First off, let me say that I think people are getting too hung up on the word "core" and the idea that the "core rules" will be exactly the same as the simplest game that they present. I've been trying to use the word "basic" for that game, and "core" for the rules in the Big3 books, which will cover playstyle options for the current editions of the games (if I understand what's been said so far.) I see it like this:
Basic Game < Big3 "Core" books < all of 5e*
*includes rules modules and material from all the Campaign Settings, Supplemental Splatbooks, adventures, etc.
I hope for a "red box" that handles just the Basic Game for those that are new or want to sample it. The idea that "I hate Rule X and don't want it taking up pages in my PHB/DMG" doesn't mesh with the stated goals of 5e so far, IMO. (Nor does it make much sense to me..."Wait, you want
fewer options for your game?")
I have to say that this is an interesting and important subject. I certainly have been a bit worried by this habit of treating everything as a module, simply because I don't think it is possible to create so many modules as some people seem to want and still be able to connect one module to another or even feasibly connect them to the core. The core needs to allow a certain amount of room for different kinds of modules. In its own way, you need to build the core in anticipation of every really major module that will be added to it. Similarly, you need to build modules so that they fit with each other. I really believe that the game can only really be built on a fairly small number of major modules, and that adding too many more modules too long after the initial release of the game won't really work.
I don't think the goal is to make
all the modules cross-compatible. I mean that some modules will be serving the same purpose and thus you only pick one of them. So yes, any given campaign will be built with a relatively small number of modules. However, even a small number of choices for each module type will lead to a lot of (possible) variations of play.
In other words, the core game can't be that simple, and any effective module will pretty much need to be created from the beginning. You can't just turn any given hope for the game into a module because there won't be enough room for them all. People really do need to settle these issues, not just hope for modularity to make everyone happy.
Keeping in mind that some modules will exchange/replace the simple modules that comprise the core....I don't think its actually that hard. This isn't computer programming, we humans are pretty flexible and if my memories of 2e are any indication, we are quite willing to patch up rules left and right with or without a module to help us. DMs and players are very fault-tolerant, something that later editions seem to forget or ignore (no judgement implied).
That being said, I believe that there will be some who just refuse to let it work for them. I'm not sure why, but many seem fixated on the idea that the "basic" game be the game
as they want to play it.
My biggest fear for the game is that WotC gets too shortsighted in its approach and initially just makes a core without building it to actually be modular. If they just think they can make a module for something later without planning ahead, then it will be a nightmare for the players who want to play with those modules.
I don't take that as a given. I really think this is the actual purpose behind the emphasis on a simple basic game. The simpler that base is and the less intertwined the mechanics are, the easier it is to add or exchange modules. The utility and ease of later modules is far more dependent on the playtesting and design effort that goes into those modules than the core rules being built specifically with them in mind. I think the 2e "Options" books were illustrative of this idea. Some of them dug in and altered some pretty deep architecture of the original game, a game that was certainly not designed with them in mind. While some of those were better than others, it wasn't their interoperability that was the issue. (IME)