The whole premise of Rails is convention over configuration. Build a set of tools optimized for the average application following "best practices". And guess what? Rails is still the best framework for producing the average Web application using 2006 best practices.
Now it's 2013. The average application server starts with Rails, then throws away ERB, routing, form_for, and the entire MVC lifecycle and slaps Ember on top of it. Technology has changed a lot in seven years (mobile, javascript, blah blah blah), and more importantly, it's only getting more client side
exactly what rails has become to me. Models and API controllers that respond_to json. The client does the rest with Backbone, handlebars, css3 animations, ...
Same. I feel that for apps that are mostly server-side, Rails and Django are the way to go, especially CRUD apps. In the quickly emerging client-side app market, they often serve as an overly complex JSON api that requires too much configuration IMO.
While I like what Meteor is attempting to do by merging the client and server into one framework, I feel their approach is somewhat haphazard. Using it left a bad taste in my mouth, though t be fair I haven't done much beyond building a basic app for evaluation purposes. I'm currently evaluating Derby.js, and so far I'm liking it much more.
That said, convention has yet to be established for clientside apps, so either framework could fail. I do think that there is a niche for a framework that provides both the server and the client, however.
what I would like to see is a Rails type framework specifically geared towards building APIs. I like asset pipeline, give me that, I like gems, I like ActiveRecord, but take views, helpers and all that jazz. I will do that on the frontend thankyouverymuch
http://bustle.com/ is a wonderful example of a consumer website that uses Ember. Square's payment dashboard is a great example of a web app that uses Ember. Tilde also maintains a list of users that wish to publicize their use of Ember: http://emberjs.com/ember-users/
"Meteor looks very interesting, but until it supports more of the above, we’re talking apples and oranges."
If only one could somehow contribute to open source projects. Oh wait... But I guess complaining and comparing a 1 year old platform with a platform that's almost 10 years old is easier.
When you consider a funding of $11.2 M [1], then you'll not be surprised at how things can go so incredibly fast.
Please next time do your investigation homework before comparing two totally different software ecosystems.
If Ruby on Rails had that much money from the beginning now you would be running the fastest, most beautifully coded/structured super-scalable real-time app in the world :)
Go is a programming language. Rails and Meteor are frameworks. Although Go does come with more web framework features than most languages (like templates and an http server), well except for PHP.
Which, just to make things clear to new comers, is just a reply to the absurdity of the claim of the original blog post which titled "Why Meteor will kill Rails".
The whole premise of Rails is convention over configuration. Build a set of tools optimized for the average application following "best practices". And guess what? Rails is still the best framework for producing the average Web application using 2006 best practices.
Now it's 2013. The average application server starts with Rails, then throws away ERB, routing, form_for, and the entire MVC lifecycle and slaps Ember on top of it. Technology has changed a lot in seven years (mobile, javascript, blah blah blah), and more importantly, it's only getting more client side