-
Notifications
You must be signed in to change notification settings - Fork 4
CS2 Discussion: Project: A clear narrative for CoffeeScript: are we muddying the water with now CS1 and CS2? #75
Comments
Hey, there has been some discussion of this in #71 My view is that CoffeeScript 2, when it is released out of beta, will be the one true version of CoffeeScript. CoffeeScript 1 is hopefully already frozen; ideally, 1.12.4 will be the last 1.x release. We certainly won’t be releasing any more features on the 1.x branch, and I doubt there will be much interest in backporting bugfixes to 1.x either. I wouldn’t be opposed to approving pull requests if others want to submit backported bugfixes, but that would be a low priority. The upgrade effort, in my opinion, is minor. There are very few breaking changes in 2, and attaching Babel to the build chain is probably something people started doing back in 1.11 with the introduction of modules. I don’t feel guilty about asking people to make the effort in order to get the latest and greatest bugfixes and features. I’m nervous about putting 2.0.0 on And yes, developers needing to output ES5 or ES3 can use CoffeeScript 2 with Babel. That’s what I would expect most developers to be doing. |
@GeoffreyBooth that all seems sane. Re putting 2.0.0 on I understand that using the |
@greghuc I’m open to publishing 2.0.0 on |
Just to throw one example out there, the Rails coffee-script gem simply pulls the latest Anyone care to check any other major downstream users of CoffeeScript, to see how those tools pull the compiler in? |
Migrated to jashkenas/coffeescript#4968 |
I've been wondering how maintaining the CS2 and CS1 branches is going to pan out, and if/when CS1 can be deprecated. Specifically, can Coffeescript return to a world where there's "one true version" (CS2) with a single set of features + documentation?
My concern is that whilst CS2 is awesome, and moves Coffeescript forward, we're at risk of muddying the waters with having two separate versions of Coffeescript medium term.
Seems to me that there's two separate developer audiences that need considering:
I'm sure @GeoffreyBooth has thought about this, but I thought raising an issue was worthwhile.
It would be good to get some clarity so there's a clear narrative the CS/JS dev community can easily understand. Plus I imagine maintaining two parallel code branches longterm sucks.
The text was updated successfully, but these errors were encountered: