The question is whether Webpack is simply a solution that just changes the problem.
Lately the major versions just seem to be chasing one another. So would
“When will Phoenix distribute with Webpack?”
simply be traded for
“When will Phoenix update the (major) version of Webpack it comes with?”
Ditto.
After looking at Brunch’s twitter account along with the GitHub repo as well as a few brunch packages, it does indeed look like it’s dying or dead.
I can honestly say, I never really appreciated Rails Asset pipeline until I started mucking around with JS build systems. I have a preference for gulp - and havn’t even tried webpack- but know I hate brunch.
Rather than just discussing this I thought I’d contribute some code. This: A working first pass at using npm scripts instead of brunch. by baldmountain · Pull Request #2740 · phoenixframework/phoenix · GitHub is a pull request to make the default use npm scripts rather than brunch. I only kind of know what I’m doing when it comes to JavaScript/CSS build pipelines but this PR seems to generate a working project with watching JS and CSS supported. (I think the JS pipeline is OK, but don’t think CSS pipeline is correct.)
@baldmountain Sucks it got shutdown so quickly without much discussion 
I pinged the Brunch dev team to see if it is still being actively developed. @paulmillr (one of the devs) replied that dependencies are still being maintained. I did not recieve confirmation that the project is still actively being developed, :shrug:
@brunch Is this project still actively maintained? Or has it been back-burner'd?
— Rich Smith (@RichJDSmith) February 4, 2018
/cc @paulmillr & @ElanShanker
@richjdsmith I’m OK with it being shut down. For learning Phoenix, or just getting started, brunch is just fine and continues to work.
Developers who are going to build a React, Angular, Vue.js (or some other JS framework) app are going to know more about JS build pipelines and are going to replace bunch as part of starting up a project. So brunch is still fine by me.
But I do think it is something that will need to be replaced at some point. There’s just no rush to do it now.
Perfectly reasonable attitude. I agree completely. I’d just like to see there be a discussion before jumping on board The Next Great Build Tool/Task Runner™
Ruby on Rails was super opinionated and that could sometimes be a massive PITA, but I still miss the Asset Pipeline.
