You didn't ruffle my feathers, but I can easily see how you'd ruffle the feathers of someone who's invested a tremendous amount of time into a new language. >

You didn't ruffle my feathers, but I can easily see how you'd ruffle the feathers of someone who's invested a tremendous amount of time into a new language. > but can't help but think the time would be better spent adding feature Y to language X, but didn't mean to start anything :-) When people create new languages, the premise is rarely along the lines of: "I like this language, but if it only had this feature." The truth is there are a whole slew of very legitimate reasons for creating a new language. Many of those reasons are predicated on the idea that the mainstream alternatives have too many wrong features and/or it would be difficult-impossible to fix those flaws by adding new features. As you already know, it's a lot harder to remove a feature from a mainstream language than it is to introduce a new feature, so languages tend to become more complicated as time goes on. New languages tend to provide a clean slate and opportunity to learn from the mistakes of its predecessors. Another reason to create new languages: New languages often act as the test bed for ideas that eventually end up in mainstream languages. As mainstream languages mature, they rely on more experimental languages to introduce ideas so they can be tested. The only way to test these ideas is to have people write programs in these languages and study the outcome. With that being said, if it wasn't for the experimental languages, mainstream languages would not likely have things like objects, or lambda functions, continuations, type inferencing, polymorphic type systems, etc. So, if you find yourself being down-voted for asking questions like, "why don't you just add feature Y to language X", it's because you're in the wrong venue. You're not the first to ask the question. I'm not the first to answer it, so you're bound to ruffle feathers.