[semi-off topic rant, start!]
In my humble opinion, you guys are being way too harsh to the O/P. There's no reason to be so dismissive.
What he is proposing doesn't really sound too much like Game Maker. Sure, it has some similarities, but those similarities are ones shared by nearly every IDE + engine package. The "Animation Editor" is the only thing out of the ordinary (though not THAT out of the ordinary) - and, XHH, that's something I would pass on, as it's unnecessary since there is plenty of excellent animation software out there (both free and commercial).
All he's saying is a project manager with class/object management, a level editor, and a publishing tool. This is exactly the same as things like Polycode, the old Torque 2D (back before it went open source and the scrapped half of it), Dreemchest, Unity, UDK, and the like come with - all of which (with the exception of Dreemchest, perhaps) are much more complicated than LOVE (for better or for worse), and much, much more complicated than Game Maker.
Also, as far as something like Game Maker being the "wrong way" or "too easy"...I completely disagree. Game Maker is a tool that is very good at what it does, and is fully capable of making great games (as has been shown time and time again). It has positives and it has negatives, just like LOVE, or Unity, or XNA, or building an engine from scratch, etc... It is not the "wrong way," it is A way. It is not "too easy," it's different. Yes, it does more for you, but in the world of game development that's a GOOD THING, assuming that it doesn't get in the way of what you want to do. There's a reason nearly every professional team licenses game engines and middle-ware - they don't want to have to do all the work, they want as much ease as they can get while still keeping the flexibility that they need. And, quite frankly, Game Maker (for example) has seen far more success as far as quality of games than LOVE ( I don't mean this in a bad way towards LOVE, AT ALL). Just look at games like Hotline Miami, Desktop Dungeons, Spelunky, Oil Blue, Cinders, Frozzd, and even games like Super Crate Box, Dustforce, Death Worm, Reflexions, Ninjammin, etc... These games may have been made the "wrong way," but they are fun games that a lot of people love, and, in many cases, they are making their creators money and even won awards - they are successful.
At the end of the day, we are here to make games. Everybody has their own thing/style, everybody has their own reasons, etc... All this silliness about Game Maker or IDEs making things too easy is...well, silly. And, besides, not everyone here is even interested in being a game programmer. Many just want to make games, and see programming as a means to an end, as a tool to get them where they want to go. If they can get where they want to go with less programming, then more power to them.
[end semi-off topic rant]
XHH:
If creating this idea is something you really want to do, then go for it. I'm sure that a lot of people would love a lightweight IDE/engine combo like this where they can dig deep into the code, but have a nice interface to make things a little easier. I think that a lot of beginners probably have issues with getting a good workflow going with something like LOVE, so at the very least your idea would really help them.
That being said:
1. I would suggest that you make yourself EXTREMELY familiar with LOVE and Lua before starting. You should know both like the back of you hand. Otherwise, you'll probably end up making bad design decisions that you'll regret later.
2. If you are mainly creating this to facilitate your own creation of games, then I would suggest that you start out by learning LOVE and making a complete game (even a simple one) or two with just LOVE and some external tools and libraries.
3. If you want to move into the "pro game designer" world, then I do suggest that you do learn how to do things at a lower level. But, in that case, I definitely wouldn't suggest that you use something like LOVE for that, as it is very high level. The traditional route here would be learning C++, and then DirectX and OpenGL, and then going on from there.
Oh, and as far as if others would use your IDE: if you made a good, lightweight piece of software, then I'm sure that people would use it. I personally would probably not use it, unless it was excellent. Why? I tend to make games/prototypes in one of two mindsets: 1. I want a nice clean set of tools that lets me do just about anything without setting any rules or styles in stone (which is when I use LOVE, at least for 2D, or something even simpler like Haxe + NME), or 2. I want more of a complete game engine that has a lot of game-centric stuff setup (which is when I'd use things like FlashPunk/HaxePunk, Polycode, and more recently WaveEngine). So, basically, this type of software is probably not something I'd use, unless it was for a massive project and I needed really solid project (source + asset) management (though Polycode would probably fill that gap for me). Most LOVErs probably wouldn't use it either, as I think that many of the people here are using LOVE
because they don't like IDEs like what you're proposing (which is probably why you received the response you did).
If you are more interested in a more complete "turnkey" package, then perhaps you should look into things like Unity or even Torque 2D (which will apparently be getting editors again relatively soon). If you don't mind being on the cutting edge, then I highly recommend that you also check out Polycode, which will end up being pretty much exactly what you want...but, it's still in development, so don't expect everything to work 100%. At this point though, it has a built-in level editor (which kind of feels like a cross between Game Maker's Room Editor and Tiled), an animation editor, project management, and a script editor, along with a full framework that can be access either with Lua or C++. There have already been a few great games made with it, so definitely something that you might want to watch.
tl;dr? I suggest that you ,ake your idea for yourself first, and don't worry about others just yet. But before even that, make sure you look around (like at some of the options I mentioned above) to make sure that what you need/want isn't already out there and available to you.
:exhales:
Hopefully that helps
. The way I've organized the code, I could easily turn it into FreePascal or C++ if I decided to for whatever reason. I know someone is going to say "ew pithon is slow!" but that's not true at all.