[Questions] Risk of adopting ES6 features


ES6 has been around for many years, we are very conservative using it. As it gets more and more adoption, maybe we should start taking advantage of it in the code generation and runtime support for BuckleScript

The question to the community is that how will this impact your use case? Do you already have polyfill to it, what’s your minimum browser support?



Both projects I worked for during the last 2 years still support Internet Explorer. Both had corporate customers. If it was my business, I’d probably pull the plug on IE already :slight_smile: Sure, we’d lose a few customers but instead would get greater velocity and smaller bundles.

BTW, none of those projects use BuckleScript, it’s either tsc or tsc+Babel. Maybe that’s not a coincidence: TypeScript feels like a safer bet for the big corps.


One of the idea/feature that attracted me to ReasonML was that you can commit the generated readable code to your repository but that’s not possible for me because most of projects today, at least where i work, are in ES6 and hence readable code to be committed in repo should be ES6 compliant as well.
And i think if we get an option to generate ES6 code, it will be easier to get more people on board and get them to use it at their work instead of only community/pet projects. Because not everyone can get whole team on boarded right away.


I think that this is a natural development and definitely a question of when, rather than if. ES6 would also help with having the output look closer to the input, amongst other things.

It makes sense to bear in mind the target audience of reason, which are likely slightly more agressive with being on the front foot than the norm. Even without that consideration, many businesses have already stopped support of IE11 due to the burden of support vs user base size. By the time an ES6 output could be implemented this will only have moved on even further.

In terms of users that may be on IE, Facebook is probably the outlier here due to the massive userbase (but I think it’s only on messenger.com), so it will likely impact FB much more than the rest of the community.

Given that I would say press on…


I think the risk is mitigated because of Babel. If people really want to target exactly ES4/5/whatever they can further transpile. Also, as I see it almost everyone is using Babel anyway somehow or the other, because of the transpilation/bundling/asset pipeline. So yeah. This is not much of a worry for us.