Looks like an interesting project, but that kind of steganography doesn't defeat an aggressive censor (look at my mention of whitespace in OP, it's in that category); the messages will be detectable from scanning. (Remember that steganography is about hiding that the message exists, not only hiding the message). Doesn't mean it isn't useful; it's just about what's your threat model. If your threat model is an all-powerful censor, that's not enough.