jayderyu - Yeah definitely convenience is a huge deal, every single click you can reduce is a big deal when it comes to getting people to learn tools. But it's just not something that's missing from C2 and is not difficult to use. The main problem is no-one taking advantage of it, which stems a lot from it seeming like a lot of work. I worked with Excal a few times earlier on doing the 'copy your work over' approach and it was hell. It was just never made obvious how easy an SVN is to set-up, and when you look for collab solutions on the forums, you mainly find 'CANNOT DO! NOT FOR TEAMS! PREPARE FOR HELL!', instead of a whole bunch of 'oh just do this'.
So, I think instead of adding features it's far more efficient for Scirra to shove what's available in people's faces. I mean come on, look at the insane amount of effort people go to to try to get their games working on mobile, and then those same people think an SVN is too hard.
Excal - I think the 'help me make my game' is a legitimate way to go, but only if you have enough experience. As you said, when you team up you feel you're all dragging in different directions, which is not a good way to go! So from the start it should be understood who is the designer, and that person should set out the systems for the rest to work off. Sure, the others should give their ideas and try to put as much of themselves into the game as possible, but the lead really needs to be able to say 'sorry, but that's not the direction of the game.'. This is the best thing for the game and those trying to learn.
Also different specialities are a must - I love making experience systems and in-depth damage systems, but as soon as it needs to be represented on screen productivity grinds to a halt. Whereas there are many people who only want to lay down images on screen and have them work. Of course it helps if everyone understands how and why other things are done, but specialisation streamlines production in a big way.
I do think it would rock the house if there was a 'module capx' section of sorts. I've learnt so much from people posting answers to their own questions when they find them, but you need to ask some specific questions in searches to come across these gems. I made this system for example, that lets you easily manipulate arrays and would love to get some feedback on it, but I don't really know where to post it and don't see many getting use from it after a day or two when it sinks in the forums. I've already expanded it to have many more applications (like pushing on axis even when duplicates, acquiring indexes of Y, etc.), and it only requires you to have functions and a family, nothing more. And the thing is, even though it's great for me, I have no idea if something similar is out there or where to find it if it is. For all I know I've just wasted a heap of time on an inferior version of something someone made months ago.
And actually I believe modularity, as totally fantastic as it would be, is easily second to multiplayer. Everyone wants to make multiplayer games, and currently they can't. Not many people want to collaborate, but they can. So modularity is clearly a second place here, although a massively important second place.
But whatever the case or viewpoint on all of the above, this is definitely a conversation that needs to appear more often with C2 :)