I'd like to start this by discussing quests, as I am currently encountering them as a design element in Dungeons & Dragons, as I prepare the Madness at Gardmore Abbey super-adventure for my 4th edition campaign. Madness at Gardmore Abbey is a very big adventure which can take more than a dozen sessions to complete, so with my group's "twice a month" playing frequency this could occupy us for more than half a year. There are 33 planned encounters in the adventure, not counting the roleplaying encounters the possibilities of which are described in a whole book full of NPCs, their goals and possible actions. Now Gardmore Abbey is very much a "sandbox" style of adventure. There are two dungeons, but they are relatively small, and most of the encounters are above ground. And there is very little limitation to the access of these encounters, no "rails" which would force any sort of linear structure. Right from the start the group has at least 4 different options on where to enter the abbey, leading to different encounters, and that isn't counting variants or all the options of climbing over the wall and pretty much just starting anywhere you like.

In terms of MMORPGs this design would be very much on the side of a "world" design, and not so much on the side of a "game" design. You have a world that you can explore relatively freely, and in any order you like. As this is D&D and the adventure goes from level 6 to 8 and not like a full MMORPG world from level 1 to 60 or more, you don't even have the problem of somebody accidentally running into an area that would be much too high in level for him, although there are obviously easier and more difficult ways to get through this adventure.

So as a DM, who plays some sort of "game designer" role in a D&D adventure, what are the challenges of such a sandbox / open world design? Would it be a good idea to just let the players loose on Gardmore Abbey and let them take some random path through the adventure?

Having been DM for over 30 years now, I am very well aware of the risk such an adventure design can lead to: The abandonment of story in favor of a hack'n'slash dungeon crawler attitude by the players. It is the classic dilemma often observed in the mega-dungeons of old, where the players come to a 4-way crossing in a dungeon and have to decide which way to go. Not having any sensible information about which way to go, a group will just choose a random path, and just try to clear out all rooms of the dungeon more or less systematically in order to get a maximum of xp and treasure and not to miss anything. In MMORPGs there is an additional issue of respawns, leading to the "Evercamp" situation where players aren't moving at all any more, because they get more xp and treasure by remaining static and waiting for the respawn than by moving around.

While it is perfectly possible to play Madness at Gardmore Abbey like this and just take some random path through the abbey until you killed all the monsters and gotten all the treasure, this is hardly ideal. Gardmore Abbey is a fantasy ecosystem on its own, with a rich history, and many interwoven stories. There is a *reason* why there are certain monsters in certain locations. And while you *can* ignore these reasons and just hack'n'slash your way through the place, it is obvious that the experience would become much better if you followed the various story-lines and went to locations with a purpose. And this is where quests come in.

Now the word "quest" has a lot of different meanings in different contexts. A medieval knight of literature on a quest for the holy grail would expect that quest to last a life-time, with good chances to never actually achieve his quest-goal. That is a far way from doing 20 daily quests, each of which is more of a minor chore like "kill 10 of these" or "gather 8 of that". Classic D&D always had quests, but didn't even call them like that. But if the wizard in the tavern gave the heroes the task to find the Lost Amulet of Chorr in the caves of the troglodytes, that was very much "the quest" of the adventure. 4th edition D&D formalized that, and added specific quest rewards, and the possibility to have both major and minor quests ongoing simultaneously in an adventure. So the Madness at Gardmore Abbey has several chapters full of quests that the players can receive from different quest-givers.

Now as anyone having played a modern MMORPG can attest, having too many quests can lead to the other extreme, where again story is abandoned and the mode of the game is more hack'n'slash. If you have a "quest hub" where you can get a dozen quests which together basically tell you to kill a certain number of every species of wildlife in the area, you'll just go on a killing spree and do a tour of the area until all the quest trackers are fulfilled. So the trick the Madness at Gardmore Abbey adventure uses (and which can be further refined by a good DM) is to feed the adventurers not more than one or two quests at the same time. That does lead to the somewhat annoying "back to the quest-giver NPC for the next quest" issue, but fortunately travel time in a pen & paper roleplaying game isn't much of an issue, the players can just say "we go there", and the DM can fast forward time.

Quests turn the Madness at Gardmore Abbey from a dungeon crawl into a series of interwoven stories, giving the players a purpose to go to specific locations with specific goals. Even if at the end of the adventure the players have gone to all the locations, killed all the monsters, and gotten all of the treasure, the experience will have been a lot more memorable than if they had just achieved the same with a systematic killing spree without purpose. Chances are that the players will forget the monsters and treasures soon, but will remember some of the stories that led them there. It is the quests that give the opportunity to not just see the D&D adventure as a series of combats, because now there is a motivation to roleplay with the NPCs to learn more of the story or to achieve goals in other ways than by fighting.

There is a design lesson to be learned here for online virtual worlds: Quests do have value in providing purpose to the action of players, but they need to be handed out a lot more sparingly than current games do. If a quest was something you only had one of, and it would take you days to achieve that quest's goal, it would be a lot more memorable than a shopping list of minor chores from the next quest hub.