When I began my Dwimmermount campaign nearly a year ago, I intended it to be an experiment of sorts, an attempt to experience for myself what it might have been like to play OD&D in the early days of the hobby. Of course, I couldn't truly recreate that experience for myself, as I entered the hobby post-LBBs and I brought to the game lots of assumptions and knowledge born out of the hobby as I experienced it. Moreover, as I know well, there's no single "early days of the hobby." Rather, there's a morass of such things, many of them complementary and compatible and some of them not. Thus, my experience would, at best, give me only a particular experience of those early days and not necessarily a particularly authentic one.
Still, I thought then -- and do now -- that it was a worthwhile endeavor in its own right. As someone who entered the hobby after those glory days and whose tastes now are more in tune with them, I wanted to give myself every opportunity to appreciate the unique virtues of those times in the only way I could. I'd say that, by and large, my Dwimmermount campaign has succeeded in its original purpose. After 21 sessions (I'll post a session summary of the most recent one soon), it shows no signs of stopping and everyone is enjoying themselves, which is the only real measure of a campaign's worth. In addition, I've learned quite a lot about what works and what doesn't, both within OD&D specifically and old school games generally. And I've acquired a number of new methods and approaches that I somehow never learned in my formative years as a referee.
The process of "unlearning" is sometimes a difficult one, particularly when it comes to world-building. I am an inveterate world-builder. I love to create new worlds and am always turning over ideas for such things in my head. It's a great blessing in many ways but it's also a curse, particularly when running a campaign, like Dwimmermount, where I'm philosophically committed to letting things evolve naturally and not according to some master plan beforehand. Consequently, the setting of the campaign is still quite limited and fuzzy beyond a certain point. Indeed, only Dwimmermount, Adamas, and Muntburg have much detail at all and while the players have heard of the existence of other places, like Yethlyreom, City of the Necromancers, or Thule, the ghost-haunted island whose ancient inhabitants forged an empire for the ages, they've never traveled there or learned much about them.
The result is a very focused campaign and I think that's part of the secret of its success. Dwimmermount is the centerpiece of the campaign and it's where the characters go, week after week, in their quest for adventure. This has also made it very easy for me to run, since I know precisely where the PCs are headed and I can prepare accordingly -- or not, as the case may be, since, depending on how far the PCs have gotten, a megadungeon-oriented campaign may not require the creation of new material on a weekly basis, previously created material not having been exhausted yet. It's a very different way of playing, to be sure, but it's been quite satisfying and it's helped me clarify quite a few things in my own mind as a result.
But old habits die hard and there's a part of me that finds it difficult to restrain myself and not flesh out the world beyond Dwimmermount more than is necessary for weekly play. I find myself fiddling with maps, imagining other city-states and nations, creating exotic religions, and conjuring up strange vistas beyond the horizon. It's hard not to do this, because I enjoy doing it and because, for most of my time in the hobby, creating worlds was what referees were expected to do. Now, don't misunderstand: I am regularly creating a world for the Dwimmermount campaign. Every time I populate a room in the megadungeon, for example, I'm adding little details that give the greater whole depth and context that it wouldn't otherwise have and I enjoy this. The difficulty is that I find myself wanting to go beyond this and detail more than my players need -- or likely will ever need -- to play enjoyably in the campaign.
So far, I've (largely) resisted the urge to go hog wild and detail the entire world. There's simply no need to do so and I've come to accept that, the less that is set in stone, the better, as it gives me a lot more freedom in my refereeing. Every detail I establish before it's needed is a link in a chain that restricts my future movement and that of my players. It's better to be patient, focus on the here and now, and fill in details of gods and nations and other worlds only when there's an immediate need to do so. It's a very different way of creating worlds than I am used to -- or temperamentally inclined towards -- but it's also a very rewarding one and one, I suspect, that's closer to the way things were done back in the early days. Even if it isn't, I've learned a lot by adopting it and highly recommend it to others, if they've never done so themselves.