James Brantly

> I'm thinking that it makes more sense to pour energy into getting stuff merged into core rather than porting to 15.0, but I'm willing to hear disagreeing viewpoints. FWIW,...

Thanks! I'll take a look into it when I can.

Sorry for the delay. I had someone else try the tests (I didn't have Chromium handy) and they passed. The error sounds like a cross-domain issue perhaps, or something similar....

It's also a little sketch that it uses "Microsoft-corp". It's difficult to verify that this is actually Microsoft.

First off, realize that I'm coming from a place of ignorance here in that I haven't really looked at the ts-loader code in detail since the great refactoring and I...

Also, is it just me or would things work a whole lot better if TypeScript's operation was asynchronous? Does anyone know if this has been brought up to them before?

There's quite a bit of information on this over at https://github.com/TypeStrong/ts-loader/issues/78. It's still an open issue for ts-loader as well, which does use the language service. The main slow point...

Just a quick 2 cents on this since something similar is implemented in ts-loader. You [can implement `resolveModuleNames` to achieve this functionality](https://github.com/TypeStrong/ts-loader/blob/34dd5c99330e3768997b7ebc5ec8a33f4d996edf/index.ts#L356-L385). This code (roughly) first tries to resolve using...

> I'm assuming that any code that you write using the Webpack resolve logic will look broken to Intellisense Yup. That's the major downside :frowning: BUT I'm hoping when https://github.com/Microsoft/TypeScript/issues/5039...

> It seems to me that a better end state would be a zero-configuration setup in which Browserify used TypeScript's module resolution directly Very interesting. I'll have to keep that...