Maybe it's just me, but I honestly have trouble taking anyone who uses mustache templates seriously because I automatically assume they've never worked on a large website.
I'd be really interested to see how Mustache does in real world performance vs., say, PHP or Ruby templates. I'm also curious about using it to offload rendering to the browser, particularly when using Ajax or UI binding.
While Mustache sometimes seems like a solution in search of problem, I do like that it's completely logicless, and so forces all logic, even display logic, into the controller.
I interviewed for a job @Twilio - they were using Handlebars, a Mustache variant. Also, here is my mini-chance to rant about that: I had 4 phone and 6 in-person interviews -- didn't get hired. Argh! How could anyone make it through that? Also - i'm a generalist, which seems to have kept me out of both Facebook and Twilio. /rant
@dustindiaz I'm curious about the usecase for Matador. Its mentioned in the docs that it uses Express. I use Express pretty heavily and it seems that most of the things made available via Matador is already available in Express. Why should I consider stacking Matador on top of a framework that already works pretty well?
Great question. And fwiw, we love Express too! After building a number of Express applications myself, I quite often ran into organizational issues (such is the loosey goosey way of Sinatra). Matador takes more of a Rails approach to organizing your code; of course, without all the extra cruft that comes with Rails.
We still tack on Express requests and responses onto your controller instances.
Most of all, the router -> controller mapping with request filtering (middleware-like) is where Matador comes in most useful. We have other features we'd like to add, especially in the scaffolding layer, but figured we'd open source early and get developer feedback.
TJ's goal (in my view) is to make a really unopinionated way for node to interact with the web in a modern way, standardizing things like routes, sessions, etc. Matador would provide some opinion on top of Express to standardize file locations, or do a Rails-like inference of inflection perhaps. (I haven't looked deeply).
I wrote my own version of this last year, and found that it was a really good learning experience, but I didn't think it was worth publishing publicly. However, there is definitely room for a scaffolded MVC prototyping framework on top of Express, imo.
yup that's certainly correct. Express is still intentionally much lower level so myself and everyone can just craft what they want without fighting some larger framework
16 comments
[ 2.8 ms ] story [ 51.5 ms ] threadI recommend you check out Chris Wanstrath's mustache library for Ruby: https://github.com/defunkt/mustache#readme
While Mustache sometimes seems like a solution in search of problem, I do like that it's completely logicless, and so forces all logic, even display logic, into the controller.
We still tack on Express requests and responses onto your controller instances.
Most of all, the router -> controller mapping with request filtering (middleware-like) is where Matador comes in most useful. We have other features we'd like to add, especially in the scaffolding layer, but figured we'd open source early and get developer feedback.
Thanks!
I wrote my own version of this last year, and found that it was a really good learning experience, but I didn't think it was worth publishing publicly. However, there is definitely room for a scaffolded MVC prototyping framework on top of Express, imo.