I'm against changing this, and I think a vote is superfluous so **I do not want a vote**.
When I first found Nim I though the style insensitivity was a bit strange. But the more I use it the more I like it. And after having had to write some code in C and Ruby that doesn't have it I'm even more for it. The fact of the matter is that you will never be able to get 100% of people to write in one style (and even if you did, is it dataBase or database, fileName or filename?), so all code that isn't style insensitive will be at risk of ending up as a mixed style project. It only takes one "bad" import. Sure you can fix this by making a lot of bindings, but why would you, when you can have style insensitivity for free. Besides, apart from the ubiquitous object/instance differentiation (which Nim also has an exception for) it's widely regarded to be a bad idea to have style-sensitive identifiers anyways. Having a variable fileName and a procedure filename would just be confusing. The only case that can be made here is that an identifier like word1_word2 would collide with wo_rd1word2 which is a problem for a very small set of word combinations where both would be normal names but not the same. Couldn't think of any example of this though. Besides style sensitivity is a relatively new concept. The only reason we have snake-case is because spaces can't be used in identifiers, and the only reason we have upper case letters as separate characters is that the ASCII group decided that it would result in safer encoding. The alternative was sending "Hello" as "<switch to uppercase symbol>H<switch to lowercase symbol>ello". The choice was made as missing the <switch to lowercase symbol> would mean the entire rest of the document would be upper-case, which would be annoying.
