[molbioguy]

You're right, but I always understood that Light would support multiple languages. The Kickstarter site mentions "The first two languages it will support are Javascript and Clojure, but the application will be written in such a way that adding new languages can happen through plugins." And I think Python will be the third based on contributions. So it's good not to discourage people that aren't using clojure.

[Sandman]

True, but it's probably a lot easier to build an IDE by concentrating on supporting one language, at least in the beginning, and thinking about support for other languages later on. Don't forget that this is not even an alpha version of Light Table yet. This is simply a demo, showing us what we might expect. I have no doubt that both JavaScript and Python will be supported in the final version of LT.

[deafbybeheading]

I think what alttab was saying is that you should still post updates without assuming you're just talking to a bunch of clojure aficionados.

E.g., I'm somewhat interested in LT but for the other languages. I'm curious enough to check out the odd blog post now and then, but if it's all gibberish, I won't come back and may lose track of the project and miss the other language support when it does land.

[alttab]

Nailed it sir.

[ibdknox]

Yeah, there's a ton to learn about how all this should and needs to work. Spreading ourselves across supporting the basics of multiple languages this early means our rate of learning decreases dramatically. We're using Clojure to prove a bunch of assumptions because it's the easiest to support quickly and so far it's been great. Our first goal is to build the best Clojure env out there and then move from that solid base into everything else.

[rhizome]

I think that might be a mistake in strategy. If what you posit is true, that means they're going to have to re-invent their way of talking about the interface and usage for each language, though probably they would start to settle their terminology on the third or fourth reinvention. If this is so, that means they're actually wasting energy now in tying the app story to Clojure concepts.

[alttab]

This is true. The story might break with ruby or JS, and no doubt the abstractions and assumptions of the behavior of how the programming language works will break the minute you go to a different execution model.

[rhizome]

It's a tightrope, for sure, and it can be hard to see when you're in the middle of it.