this is not the fault of the pointcrawl but the fault of the swamp. If I want to handle travel in my campaign as a hexcrawl, then I have to change my ideas of how the players get to the ruin in the swamp.
Another issue that’s even tougher is using pointcrawling inside a huge dungeon that is mostly empty and irrelevant rooms and tunnels. A pointcrawl seems like a good idea to only fully play out the interesting sections of the whole dungeon. The chapple, the lab, the monster pens, the gatehouse, and so on. In a somewhat open castle this is no problem. Players can see the keep, the chapple, the stables, and the gatehouse from afar and chose their destination as they leave one area. But in an underground dungeon that doesn’t work. You don’t know what’s behind a door or corner until you enter the new area. It’s a completely different way people are navigating such environments.
If players are in the gatehouse and have the options to go from there to either the stables or the barracks, then the players have to know that these are the two things they can pick from. Otherwise it’s just randomness that takes them to the next area, not a meaningful choice. In a sci-fi setting this could be easy. Just have signs on the walls that tell you where you can go from here and how you get there. In a cave network or ancient ruin, that’s not a feasable approach, though.
Solutions? I don’t have any. I am still working on it, but so far I’ve made little progress beyond identifying the prolem. But my efforts will continue and maybe I’ll come up with something smart one day.
For the swamp thing: Remember the edges on the point crawl graph don’t have to be actual paths, and the nodes don’t have to be visible landmarks.
There could be thinks like: “Where the old causeway meets the river, leave the path in northwestern direction. After three miles you hit a dry knob, follow it’s southern flank for about half a mile, there you find the ruin of GacMuffins Hall.”
Have the players roll to see if they loose the direction, on the way from the causeway to the knob, due to low visibility in the thick vegetation.
For Dungeons I think the builder and inhabitants of the structure would have the same problems. So you could have markings to show the way. Now the markings don’t have to be easy to see, can be hidden, or buried.
The example you give sounds to me like being spot on to navigation by landmarks. Also, it’s an example of traveling from a known point to a known point. The diffuse idea that I had was that the players are looking for a ruin that is somewhere in the swamp without having any real idea of how to get there. Which I think is a blatantly flawed premise. Unless they did a tight raster search of the whole area it could be reasonably said that such a task is impossible. A swamp can be relatively small, but I am working on my setting as a whole continent of uncharted wilderness.
To find a specific place in the wilderness (as opposed to randomly stumbling into one), an adventure has to be designed to allow the players to navigate by landmarks and along marked paths. Otherwise it would be realistically impossible that the players will find it. Exploration can be added to this by offering the players multiple possible routes that will have them run into unexpected sites that happen to be located on the path that connects two landmarks. If they had charted their routes along other landmarks they would never have discovered those sites.
I think the solution for the dungeon problem lies in a similar direction. While you can convert a wilderness between hex map and point map back and forth, the same is not the case for point maps and grid maps in dungeons. A point map dungeon probably has to be designed with quite different structures in mind.
As you see, I already got some ideas last night that I’m experimenting with right now. I think this is crackable.