back
28 comments
A hair more informative than a list of technologies and some slogans: https://github.com/brunch/brunch/blob/master/readme.md

Really, that's the only thing that comes dangerously close to a concise explanation of what this project actually is - and it makes it sound like a couple of programs that generate a fairly typical MVC website folder structure and auto-compile a few types of files whenever they're changed.

If Brunch is more than that, it might be worthwhile to at least have a gloss of what else it does around, even if the rest of the documentation is in progress. Otherwise, the understanding of any awesomeness is going to be limited to people who're going to plumb the code or went to that Vienna user group presentation. :)

thank you for your feedback! we definitely need not only to improve documentation but also figure out how to concisely communicate what it is all about.

very valuable advice, that is exactly what we aimed for :)

Good deal; I'll keep an eye out. :)
Similar project: Capt - https://github.com/bnolan/capt by Ben Nolan ( Anyone remembers behaviors.js for Prototype way way before jQuery ? )

Also see: Faux - https://github.com/unspace/faux

This would be perfect for a project I'm working on if it supported HAML-JS/SASS (optionally replacing eco/stylus).

Is the framework extendable in this respect?

we currently focus on recommending a certain set of components and aim to make their integration seamless.

once we got that and the documentation and code examples are improved we will look into making it easier to switch certain components for people who want something different. comparable to rails (ruby) where you can switch activerecord with something different.

so we certainly don't intend to make replacing components a hassle :)

for what it's worth, the meat here is really only in jquery, backbone, and underscore. the rest is either redundant or preferential, no?

edit: downvotes welcome, given an answer to my question.

thank you for your question.

you have a point. I still see that it is very valuable to provide tooling, documentation and conventions for this toolchain since more and more people use it but every project you see on HN/github/… looks a bit different.

I believe these kinds of opinionated conventions you know from other frameworks really add benefit. So I think there are quite some backbone users who intentionally chose backbone because it does not dictate as much as other client side frameworks do, but there are also many people who want to use backbone and coffeescript and would love to have an easer getting started experience :)

interesting topic.

we just presented brunch at the vienna javascript user group (viennajs.org).

we still need some input on good client side testing tools that would make sense to be integrated with a js framework. further we need to get rid of some rough edges and will focus on providing documentation and examples.

currently we would like to get feedback on the chosen toolchain and love to hear suggestions on how to improve their integration (we currently ship a file watcher, still need to add uglify support and different build targets - build.phonegap.com, chrome webstore, …)

cheers, tosh

This looks quite neat indeed, and I can't wait to start playing around with it.

That said, I'd definitely like to see some integrated support beyond Stylus for things like SASS, LESS, and/or Compass.

Keep up the good work!

Ditto. Stylus looks great, and its in npm so thats a win. But Compass has the most watchers on GitHub and therefore I'd assume the most usage.

But all-in-all, awesome!

thank you for your feedback. npm support was a big win for us since we tried to get the dependencies (especially on languages and package managers) simple.

we definitely will keep sass/scss in mind since it is very popular in the rails community. once the core components work well together we will look into customization (also for the templating component).

that said, I believe there is value in a recommended set of components to have consistent documentation, code examples and a straightforward getting-started experience. but for people who know what to do it should be easy to switch components (comparable to replacing activerecord in rails). easy things should be easy, hard things should be possible :)

In my experience Compass has been painfully slow to compile. (Haven't tried out Stylus yet.)
Here are some Hacker News threads related to client side testing:

http://news.ycombinator.com/item?id=2038663

http://news.ycombinator.com/item?id=2142104

http://news.ycombinator.com/item?id=2055867

great pointer! zombie looks pretty promising and makes a good fit with the node ecosystem.
I'm currently working on a haml parser that supports coffescript. Would be awesome if the template logic would be agnostic.
We are definitely aiming to provide ways to integrate other templating frameworks. Keep us updated on your progress! Haml combined with CoffeeScript would be awesome.
This looks seriously useful for my next weekend project. That said, I'm looking for a caching, RESTful persistence layer for Backbone that will sync to the network if available, and otherwise caches data locally... anybody know of one?
Looks like a really good fit to what's currently going on in the community. Get working hard on the documentation and keep HN posted on how you progress!
that's what we are aiming for :)
kool-aid overdose! It makes me want to build something just for the sake of playing with cool tech
please do :) backbone, underscore & coffeescript really feel right and we'd love to see more people using them together.
Looks awesome. Wish there was a way to host a free app with it, Google App Engine style.
Nice presentation, thanks again!
had a look at it a couple of days ago. looks indeed very promising!
seen the presentation at viennajs. looks very promising.
yeah. very good deal. keept it up!
Simply based on name alone it reminds me of my project Crumb: http://breefield.com/crumb/