Any helpful ideas?
If it is for pay, or possible pay, then by all means dive in and see what you can do.
But a purely for fun project with copious alternatives that handle the vast majority of what you're trying to do and have the ability to be extended, means you'll eventually run out of steam and realize you can just use these other tools. I've been working on a twitter app for months and I've run out of steam. I have plenty of ideas and improvements, and as none of the ones out there really allow extensions, I'm still plugging away at my one, but I've taken absurdly long to get a lightly functional app.
Be warned, that way lies madness.
Pick a set of features you want to start with and how you will interact with it. Pick what language/API/platform/etc you want it to work under and start reading docs and thinking how to go about writing your own.
Not to discourage you, but you probably won't be able to match most editors having never written one before. If you just want a hobby project, it be useful to start it for the experience. It might teach you a lot about software, or however, it might just frustrate you.
edit: What is your experience in? What kind of code do you write now?
A good editor can be a good user interface to all kinds of interesting things, like language and compiler development.