Would their be a big difference in using a map generated from a file for a project? I mean, I think personally I'd prefer to have a mapper map a project for me via another application programmed just for that, rather than messing with the source and cutting stuff out for what they'll need.

Would it take too much time loading the mentioned map from a file to be even worth it? Do you think it could be an effective way of running a world?
You could always get Dropbox, put the map file in a specific folder, then give someone access to that folder. The map should still read your code to get objects, but they can't access it. Theoretically, anyways.
I don't think it would, as presumably without access to where the code is, dropbox on their machine won't replicate it and so it's not available locally for them to build against.
That would be really cool if it worked that way. I figured I have three options.

1. I could give them the whole source. Obviously this is a really bad choice.
2. I could create another environment and then copy over the icons, and the code for the icons so they're structured properly in the tree.
3. I could create a mapping program which will save all the atoms, then would be loaded from the host whenever the world started.

I'm not sure how well the third one would work, as I've never done it before. It seems like it would take a lot longer to load and all. Placing everything on the map each boot.

I was kind of thinking the mapping program just so resource ripping is a tad harder and takes a tad bit longer. From what I hear the RSC extractors are harder to find, and since they won't have the images and pixel art first hand, they'd end up having to print screen everything. Generally it won't come out the same as the original either, I don't think.

All in all, if someones going to get resources from a project, they will. I'm not not sure if this extra step is even worth it, and if it is if it will actually run smoothly enough to be worth it.
Loading a map from savefiles is obviously going to be slower than loading it from a compiled .dmm file.

1. Sharing the entire source is usually bad.
2. This is what I've done before, and it works so long as you keep the paths exactly the same, so it's compatible with the source code. Of course, it's still giving out all the icons, which hasn't affected me before (the people I worked with aren't rippin' idiots), but it's safe to be safe if that's what you choose to do.
3. It's possible, and potentially more flexible than Dream Maker's map editor, but I don't think it's worth doing if you can just do #2, or even #1.

I've seen Lummox talk about native map saving/loading. Maybe look in to that?
If you're still interested in option #2, I just finished a tool that should make it easier to set up:

Map Data Extractor

Emphasis on just, mind you, so there may still be a few kinks to work out.

Also, you don't have to include the icon files if you include an RSC, but they're pitifully easy to extract if you have the RSC (a simple fcopy() or ftp() will do it). And the native support for loading maps may still be a long ways off. That said, if you plan to publicly release a map editor, you can't beat option #3 as players can share and play maps without waiting for you to recompile the game.
Now that's quite a thought. The ability to update the map without having to recompile the game. I never thought about it that way, but it sounds like it would be really convenient.

The ability to update the map on reboot, or even at runtime sounds like a really cool function. Especially if used for events, and things similar to that.

I'll probably highly consider option number 3 due to the fact that it can be so flexible. DarkCampainger, I just took a quick look at your Map Data Extractor and it looks amazing. I'll be sure to give it a go and see how well everything goes when I start developing the map maker.