My Ideal Published Megadungeon

Over at Grognardia, James asks a question that I have spent quite some time thinking about:

So, today's question is this: assuming you're interested in megadungeons -- if you're not, please don't use the comments section to express your disinterest -- what would be your preferred format? Feel free and assume that there are no limits and go with what you would consider to be the ideal format for presenting a true old school megadungeon.

I have never been a huge fan of modules. Sure, I have used them and will continue to integrate them into my campaigns, but they are not even close to being a regular feature in any campaign I've ever run. The reason for this is utility.

The vast majority of modules have a huge amount of information about every single aspect of the dungeon — especially rooms. While I enjoy reading this stuff, all that information gets in the way of running the module. Instead of being able to quickly ascertain information about an individual room, I have to wade through a bunch of information I don't really need when interacting with players. The game slows down and I inevitably miss a key bit of information that gums up the works and makes my evening less-than-fun.

Therefore, my ideal megadungeon module would be something that I could easily utilize in play. There are four ways that I would do this:

1. Make the dungeon background interesting but easily adaptable. In a typical megadungeon of my own design, for example, I imagine that it has gone through three phases: the original builders; those that overthrew the builders; current occupiers. In my ideal megadungeon publication, the module would use this basic outline and then give an example of how to utilize it. Thus, I am invited to use it as written, completely overhaul it, or adapt it for my campaign — whichever is most useful for me and my current need.
2. Give visual cues on the map as to who and when certain parts of the dungeon were built and the factions that currently occupy them. Thus, at a glance I have enough information to describe a room and how it differs from the hallway the players just came from. This can be done simply by shading certain areas and keying the shading as Original; Conquerers; Current; (insert faction name here).
3. Have several different short entries for each room. Each entry should be a phrase that describes the utility of the room for each era. If released as a .pdf, using color would be ideal. For example: red = Original; blue = Conquerers; black = Current. For a printable version and/or a BW printed version, plain, underlined and boxed texts could be used. This way I have a wealth of information to give players with built-in layers. If the room was originally a vestry, used as a weapons room by the conquerers and a storage room by the current occupiers I have information for those that give the room a cursory glance (storage room); those that look closely at the contents (what is stored there); and for those that look beyond (weapons racks and religious symbols on the wall). All this at a glance.
4. Finally, as was done in B1, provide a blank space for customized monsters as well as a normal stat line for the monsters that the author envisioned for his own version of the dungeon. I realize that this adds space and therefore cost to a print run; however, given the space saved by having extremely short room descriptions I think it worth it.

Two final thoughts: for those that want crunch and background material, each level can have a summary of history and current factions at the beginning of each level (much in the same way that Michael Curtis did with Stonehell). Finally, have the maps in a separate file/booklet. This way, at the table I can have both the map AND the description right in front of me without having to flip back and forth between pages.