Backbone.js gives structure to web applications by providing models with key-value binding and custom events, collections with a rich API of enumerable functions, views with declarative event handling, and connects it all to your existing API over a RESTful JSON interface.

The project is hosted on GitHub, and the annotated source code is available, as well as an online test suite, an example application, a list of tutorials and a long list of real-world projects that use Backbone. Backbone is available for use under the MIT software license.

You can report bugs and discuss features on the GitHub issues page, on Freenode IRC in the #documentcloud channel, post questions to the Google Group, add pages to the wiki or send tweets to @documentcloud.

Backbone is an open-source component of DocumentCloud.

Downloads & Dependencies (Right-click, and use "Save As")

Development Version (1.0.0) 58kb, Full source, tons of comments
Production Version (1.0.0) 7.1kb, Packed and gzipped
(Source Map)
Edge Version (master) Unreleased, use at your own risk

Backbone's only hard dependency is Underscore.js ( >= 1.4.3). For RESTful persistence, history support via Backbone.Router and DOM manipulation with Backbone.View, include json2.js, and either jQuery ( >= 1.7.0) or Zepto.


When working on a web application that involves a lot of JavaScript, one of the first things you learn is to stop tying your data to the DOM. It's all too easy to create JavaScript applications that end up as tangled piles of jQuery selectors and callbacks, all trying frantically to keep data in sync between the HTML UI, your JavaScript logic, and the database on your server. For rich client-side applications, a more structured approach is often helpful.

With Backbone, you represent your data as Models, which can be created, validated, destroyed, and saved to the server. Whenever a UI action causes an attribute of a model to change, the model triggers a "change" event; all the Views that display the model's state can be notified of the change, so that they are able to respond accordingly, re-rendering themselves with the new information. In a finished Backbone app, you don't have to write the glue code that looks into the DOM to find an element with a specific id, and update the HTML manually — when the model changes, the views simply update themselves.

If you're new here, and aren't yet quite sure what Backbone is for, start by browsing the list of Backbone-based projects.

Many of the examples that follow are runnable. Click the play button to execute them.

Upgrading to 1.0

Backbone 1.0 should be a fairly painless upgrade from the 0.9 series. If you're upgrading from an older version, be sure to check out the change log. In brief, a few of the larger breaking changes are:


Events is a module that can be mixed in to any object, giving the object the ability to bind and trigger custom named events. Events do not have to be declared before they are bound, and may take passed arguments. For example:

var object = {};

_.extend(object, Backbone.Events);

object.on("alert", function(msg) {
  alert("Triggered " + msg);

object.trigger("alert", "an event");

For example, to make a handy event dispatcher that can coordinate events among different areas of your application: var dispatcher = _.clone(Backbone.Events)

onobject.on(event, callback, [context])Alias: bind
Bind a callback function to an object. The callback will be invoked whenever the event is fired. If you have a large number of different events on a page, the convention is to use colons to namespace them: "poll:start", or "change:selection". The event string may also be a space-delimited list of several events...

book.on("change:title change:author", ...);

To supply a context value for this when the callback is invoked, pass the optional third argument: model.on('change', this.render, this)

Callbacks bound to the special "all" event will be triggered when any event occurs, and are passed the name of the event as the first argument. For example, to proxy all events from one object to another:

proxy.on("all", function(eventName) {

All Backbone event methods also support an event map syntax, as an alternative to positional arguments:

  "change:title": titleView.update,
  "change:author": authorPane.update,
  "destroy": bookView.remove
});[event], [callback], [context])Alias: unbind
Remove a previously-bound callback function from an object. If no context is specified, all of the versions of the callback with different contexts will be removed. If no callback is specified, all callbacks for the event will be removed. If no event is specified, callbacks for all events will be removed.

// Removes just the `onChange` callback."change", onChange);

// Removes all "change" callbacks."change");

// Removes the `onChange` callback for all events., onChange);

// Removes all callbacks for `context` for all events., null, context);

// Removes all callbacks on `object`.;

Note that calling, for example, will indeed remove all events on the model — including events that Backbone uses for internal bookkeeping.

triggerobject.trigger(event, [*args])
Trigger callbacks for the given event, or space-delimited list of events. Subsequent arguments to trigger will be passed along to the event callbacks.

onceobject.once(event, callback, [context])
Just like on, but causes the bound callback to only fire once before being removed. Handy for saying "the next time that X happens, do this".

listenToobject.listenTo(other, event, callback)
Tell an object to listen to a particular event on an other object. The advantage of using this form, instead of other.on(event, callback), is that listenTo allows the object to keep track of the events, and they can be removed all at once later on.

view.listenTo(model, 'change', view.render);

stopListeningobject.stopListening([other], [event], [callback])
Tell an object to stop listening to events. Either call stopListening with no arguments to have the object remove all of its registered callbacks ... or be more precise by telling it to remove just the events it's listening to on a specific object, or a specific event, or just a specific callback.



listenToOnceobject.listenToOnce(other, event, callback)
Just like listenTo, but causes the bound callback to only fire once before being removed.

Catalog of Events
Here's the complete list of built-in Backbone events, with arguments. You're also free to trigger your own events on Models, Collections and Views as you see fit. The Backbone object itself mixes in Events, and can be used to emit any global events that your application needs.


Models are the heart of any JavaScript application, containing the interactive data as well as a large part of the logic surrounding it: conversions, validations, computed properties, and access control. You extend Backbone.Model with your domain-specific methods, and Model provides a basic set of functionality for managing changes.

The following is a contrived example, but it demonstrates defining a model with a custom method, setting an attribute, and firing an event keyed to changes in that specific attribute. After running this code once, sidebar will be available in your browser's console, so you can play around with it.

var Sidebar = Backbone.Model.extend({
  promptColor: function() {
    var cssColor = prompt("Please enter a CSS color:");
    this.set({color: cssColor});

window.sidebar = new Sidebar;

sidebar.on('change:color', function(model, color) {
  $('#sidebar').css({background: color});

sidebar.set({color: 'white'});


extendBackbone.Model.extend(properties, [classProperties])
To create a Model class of your own, you extend Backbone.Model and provide instance properties, as well as optional classProperties to be attached directly to the constructor function.

extend correctly sets up the prototype chain, so subclasses created with extend can be further extended and subclassed as far as you like.

var Note = Backbone.Model.extend({

  initialize: function() { ... },

  author: function() { ... },

  coordinates: function() { ... },

  allowedToEdit: function(account) {
    return true;


var PrivateNote = Note.extend({

  allowedToEdit: function(account) {
    return account.owns(this);


Brief aside on super: JavaScript does not provide a simple way to call super — the function of the same name defined higher on the prototype chain. If you override a core function like set, or save, and you want to invoke the parent object's implementation, you'll have to explicitly call it, along these lines:

var Note = Backbone.Model.extend({
  set: function(attributes, options) {
    Backbone.Model.prototype.set.apply(this, arguments);

constructor / initializenew Model([attributes], [options])
When creating an instance of a model, you can pass in the initial values of the attributes, which will be set on the model. If you define an initialize function, it will be invoked when the model is created.

new Book({
  title: "One Thousand and One Nights",
  author: "Scheherazade"

In rare cases, if you're looking to get fancy, you may want to override constructor, which allows you to replace the actual constructor function for your model.

var Library = Backbone.Model.extend({
  constructor: function() {
    this.books = new Books();
    Backbone.Model.apply(this, arguments);
  parse: function(data, options) {
    return data.library;

If you pass a {collection: ...} as the options, the model gains a collection property that will be used to indicate which collection the model belongs to, and is used to help compute the model's url. The model.collection property is otherwise added automatically when you first add a model to a collection.

If {parse: true} is passed as an option, the attributes will first be converted by parse before being set on the model.

Get the current value of an attribute from the model. For example: note.get("title")

setmodel.set(attributes, [options])
Set a hash of attributes (one or many) on the model. If any of the attributes change the model's state, a "change" event will be triggered, unless {silent: true} is passed as an option. Change events for specific attributes are also triggered, and you can bind to those as well, for example: change:title, and change:content. You may also pass individual keys and values.

note.set({title: "March 20", content: "In his eyes she eclipses..."});

book.set("title", "A Scandal in Bohemia");

Similar to get, but returns the HTML-escaped version of a model's attribute. If you're interpolating data from the model into HTML, using escape to retrieve attributes will prevent XSS attacks.

var hacker = new Backbone.Model({
  name: "<script>alert('xss')</script>"


Returns true if the attribute is set to a non-null or non-undefined value.

if (note.has("title")) {

unsetmodel.unset(attribute, [options])
Remove an attribute by deleting it from the internal attributes hash. Fires a "change" event unless silent is passed as an option.

Removes all attributes from the model, including the id attribute. Fires a "change" event unless silent is passed as an option.
A special property of models, the id is an arbitrary string (integer id or UUID). If you set the id in the attributes hash, it will be copied onto the model as a direct property. Models can be retrieved by id from collections, and the id is used to generate model URLs by default.

A model's unique identifier is stored under the id attribute. If you're directly communicating with a backend (CouchDB, MongoDB) that uses a different unique key, you may set a Model's idAttribute to transparently map from that key to id.

var Meal = Backbone.Model.extend({
  idAttribute: "_id"

var cake = new Meal({ _id: 1, name: "Cake" });
alert("Cake id: " +;

A special property of models, the cid or client id is a unique identifier automatically assigned to all models when they're first created. Client ids are handy when the model has not yet been saved to the server, and does not yet have its eventual true id, but already needs to be visible in the UI.

The attributes property is the internal hash containing the model's state — usually (but not necessarily) a form of the JSON object representing the model data on the server. It's often a straightforward serialization of a row from the database, but it could also be client-side computed state.

Please use set to update the attributes instead of modifying them directly. If you'd like to retrieve and munge a copy of the model's attributes, use _.clone(model.attributes) instead.

Due to the fact that Events accepts space separated lists of events, attribute names should not include spaces.

The changed property is the internal hash containing all the attributes that have changed since the last "change" event was triggered. Please do not update changed directly since its state is internally maintained by set. A copy of changed can be acquired from changedAttributes.

defaultsmodel.defaults or model.defaults()
The defaults hash (or function) can be used to specify the default attributes for your model. When creating an instance of the model, any unspecified attributes will be set to their default value.

var Meal = Backbone.Model.extend({
  defaults: {
    "appetizer":  "caesar salad",
    "entree":     "ravioli",
    "dessert":    "cheesecake"

alert("Dessert will be " + (new Meal).get('dessert'));

Remember that in JavaScript, objects are passed by reference, so if you include an object as a default value, it will be shared among all instances. Instead, define defaults as a function.

Return a copy of the model's attributes for JSON stringification. This can be used for persistence, serialization, or for augmentation before being sent to the server. The name of this method is a bit confusing, as it doesn't actually return a JSON string — but I'm afraid that it's the way that the JavaScript API for JSON.stringify works.

var artist = new Backbone.Model({
  firstName: "Wassily",
  lastName: "Kandinsky"

artist.set({birthday: "December 16, 1866"});


syncmodel.sync(method, model, [options])
Uses Backbone.sync to persist the state of a model to the server. Can be overridden for custom behavior.

Resets the model's state from the server by delegating to Backbone.sync. Returns a jqXHR. Useful if the model has never been populated with data, or if you'd like to ensure that you have the latest server state. A "change" event will be triggered if the server's state differs from the current attributes. Accepts success and error callbacks in the options hash, which are passed (model, response, options) and (model, xhr, options) as arguments, respectively.

// Poll every 10 seconds to keep the channel model up-to-date.
setInterval(function() {
}, 10000);[attributes], [options])
Save a model to your database (or alternative persistence layer), by delegating to Backbone.sync. Returns a jqXHR if validation is successful and false otherwise. The attributes hash (as in set) should contain the attributes you'd like to change — keys that aren't mentioned won't be altered — but, a complete representation of the resource will be sent to the server. As with set, you may pass individual keys and values instead of a hash. If the model has a validate method, and validation fails, the model will not be saved. If the model isNew, the save will be a "create" (HTTP POST), if the model already exists on the server, the save will be an "update" (HTTP PUT).

If instead, you'd only like the changed attributes to be sent to the server, call, {patch: true}). You'll get an HTTP PATCH request to the server with just the passed-in attributes.

Calling save with new attributes will cause a "change" event immediately, a "request" event as the Ajax request begins to go to the server, and a "sync" event after the server has acknowledged the successful change. Pass {wait: true} if you'd like to wait for the server before setting the new attributes on the model.

In the following example, notice how our overridden version of Backbone.sync receives a "create" request the first time the model is saved and an "update" request the second time.

Backbone.sync = function(method, model) {
  alert(method + ": " + JSON.stringify(model)); = 1;

var book = new Backbone.Model({
  title: "The Rough Riders",
  author: "Theodore Roosevelt"
});;{author: "Teddy"});

save accepts success and error callbacks in the options hash, which are passed (model, response, options) and (model, xhr, options) as arguments, respectively. If a server-side validation fails, return a non-200 HTTP response code, along with an error response in text or JSON."author", "F.D.R.", {error: function(){ ... }});

Destroys the model on the server by delegating an HTTP DELETE request to Backbone.sync. Returns a jqXHR object, or false if the model isNew. Accepts success and error callbacks in the options hash, which are passed (model, response, options) and (model, xhr, options) as arguments, respectively. Triggers a "destroy" event on the model, which will bubble up through any collections that contain it, a "request" event as it begins the Ajax request to the server, and a "sync" event, after the server has successfully acknowledged the model's deletion. Pass {wait: true} if you'd like to wait for the server to respond before removing the model from the collection.

book.destroy({success: function(model, response) {

Underscore Methods (6)
Backbone proxies to Underscore.js to provide 6 object functions on Backbone.Model. They aren't all documented here, but you can take a look at the Underscore documentation for the full details…

user.pick('first_name', 'last_name', 'email');

chapters.keys().join(', ');

validatemodel.validate(attributes, options)
This method is left undefined, and you're encouraged to override it with your custom validation logic, if you have any that can be performed in JavaScript. By default validate is called before save, but can also be called before set if {validate:true} is passed. The validate method is passed the model attributes, as well as the options from set or save. If the attributes are valid, don't return anything from validate; if they are invalid, return an error of your choosing. It can be as simple as a string error message to be displayed, or a complete error object that describes the error programmatically. If validate returns an error, save will not continue, and the model attributes will not be modified on the server. Failed validations trigger an "invalid" event, and set the validationError property on the model with the value returned by this method.

var Chapter = Backbone.Model.extend({
  validate: function(attrs, options) {
    if (attrs.end < attrs.start) {
      return "can't end before it starts";

var one = new Chapter({
  title : "Chapter One: The Beginning"

one.on("invalid", function(model, error) {
  alert(model.get("title") + " " + error);
  start: 15,
  end:   10

"invalid" events are useful for providing coarse-grained error messages at the model or collection level.

The value returned by validate during the last failed validation.

Run validate to check the model state.

var Chapter = Backbone.Model.extend({
  validate: function(attrs, options) {
    if (attrs.end < attrs.start) {
      return "can't end before it starts";

var one = new Chapter({
  title : "Chapter One: The Beginning"

  start: 15,
  end:   10

if (!one.isValid()) {
  alert(one.get("title") + " " + one.validationError);

Returns the relative URL where the model's resource would be located on the server. If your models are located somewhere else, override this method with the correct logic. Generates URLs of the form: "[collection.url]/[id]" by default, but you may override by specifying an explicit urlRoot if the model's collection shouldn't be taken into account. You can also pass in the model's url as an option when instantiating it.

Delegates to Collection#url to generate the URL, so make sure that you have it defined, or a urlRoot property, if all models of this class share a common root URL. A model with an id of 101, stored in a Backbone.Collection with a url of "/documents/7/notes", would have this URL: "/documents/7/notes/101"

urlRootmodel.urlRoot or model.urlRoot()
Specify a urlRoot if you're using a model outside of a collection, to enable the default url function to generate URLs based on the model id. "[urlRoot]/id"
Normally, you won't need to define this. Note that urlRoot may also be a function.

var Book = Backbone.Model.extend({urlRoot : '/books'});

var solaris = new Book({id: "1083-lem-solaris"});


parsemodel.parse(response, options)
parse is called whenever a model's data is returned by the server, in fetch, and save. The function is passed the raw response object, and should return the attributes hash to be set on the model. The default implementation is a no-op, simply passing through the JSON response. Override this if you need to work with a preexisting API, or better namespace your responses.

If you're working with a Rails backend that has a version prior to 3.1, you'll notice that its default to_json implementation includes a model's attributes under a namespace. To disable this behavior for seamless Backbone integration, set:

ActiveRecord::Base.include_root_in_json = false

Returns a new instance of the model with identical attributes.

Has this model been saved to the server yet? If the model does not yet have an id, it is considered to be new.

Has the model changed since the last "change" event? If an attribute is passed, returns true if that specific attribute has changed.

Note that this method, and the following change-related ones, are only useful during the course of a "change" event.

book.on("change", function() {
  if (book.hasChanged("title")) {

Retrieve a hash of only the model's attributes that have changed, or false if there are none. Optionally, an external attributes hash can be passed in, returning the attributes in that hash which differ from the model. This can be used to figure out which portions of a view should be updated, or what calls need to be made to sync the changes to the server.

During a "change" event, this method can be used to get the previous value of a changed attribute.

var bill = new Backbone.Model({
  name: "Bill Smith"

bill.on("change:name", function(model, name) {
  alert("Changed name from " + bill.previous("name") + " to " + name);

bill.set({name : "Bill Jones"});

Return a copy of the model's previous attributes. Useful for getting a diff between versions of a model, or getting back to a valid state after an error occurs.


Collections are ordered sets of models. You can bind "change" events to be notified when any model in the collection has been modified, listen for "add" and "remove" events, fetch the collection from the server, and use a full suite of Underscore.js methods.

Any event that is triggered on a model in a collection will also be triggered on the collection directly, for convenience. This allows you to listen for changes to specific attributes in any model in a collection, for example: Documents.on("change:selected", ...)

extendBackbone.Collection.extend(properties, [classProperties])
To create a Collection class of your own, extend Backbone.Collection, providing instance properties, as well as optional classProperties to be attached directly to the collection's constructor function.

Override this property to specify the model class that the collection contains. If defined, you can pass raw attributes objects (and arrays) to add, create, and reset, and the attributes will be converted into a model of the proper type.

var Library = Backbone.Collection.extend({
  model: Book

A collection can also contain polymorphic models by overriding this property with a function that returns a model.

var Library = Backbone.Collection.extend({

  model: function(attrs, options) {
    if (condition) {
      return new PublicDocument(attrs, options);
    } else {
      return new PrivateDocument(attrs, options);


constructor / initializenew Collection([models], [options])
When creating a Collection, you may choose to pass in the initial array of models. The collection's comparator may be included as an option. Passing false as the comparator option will prevent sorting. If you define an initialize function, it will be invoked when the collection is created. There are several options that, if provided, are attached to the collection directly: url, model and comparator.

var tabs = new TabSet([tab1, tab2, tab3]);
var spaces = new Backbone.Collection([], {
  model: Space,
  url: '/spaces'

Raw access to the JavaScript array of models inside of the collection. Usually you'll want to use get, at, or the Underscore methods to access model objects, but occasionally a direct reference to the array is desired.

Return an array containing the attributes hash of each model in the collection. This can be used to serialize and persist the collection as a whole. The name of this method is a bit confusing, because it conforms to JavaScript's JSON API.

var collection = new Backbone.Collection([
  {name: "Tim", age: 5},
  {name: "Ida", age: 26},
  {name: "Rob", age: 55}


synccollection.sync(method, collection, [options])
Uses Backbone.sync to persist the state of a collection to the server. Can be overridden for custom behavior.

Underscore Methods (28)
Backbone proxies to Underscore.js to provide 28 iteration functions on Backbone.Collection. They aren't all documented here, but you can take a look at the Underscore documentation for the full details…

books.each(function(book) {

var titles = {
  return book.get("title");

var publishedBooks = books.filter(function(book) {
  return book.get("published") === true;

var alphabetical = books.sortBy(function(book) {

addcollection.add(models, [options])
Add a model (or an array of models) to the collection. Fires an "add" event, which you can pass {silent: true} to suppress. If a model property is defined, you may also pass raw attributes objects, and have them be vivified as instances of the model. Pass {at: index} to splice the model into the collection at the specified index. If you're adding models to the collection that are already in the collection, they'll be ignored, unless you pass {merge: true}, in which case their attributes will be merged into the corresponding models, firing any appropriate "change" events.

var ships = new Backbone.Collection;

ships.on("add", function(ship) {
  alert("Ahoy " + ship.get("name") + "!");

  {name: "Flying Dutchman"},
  {name: "Black Pearl"}

Note that adding the same model (a model with the same id) to a collection more than once
is a no-op.

removecollection.remove(models, [options])
Remove a model (or an array of models) from the collection. Fires a "remove" event, which you can use silent to suppress. The model's index before removal is available to listeners as options.index.

resetcollection.reset([models], [options])
Adding and removing models one at a time is all well and good, but sometimes you have so many models to change that you'd rather just update the collection in bulk. Use reset to replace a collection with a new list of models (or attribute hashes), triggering a single "reset" event at the end. Pass {silent: true} to suppress the "reset" event. For convenience, within a "reset" event, the list of any previous models is available as options.previousModels.

Here's an example using reset to bootstrap a collection during initial page load, in a Rails application:

  var accounts = new Backbone.Collection;
  accounts.reset(<%= @accounts.to_json %>);

Calling collection.reset() without passing any models as arguments will empty the entire collection.

setcollection.set(models, [options])
The set method tries to peform a "smart" update of the collection with the passed list of models. If a model in the list isn't yet in the collection it will be added; if the model is already in the collection its attributes will be merged; and if the collection contains any models that aren't present in the list, they'll be removed. All of the appropriate "add", "remove", and "change" events are fired as this happens. If you'd like to customize the behavior, you can disable it with options: {add: false}, {remove: false}, or {merge: false}.

var vanHalen = new Collection([eddie, alex, stone, roth]);

vanHalen.update([eddie, alex, stone, hagar]);

// Fires a "remove" event for roth, and an "add" event for "hagar".
// Updates any of stone, alex, and eddie's attributes that may have
// changed over the years.

Get a model from a collection, specified by an id, a cid, or by passing in a model.

var book = Library.get(110);
Get a model from a collection, specified by index. Useful if your collection is sorted, and if your collection isn't sorted, at will still retrieve models in insertion order.

pushcollection.push(model, [options])
Add a model at the end of a collection. Takes the same options as add.

Remove and return the last model from a collection. Takes the same options as remove.

unshiftcollection.unshift(model, [options])
Add a model at the beginning of a collection. Takes the same options as add.

Remove and return the first model from a collection. Takes the same options as remove.

slicecollection.slice(begin, end)
Return a shallow copy of this collection's models, using the same options as native Array#slice.

Like an array, a Collection maintains a length property, counting the number of models it contains.

By default there is no comparator for a collection. If you define a comparator, it will be used to maintain the collection in sorted order. This means that as models are added, they are inserted at the correct index in collection.models. A comparator can be defined as a sortBy (pass a function that takes a single argument), as a sort (pass a comparator function that expects two arguments), or as a string indicating the attribute to sort by.

"sortBy" comparator functions take a model and return a numeric or string value by which the model should be ordered relative to others. "sort" comparator functions take two models, and return -1 if the first model should come before the second, 0 if they are of the same rank and 1 if the first model should come after.

Note how even though all of the chapters in this example are added backwards, they come out in the proper order:

var Chapter  = Backbone.Model;
var chapters = new Backbone.Collection;

chapters.comparator = function(chapter) {
  return chapter.get("page");

chapters.add(new Chapter({page: 9, title: "The End"}));
chapters.add(new Chapter({page: 5, title: "The Middle"}));
chapters.add(new Chapter({page: 1, title: "The Beginning"}));


Collections with a comparator will not automatically re-sort if you later change model attributes, so you may wish to call sort after changing model attributes that would affect the order.

Force a collection to re-sort itself. You don't need to call this under normal circumstances, as a collection with a comparator will sort itself whenever a model is added. To disable sorting when adding a model, pass {sort: false} to add. Calling sort triggers a "sort" event on the collection, unless silenced by passing {silent: true}.

Pluck an attribute from each model in the collection. Equivalent to calling map, and returning a single attribute from the iterator.

var stooges = new Backbone.Collection([
  {name: "Curly"},
  {name: "Larry"},
  {name: "Moe"}

var names = stooges.pluck("name");


Return an array of all the models in a collection that match the passed attributes. Useful for simple cases of filter.

var friends = new Backbone.Collection([
  {name: "Athos",      job: "Musketeer"},
  {name: "Porthos",    job: "Musketeer"},
  {name: "Aramis",     job: "Musketeer"},
  {name: "d'Artagnan", job: "Guard"},

var musketeers = friends.where({job: "Musketeer"});


Just like where, but directly returns only the first model in the collection that matches the passed attributes.

urlcollection.url or collection.url()
Set the url property (or function) on a collection to reference its location on the server. Models within the collection will use url to construct URLs of their own.

var Notes = Backbone.Collection.extend({
  url: '/notes'

// Or, something more sophisticated:

var Notes = Backbone.Collection.extend({
  url: function() {
    return this.document.url() + '/notes';

parsecollection.parse(response, options)
parse is called by Backbone whenever a collection's models are returned by the server, in fetch. The function is passed the raw response object, and should return the array of model attributes to be added to the collection. The default implementation is a no-op, simply passing through the JSON response. Override this if you need to work with a preexisting API, or better namespace your responses.

var Tweets = Backbone.Collection.extend({
  // The Twitter Search API returns tweets under "results".
  parse: function(response) {
    return response.results;

Returns a new instance of the collection with an identical list of models.

Fetch the default set of models for this collection from the server, resetting the collection when they arrive. The options hash takes success and error callbacks which will be passed (collection, response, options) and (collection, xhr, options) as arguments, respectively. When the model data returns from the server, the collection will be (efficiently) reset, unless you pass {update: true}, in which case it will use update to (intelligently) merge the fetched models. Delegates to Backbone.sync under the covers for custom persistence strategies and returns a jqXHR. The server handler for fetch requests should return a JSON array of models.

Backbone.sync = function(method, model) {
  alert(method + ": " + model.url);

var accounts = new Backbone.Collection;
accounts.url = '/accounts';


The behavior of fetch can be customized by using the available update options. For example, to fetch a collection, getting an "add" event for every new model, and a "change" event for every changed existing model, without removing anything: collection.fetch({update: true, remove: false})

jQuery.ajax options can also be passed directly as fetch options, so to fetch a specific page of a paginated collection: Documents.fetch({data: {page: 3}})

Note that fetch should not be used to populate collections on page load — all models needed at load time should already be bootstrapped in to place. fetch is intended for lazily-loading models for interfaces that are not needed immediately: for example, documents with collections of notes that may be toggled open and closed.

createcollection.create(attributes, [options])
Convenience to create a new instance of a model within a collection. Equivalent to instantiating a model with a hash of attributes, saving the model to the server, and adding the model to the set after being successfully created. Returns the new model. If client-side validation failed, the model will be unsaved, with validation errors. In order for this to work, you should set the model property of the collection. The create method can accept either an attributes hash or an existing, unsaved model object.

Creating a model will cause an immediate "add" event to be triggered on the collection, a "request" event as the new model is sent to the server, as well as a "sync" event, once the server has responded with the successful creation of the model. Pass {wait: true} if you'd like to wait for the server before adding the new model to the collection.

var Library = Backbone.Collection.extend({
  model: Book

var nypl = new Library;

var othello = nypl.create({
  title: "Othello",
  author: "William Shakespeare"


Web applications often provide linkable, bookmarkable, shareable URLs for important locations in the app. Until recently, hash fragments (#page) were used to provide these permalinks, but with the arrival of the History API, it's now possible to use standard URLs (/page). Backbone.Router provides methods for routing client-side pages, and connecting them to actions and events. For browsers which don't yet support the History API, the Router handles graceful fallback and transparent translation to the fragment version of the URL.

During page load, after your application has finished creating all of its routers, be sure to call Backbone.history.start(), or Backbone.history.start({pushState: true}) to route the initial URL.

extendBackbone.Router.extend(properties, [classProperties])
Get started by creating a custom router class. Define actions that are triggered when certain URL fragments are matched, and provide a routes hash that pairs routes to actions. Note that you'll want to avoid using a leading slash in your route definitions:

var Workspace = Backbone.Router.extend({

  routes: {
    "help":                 "help",    // #help
    "search/:query":        "search",  // #search/kiwis
    "search/:query/p:page": "search"   // #search/kiwis/p7

  help: function() {

  search: function(query, page) {


The routes hash maps URLs with parameters to functions on your router (or just direct function definitions, if you prefer), similar to the View's events hash. Routes can contain parameter parts, :param, which match a single URL component between slashes; and splat parts *splat, which can match any number of URL components. Part of a route can be made optional by surrounding it in parentheses (/:optional).

For example, a route of "search/:query/p:page" will match a fragment of #search/obama/p2, passing "obama" and "2" to the action.

A route of "file/*path" will match #file/nested/folder/file.txt, passing "nested/folder/file.txt" to the action.

A route of "docs/:section(/:subsection)" will match #docs/faq and #docs/faq/installing, passing "faq" to the action in the first case, and passing "faq" and "installing" to the action in the second.

When the visitor presses the back button, or enters a URL, and a particular route is matched, the name of the action will be fired as an event, so that other objects can listen to the router, and be notified. In the following example, visiting #help/uploading will fire a route:help event from the router.

routes: {
  "help/:page":         "help",
  "download/*path":     "download",
  "folder/:name":       "openFolder",
  "folder/:name-:mode": "openFolder"
router.on("route:help", function(page) {

constructor / initializenew Router([options])
When creating a new router, you may pass its routes hash directly as an option, if you choose. All options will also be passed to your initialize function, if defined.

routerouter.route(route, name, [callback])
Manually create a route for the router, The route argument may be a routing string or regular expression. Each matching capture from the route or regular expression will be passed as an argument to the callback. The name argument will be triggered as a "route:name" event whenever the route is matched. If the callback argument is omitted router[name] will be used instead.

initialize: function(options) {

  // Matches #page/10, passing "10"
  this.route("page/:number", "page", function(number){ ... });

  // Matches /117-a/b/c/open, passing "117-a/b/c" to
  this.route(/^(.*?)\/open$/, "open");


open: function(id) { ... }

navigaterouter.navigate(fragment, [options])
Whenever you reach a point in your application that you'd like to save as a URL, call navigate in order to update the URL. If you wish to also call the route function, set the trigger option to true. To update the URL without creating an entry in the browser's history, set the replace option to true.

openPage: function(pageNumber) {;
  this.navigate("page/" + pageNumber);

# Or ...

app.navigate("help/troubleshooting", {trigger: true});

# Or ...

app.navigate("help/troubleshooting", {trigger: true, replace: true});


History serves as a global router (per frame) to handle hashchange events or pushState, match the appropriate route, and trigger callbacks. You shouldn't ever have to create one of these yourself since Backbone.history already contains one.

pushState support exists on a purely opt-in basis in Backbone. Older browsers that don't support pushState will continue to use hash-based URL fragments, and if a hash URL is visited by a pushState-capable browser, it will be transparently upgraded to the true URL. Note that using real URLs requires your web server to be able to correctly render those pages, so back-end changes are required as well. For example, if you have a route of /documents/100, your web server must be able to serve that page, if the browser visits that URL directly. For full search-engine crawlability, it's best to have the server generate the complete HTML for the page ... but if it's a web application, just rendering the same content you would have for the root URL, and filling in the rest with Backbone Views and JavaScript works fine.

When all of your Routers have been created, and all of the routes are set up properly, call Backbone.history.start() to begin monitoring hashchange events, and dispatching routes.

To indicate that you'd like to use HTML5 pushState support in your application, use Backbone.history.start({pushState: true}). If you'd like to use pushState, but have browsers that don't support it natively use full page refreshes instead, you can add {hashChange: false} to the options.

If your application is not being served from the root url / of your domain, be sure to tell History where the root really is, as an option: Backbone.history.start({pushState: true, root: "/public/search/"})

When called, if a route succeeds with a match for the current URL, Backbone.history.start() returns true. If no defined route matches the current URL, it returns false.

If the server has already rendered the entire page, and you don't want the initial route to trigger when starting History, pass silent: true.

Because hash-based history in Internet Explorer relies on an <iframe>, be sure to only call start() after the DOM is ready.

  new WorkspaceRouter();
  new HelpPaneRouter();
  Backbone.history.start({pushState: true});


Backbone.sync is the function that Backbone calls every time it attempts to read or save a model to the server. By default, it uses jQuery.ajax to make a RESTful JSON request and returns a jqXHR. You can override it in order to use a different persistence strategy, such as WebSockets, XML transport, or Local Storage.

The method signature of Backbone.sync is sync(method, model, [options])

With the default implementation, when Backbone.sync sends up a request to save a model, its attributes will be passed, serialized as JSON, and sent in the HTTP body with content-type application/json. When returning a JSON response, send down the attributes of the model that have been changed by the server, and need to be updated on the client. When responding to a "read" request from a collection (Collection#fetch), send down an array of model attribute objects.

Whenever a model or collection begins a sync with the server, a "request" event is emitted. If the request completes successfully you'll get a "sync" event, and an "error" event if not.

The sync function may be overriden globally as Backbone.sync, or at a finer-grained level, by adding a sync function to a Backbone collection or to an individual model.

The default sync handler maps CRUD to REST like so:

As an example, a Rails handler responding to an "update" call from Backbone might look like this: (In real code, never use update_attributes blindly, and always whitelist the attributes you allow to be changed.)

def update
  account = Account.find params[:id]
  account.update_attributes params
  render :json => account

One more tip for integrating Rails versions prior to 3.1 is to disable the default namespacing for to_json calls on models by setting ActiveRecord::Base.include_root_in_json = false

ajaxBackbone.ajax = function(request) { ... };
If you want to use a custom AJAX function, or your endpoint doesn't support the jQuery.ajax API and you need to tweak things, you can do so by setting Backbone.ajax.

emulateHTTPBackbone.emulateHTTP = true
If you want to work with a legacy web server that doesn't support Backbone's default REST/HTTP approach, you may choose to turn on Backbone.emulateHTTP. Setting this option will fake PUT and DELETE requests with a HTTP POST, setting the X-HTTP-Method-Override header with the true method. If emulateJSON is also on, the true method will be passed as an additional _method parameter.

Backbone.emulateHTTP = true;;  // POST to "/collection/id", with "_method=PUT" + header.

emulateJSONBackbone.emulateJSON = true
If you're working with a legacy web server that can't handle requests encoded as application/json, setting Backbone.emulateJSON = true; will cause the JSON to be serialized under a model parameter, and the request to be made with a application/x-www-form-urlencoded mime type, as if from an HTML form.


Backbone views are almost more convention than they are code — they don't determine anything about your HTML or CSS for you, and can be used with any JavaScript templating library. The general idea is to organize your interface into logical views, backed by models, each of which can be updated independently when the model changes, without having to redraw the page. Instead of digging into a JSON object, looking up an element in the DOM, and updating the HTML by hand, you can bind your view's render function to the model's "change" event — and now everywhere that model data is displayed in the UI, it is always immediately up to date.

extendBackbone.View.extend(properties, [classProperties])
Get started with views by creating a custom view class. You'll want to override the render function, specify your declarative events, and perhaps the tagName, className, or id of the View's root element.

var DocumentRow = Backbone.View.extend({

  tagName: "li",

  className: "document-row",

  events: {
    "click .icon":          "open",
    "click .button.edit":   "openEditDialog",
    "click .button.delete": "destroy"

  initialize: function() {
    this.listenTo(this.model, "change", this.render);

  render: function() {


Properties like tagName, id, className, el, and events may also be defined as a function, if you want to wait to define them until runtime.

constructor / initializenew View([options])
When creating a new View, the options you pass — after being merged into any default options already present on the view — are attached to the view as this.options for future reference. There are several special options that, if passed, will be attached directly to the view: model, collection, el, id, className, tagName and attributes. If the view defines an initialize function, it will be called when the view is first created. If you'd like to create a view that references an element already in the DOM, pass in the element as an option: new View({el: existingElement})

var doc = Documents.first();

new DocumentRow({
  model: doc,
  id: "document-row-" +

All views have a DOM element at all times (the el property), whether they've already been inserted into the page or not. In this fashion, views can be rendered at any time, and inserted into the DOM all at once, in order to get high-performance UI rendering with as few reflows and repaints as possible. this.el is created from the view's tagName, className, id and attributes properties, if specified. If not, el is an empty div.

var ItemView = Backbone.View.extend({
  tagName: 'li'

var BodyView = Backbone.View.extend({
  el: 'body'

var item = new ItemView();
var body = new BodyView();

alert(item.el + ' ' + body.el);

A cached jQuery object for the view's element. A handy reference instead of re-wrapping the DOM element all the time.



If you'd like to apply a Backbone view to a different DOM element, use setElement, which will also create the cached $el reference and move the view's delegated events from the old element to the new one.

A hash of attributes that will be set as HTML DOM element attributes on the view's el (id, class, data-properties, etc.), or a function that returns such a hash.

$ (jQuery)view.$(selector)
If jQuery is included on the page, each view has a $ function that runs queries scoped within the view's element. If you use this scoped jQuery function, you don't have to use model ids as part of your query to pull out specific elements in a list, and can rely much more on HTML class attributes. It's equivalent to running: view.$el.find(selector)

ui.Chapter = Backbone.View.extend({
  serialize : function() {
    return {
      title: this.$(".title").text(),
      start: this.$(".start-page").text(),
      end:   this.$(".end-page").text()

The default implementation of render is a no-op. Override this function with your code that renders the view template from model data, and updates this.el with the new HTML. A good convention is to return this at the end of render to enable chained calls.

var Bookmark = Backbone.View.extend({
  template: _.template(…),
  render: function() {
    return this;

Backbone is agnostic with respect to your preferred method of HTML templating. Your render function could even munge together an HTML string, or use document.createElement to generate a DOM tree. However, we suggest choosing a nice JavaScript templating library. Mustache.js, Haml-js, and Eco are all fine alternatives. Because Underscore.js is already on the page, _.template is available, and is an excellent choice if you prefer simple interpolated-JavaScript style templates.

Whatever templating strategy you end up with, it's nice if you never have to put strings of HTML in your JavaScript. At DocumentCloud, we use Jammit in order to package up JavaScript templates stored in /app/views as part of our main core.js asset package.

Removes a view from the DOM, and calls stopListening to remove any bound events that the view has listenTo'd.

Uses jQuery's on function to provide declarative callbacks for DOM events within a view. If an events hash is not passed directly, uses as the source. Events are written in the format {"event selector": "callback"}. The callback may be either the name of a method on the view, or a direct function body. Omitting the selector causes the event to be bound to the view's root element (this.el). By default, delegateEvents is called within the View's constructor for you, so if you have a simple events hash, all of your DOM events will always already be connected, and you will never have to call this function yourself.

The events property may also be defined as a function that returns an events hash, to make it easier to programmatically define your events, as well as inherit them from parent views.

Using delegateEvents provides a number of advantages over manually using jQuery to bind events to child elements during render. All attached callbacks are bound to the view before being handed off to jQuery, so when the callbacks are invoked, this continues to refer to the view object. When delegateEvents is run again, perhaps with a different events hash, all callbacks are removed and delegated afresh — useful for views which need to behave differently when in different modes.

A view that displays a document in a search result might look something like this:

var DocumentView = Backbone.View.extend({

  events: {
    "dblclick"                : "open",
    "click .icon.doc"         : "select",
    "contextmenu .icon.doc"   : "showMenu",
    "click .show_notes"       : "toggleNotes",
    "click .title .lock"      : "editAccessLevel",
    "mouseover .title .date"  : "showTooltip"

  render: function() {
    return this;

  open: function() {"viewer_url"));

  select: function() {
    this.model.set({selected: true});



Removes all of the view's delegated events. Useful if you want to disable or remove a view from the DOM temporarily.


Backbone.noConflictvar backbone = Backbone.noConflict();
Returns the Backbone object back to its original value. You can use the return value of Backbone.noConflict() to keep a local reference to Backbone. Useful for embedding Backbone on third-party websites, where you don't want to clobber the existing Backbone.

var localBackbone = Backbone.noConflict();
var model = localBackbone.Model.extend(...);

Backbone.$Backbone.$ = $;
If you have multiple copies of jQuery on the page, or simply want to tell Backbone to use a particular object as its DOM / Ajax library, this is the property for you.


The list of examples that follows, while long, is not exhaustive. If you've worked on an app that uses Backbone, please add it to the wiki page of Backbone apps.

Jérôme Gravel-Niquet has contributed a Todo List application that is bundled in the repository as Backbone example. If you're wondering where to get started with Backbone in general, take a moment to read through the annotated source. The app uses a LocalStorage adapter to transparently save all of your todos within your browser, instead of sending them to a server. Jérôme also has a version hosted at



The DocumentCloud workspace is built on Backbone.js, with Documents, Projects, Notes, and Accounts all as Backbone models and collections. If you're interested in history — both Underscore.js and Backbone.js were originally extracted from the DocumentCloud codebase, and packaged into standalone JS libraries.

DocumentCloud Workspace

USA Today

USA Today takes advantage of the modularity of Backbone's data/model lifecycle — which makes it simple to create, inherit, isolate, and link application objects — to keep the codebase both manageable and efficient. The new website also makes heavy use of the Backbone Router to control the page for both pushState-capable and legacy browsers. Finally, the team took advantage of Backbone's Event module to create a PubSub API that allows third parties and analytics packages to hook into the heart of the app.

USA Today


New Rdio was developed from the ground up with a component based framework based on Backbone.js. Every component on the screen is dynamically loaded and rendered, with data provided by the Rdio API. When changes are pushed, every component can update itself without reloading the page or interrupting the user's music. All of this relies on Backbone's views and models, and all URL routing is handled by Backbone's Router. When data changes are signaled in realtime, Backbone's Events notify the interested components in the data changes. Backbone forms the core of the new, dynamic, realtime Rdio web and desktop applications.


LinkedIn Mobile

LinkedIn used Backbone.js to create its next-generation HTML5 mobile web app. Backbone made it easy to keep the app modular, organized and extensible so that it was possible to program the complexities of LinkedIn's user experience. The app also uses Zepto, Underscore.js, SASS, iScroll, HTML5 LocalStorage and Canvas. The tech team blogged about their experiences using LocalStorage to improve mobile performance.

LinkedIn Mobile


Hulu used Backbone.js to build its next generation online video experience. With Backbone as a foundation, the web interface was rewritten from scratch so that all page content can be loaded dynamically with smooth transitions as you navigate. Backbone makes it easy to move through the app quickly without the reloading of scripts and embedded videos, while also offering models and collections for additional data manipulation support.



Quartz sees itself as a digitally native news outlet for the new global economy. Because Quartz believes in the future of open, cross-platform web applications, they selected Backbone and Underscore to fetch, sort, store, and display content from a custom WordPress API. Although uses responsive design for phone, tablet, and desktop browsers, it also takes advantage of Backbone events and views to render device-specific templates in some cases.



MetaLab used Backbone.js to create Flow, a task management app for teams. The workspace relies on Backbone.js to construct task views, activities, accounts, folders, projects, and tags. You can see the internals under window.Flow.


Gilt Groupe

Gilt Groupe uses Backbone.js to build multiple applications across their family of sites. Gilt's mobile website uses Backbone and Zepto.js to create a blazing-fast shopping experience for users on-the-go, while Gilt Live combines Backbone with WebSockets to display the items that customers are buying in real-time. Gilt's search functionality also uses Backbone to filter and sort products efficiently by moving those actions to the client-side.

Gilt Groupe


NewsBlur is an RSS feed reader and social news network with a fast and responsive UI that feels like a native desktop app. Backbone.js was selected for a major rewrite and transition from spaghetti code because of its powerful yet simple feature set, easy integration, and large community. If you want to poke around under the hood, NewsBlur is also entirely open-source.

Newsblur is the software-as-a-service version of WordPress. It uses Backbone.js Models, Collections, and Views in its Notifications system. Backbone.js was selected because it was easy to fit into the structure of the application, not the other way around. Automattic (the company behind is integrating Backbone.js into the Stats tab and other features throughout the homepage. Notifications


Foursquare is a fun little startup that helps you meet up with friends, discover new places, and save money. Backbone Models are heavily used in the core JavaScript API layer and Views power many popular features like the homepage map and lists.



Bitbucket is a free source code hosting service for Git and Mercurial. Through its models and collections, Backbone.js has proved valuable in supporting Bitbucket's REST API, as well as newer components such as in-line code comments and approvals for pull requests. Mustache templates provide server and client-side rendering, while a custom Google Closure inspired life-cycle for widgets allows Bitbucket to decorate existing DOM trees and insert new ones.



Disqus chose Backbone.js to power the latest version of their commenting widget. Backbone’s small footprint and easy extensibility made it the right choice for Disqus’ distributed web application, which is hosted entirely inside an iframe and served on thousands of large web properties, including IGN, Wired, CNN, MLB, and more.


Khan Academy

Khan Academy is on a mission to provide a free world-class education to anyone anywhere. With thousands of videos, hundreds of JavaScript-driven exercises, and big plans for the future, Khan Academy uses Backbone to keep frontend code modular and organized. User profiles and goal setting are implemented with Backbone, jQuery and Handlebars, and most new feature work is being pushed to the client side, greatly increasing the quality of the API.

Khan Academy


Do is a social productivity app that makes it easy to work on tasks, track projects, and take notes with your team. The web application was built from the ground up to work seamlessly on your smartphone, tablet and computer. The team used Backbone, CoffeeScript and Handlebars to build a full-featured app in record time and rolled their own extensions for complex navigation and model sync support.



IRCCloud is an always-connected IRC client that you use in your browser — often leaving it open all day in a tab. The sleek web interface communicates with an Erlang backend via websockets and the IRCCloud API. It makes heavy use of Backbone.js events, models, views and routing to keep your IRC conversations flowing in real time.



Pitchfork uses Backbone.js to power its site-wide audio player,, location routing, a write-thru page fragment cache, and more. Backbone.js (and Underscore.js) helps the team create clean and modular components, move very quickly, and focus on the site, not the spaghetti.



Spin pulls in the latest news stories from their internal API onto their site using Backbone models and collections, and a custom sync method. Because the music should never stop playing, even as you click through to different "pages", Spin uses a Backbone router for navigation within the site.


Walmart Mobile

Walmart used Backbone.js to create the new version of their mobile web application and created two new frameworks in the process. Thorax provides mixins, inheritable events, as well as model and collection view bindings that integrate directly with Handlebars templates. Lumbar allows the application to be split into modules which can be loaded on demand, and creates platform specific builds for the portions of the web application that are embedded in Walmart's native Android and iOS applications.

Walmart Mobile

Groupon Now!

Groupon Now! helps you find local deals that you can buy and use right now. When first developing the product, the team decided it would be AJAX heavy with smooth transitions between sections instead of full refreshes, but still needed to be fully linkable and shareable. Despite never having used Backbone before, the learning curve was incredibly quick — a prototype was hacked out in an afternoon, and the team was able to ship the product in two weeks. Because the source is minimal and understandable, it was easy to add several Backbone extensions for Groupon Now!: changing the router to handle URLs with querystring parameters, and adding a simple in-memory store for caching repeated requests for the same data.

Groupon Now!


37Signals chose Backbone.js to create the calendar feature of its popular project management software Basecamp. The Basecamp Calendar uses Backbone.js models and views in conjunction with the Eco templating system to present a polished, highly interactive group scheduling interface.

Basecamp Calendar

Slavery Footprint

Slavery Footprint allows consumers to visualize how their consumption habits are connected to modern-day slavery and provides them with an opportunity to have a deeper conversation with the companies that manufacture the goods they purchased. Based in Oakland, California, the Slavery Footprint team works to engage individuals, groups, and businesses to build awareness for and create deployable action against forced labor, human trafficking, and modern-day slavery through online tools, as well as off-line community education and mobilization programs.

Slavery Footprint


Stripe provides an API for accepting credit cards on the web. Stripe's management interface was recently rewritten from scratch in Coffeescript using Backbone.js as the primary framework, Eco for templates, Sass for stylesheets, and Stitch to package everything together as CommonJS modules. The new app uses Stripe's API directly for the majority of its actions; Backbone.js models made it simple to map client-side models to their corresponding RESTful resources.



Airbnb uses Backbone in many of its products. It started with Airbnb Mobile Web (built in six weeks by a team of three) and has since grown to Wish Lists, Match, Search, Communities, Payments, and Internal Tools.



Diaspora is a distributed social network, formed from a number of independently operated pods. You own your personal data, and control with whom you share. All of Diaspora is open-source code, built with Rails and Backbone.js.


SoundCloud Mobile

SoundCloud is the leading sound sharing platform on the internet, and Backbone.js provides the foundation for SoundCloud Mobile. The project uses the public SoundCloud API as a data source (channeled through a nginx proxy), jQuery templates for the rendering, Qunit and PhantomJS for the testing suite. The JS code, templates and CSS are built for the production deployment with various Node.js tools like ready.js, Jake, jsdom. The Backbone.History was modified to support the HTML5 history.pushState. Backbone.sync was extended with an additional SessionStorage based cache layer.

SoundCloud is a place to discover art you'll love. is built on Rails, using Grape to serve a robust JSON API. The main site is a single page app written in Coffeescript and uses Backbone to provide structure around this API. An admin panel and partner CMS have also been extracted into their own API-consuming Backbone projects.


When Pandora redesigned their site in HTML5, they chose Backbone.js to help manage the user interface and interactions. For example, there's a model that represents the "currently playing track", and multiple views that automatically update when the current track changes. The station list is a collection, so that when stations are added or changed, the UI stays up to date.



Inkling is a cross-platform way to publish interactive learning content. Inkling for Web uses Backbone.js to make hundreds of complex books — from student textbooks to travel guides and programming manuals — engaging and accessible on the web. Inkling supports WebGL-enabled 3D graphics, interactive assessments, social sharing, and a system for running practice code right in the book, all within a single page Backbone-driven app. Early on, the team decided to keep the site lightweight by using only Backbone.js and raw JavaScript. The result? Complete source code weighing in at a mere 350kb with feature-parity across the iPad, iPhone and web clients. Give it a try with this excerpt from JavaScript: The Definitive Guide.


Code School

Code School courses teach people about various programming topics like CoffeeScript, CSS, Ruby on Rails, and more. The new Code School course challenge page is built from the ground up on Backbone.js, using everything it has to offer: the router, collections, models, and complex event handling. Before, the page was a mess of jQuery DOM manipulation and manual Ajax calls. Backbone.js helped introduce a new way to think about developing an organized front-end application in Javascript.

Code School


CloudApp is simple file and link sharing for the Mac. Backbone.js powers the web tools which consume the documented API to manage Drops. Data is either pulled manually or pushed by Pusher and fed to Mustache templates for rendering. Check out the annotated source code to see the magic.



SeatGeek's stadium ticket maps were originally developed with Prototype.js. Moving to Backbone.js and jQuery helped organize a lot of the UI code, and the increased structure has made adding features a lot easier. SeatGeek is also in the process of building a mobile interface that will be Backbone.js from top to bottom.



Easel is an in-browser, high fidelity web design tool that integrates with your design and development process. The Easel team uses CoffeeScript, Underscore.js and Backbone.js for their rich visual editor as well as other management functions throughout the site. The structure of Backbone allowed the team to break the complex problem of building a visual editor into manageable components and still move quickly.



Prose is a content editor for GitHub, optimized for managing websites built with Jekyll and Github Pages. Prose is itself implemented as a static Jekyll site, using Backbone.js to render the views and handle the routes, as well as Github.js, a small data abstraction layer for manipulating files directly on Github. Read more in the official introduction post, or take a look at the source code.



Jolicloud is an open and independent platform and operating system that provides music playback, video streaming, photo browsing and document editing — transforming low cost computers into beautiful cloud devices. The new Jolicloud HTML5 app was built from the ground up using Backbone and talks to the Jolicloud Platform, which is based on Node.js. Jolicloud works offline using the HTML5 AppCache, extends Backbone.sync to store data in IndexedDB or localStorage, and communicates with the Joli OS via WebSockets.


Battlefield Play4Free

Battlefield Play4Free is the latest free-to-play first person shooter from the same team that created Battlefield Heroes. The in-game HTML5 front-end for makes heavy use of Backbone's views, models and collections to help keep the code modular and structured.

Battlefield Play4Free


Syllabus is the new live blogging platform used by The Verge and other Vox Media sites. Syllabus uses Backbone on both ends: an editorial dashboard and the live blog page itself. In the back, Backbone is used to provide a single-page experience for uploading, writing, editing and publishing content. On the live blog, Backbone manages fetching a JSON API feed, and updating the infinite-scrolling river of updates with new and revised content.

Syllabus provides a space where photographers, artists and designers freely arrange their visual art on virtual walls. runs on Rails, but does not use much of the traditional stack, as the entire frontend is designed as a single page web app, using Backbone.js, Brunch and CoffeeScript.


Our fellow Knight Foundation News Challenge winners, MapBox, created an open-source map design studio with Backbone.js: TileMill. TileMill lets you manage map layers based on shapefiles and rasters, and edit their appearance directly in the browser with the Carto styling language. Note that the gorgeous MapBox homepage is also a Backbone.js app.



Blossom is a lightweight project management tool for lean teams. Backbone.js is heavily used in combination with CoffeeScript to provide a smooth interaction experience. The app is packaged with Brunch. The RESTful backend is built with Flask on Google App Engine.



Trello is a collaboration tool that organizes your projects into boards. A Trello board holds many lists of cards, which can contain checklists, files and conversations, and may be voted on and organized with labels. Updates on the board happen in real time. The site was built ground up using Backbone.js for all the models, views, and routes.



Cristi Balan and Irina Dumitrascu created Tzigla, a collaborative drawing application where artists make tiles that connect to each other to create surreal drawings. Backbone models help organize the code, routers provide bookmarkable deep links, and the views are rendered with haml.js and Zepto. Tzigla is written in Ruby (Rails) on the backend, and CoffeeScript on the frontend, with Jammit prepackaging the static assets.



Why use Backbone, not [other framework X]?
If your eye hasn't already been caught by the adaptibility and elan on display in the above list of examples, we can get more specific: Backbone.js aims to provide the common foundation that data-rich web applications with ambitious interfaces require — while very deliberately avoiding painting you into a corner by making any decisions that you're better equipped to make yourself.

There's More Than One Way To Do It
It's common for folks just getting started to treat the examples listed on this page as some sort of gospel truth. In fact, Backbone.js is intended to be fairly agnostic about many common patterns in client-side code. For example...

References between Models and Views can be handled several ways. Some people like to have direct pointers, where views correspond 1:1 with models (model.view and view.model). Others prefer to have intermediate "controller" objects that orchestrate the creation and organization of views into a hierarchy. Others still prefer the evented approach, and always fire events instead of calling methods directly. All of these styles work well.

Batch operations on Models are common, but often best handled differently depending on your server-side setup. Some folks don't mind making individual Ajax requests. Others create explicit resources for RESTful batch operations: /notes/batch/destroy?ids=1,2,3,4. Others tunnel REST over JSON, with the creation of "changeset" requests:

    "create":  [array of models to create]
    "update":  [array of models to update]
    "destroy": [array of model ids to destroy]

Feel free to define your own events. Backbone.Events is designed so that you can mix it in to any JavaScript object or prototype. Since you can use any string as an event, it's often handy to bind and trigger your own custom events: model.on("selected:true") or model.on("editing")

Render the UI as you see fit. Backbone is agnostic as to whether you use Underscore templates, Mustache.js, direct DOM manipulation, server-side rendered snippets of HTML, or jQuery UI in your render function. Sometimes you'll create a view for each model ... sometimes you'll have a view that renders thousands of models at once, in a tight loop. Both can be appropriate in the same app, depending on the quantity of data involved, and the complexity of the UI.

Nested Models & Collections
It's common to nest collections inside of models with Backbone. For example, consider a Mailbox model that contains many Message models. One nice pattern for handling this is have a this.messages collection for each mailbox, enabling the lazy-loading of messages, when the mailbox is first opened ... perhaps with MessageList views listening for "add" and "remove" events.

var Mailbox = Backbone.Model.extend({

  initialize: function() {
    this.messages = new Messages;
    this.messages.url = '/mailbox/' + + '/messages';
    this.messages.on("reset", this.updateCounts);



var inbox = new Mailbox;

// And then, when the Inbox is opened:


If you're looking for something more opinionated, there are a number of Backbone plugins that add sophisticated associations among models, available on the wiki.

Backbone doesn't include direct support for nested models and collections or "has many" associations because there are a number of good patterns for modeling structured data on the client side, and Backbone should provide the foundation for implementing any of them. You may want to…

Loading Bootstrapped Models
When your app first loads, it's common to have a set of initial models that you know you're going to need, in order to render the page. Instead of firing an extra AJAX request to fetch them, a nicer pattern is to have their data already bootstrapped into the page. You can then use reset to populate your collections with the initial data. At DocumentCloud, in the ERB template for the workspace, we do something along these lines:

  var accounts = new Backbone.Collection;
  accounts.reset(<%= @accounts.to_json %>);
  var projects = new Backbone.Collection;
  projects.reset(<%= @projects.to_json(:collaborators => true) %>);

You have to escape </ within the JSON string, to prevent javascript injection attacks.

Extending Backbone
Many JavaScript libraries are meant to be insular and self-enclosed, where you interact with them by calling their public API, but never peek inside at the guts. Backbone.js is not that kind of library.

Because it serves as a foundation for your application, you're meant to extend and enhance it in the ways you see fit — the entire source code is annotated to make this easier for you. You'll find that there's very little there apart from core functions, and most of those can be overriden or augmented should you find the need. If you catch yourself adding methods to Backbone.Model.prototype, or creating your own base subclass, don't worry — that's how things are supposed to work.

How does Backbone relate to "traditional" MVC?
Different implementations of the Model-View-Controller pattern tend to disagree about the definition of a controller. If it helps any, in Backbone, the View class can also be thought of as a kind of controller, dispatching events that originate from the UI, with the HTML template serving as the true view. We call it a View because it represents a logical chunk of UI, responsible for the contents of a single DOM element.

Comparing the overall structure of Backbone to a server-side MVC framework like Rails, the pieces line up like so:

Binding "this"
Perhaps the single most common JavaScript "gotcha" is the fact that when you pass a function as a callback, its value for this is lost. With Backbone, when dealing with events and callbacks, you'll often find it useful to rely on _.bind and _.bindAll from Underscore.js.

When binding callbacks to Backbone events, you can choose to pass an optional third argument to specify the this that will be used when the callback is later invoked:

var MessageList = Backbone.View.extend({

  initialize: function() {
    var messages = this.collection;
    messages.on("reset", this.render, this);
    messages.on("add", this.addMessage, this);
    messages.on("remove", this.removeMessage, this);


// Later, in the app...


Working with Rails
Backbone.js was originally extracted from a Rails application; getting your client-side (Backbone) Models to sync correctly with your server-side (Rails) Models is painless, but there are still a few things to be aware of.

By default, Rails versions prior to 3.1 add an extra layer of wrapping around the JSON representation of models. You can disable this wrapping by setting:

ActiveRecord::Base.include_root_in_json = false

... in your configuration. Otherwise, override parse to pull model attributes out of the wrapper. Similarly, Backbone PUTs and POSTs direct JSON representations of models, where by default Rails expects namespaced attributes. You can have your controllers filter attributes directly from params, or you can override toJSON in Backbone to add the extra wrapping Rails expects.

Change Log

1.0.0March 20, 2013Diff
0.9.10Jan. 15, 2013Diff
0.9.9Dec. 13, 2012DiffDocs
0.9.2March 21, 2012DiffDocs
0.9.1Feb. 2, 2012DiffDocs
0.9.0Jan. 30, 2012DiffDocs

0.5.3August 9, 2011DiffDocs
A View's events property may now be defined as a function, as well as an object literal, making it easier to programmatically define and inherit events. groupBy is now proxied from Underscore as a method on Collections. If the server has already rendered everything on page load, pass Backbone.history.start({silent: true}) to prevent the initial route from triggering. Bugfix for pushState with encoded URLs.

0.5.2July 26, 2011DiffDocs
The bind function, can now take an optional third argument, to specify the this of the callback function. Multiple models with the same id are now allowed in a collection. Fixed a bug where calling .fetch(jQueryOptions) could cause an incorrect URL to be serialized. Fixed a brief extra route fire before redirect, when degrading from pushState.

0.5.1July 5, 2011DiffDocs
Cleanups from the 0.5.0 release, to wit: improved transparent upgrades from hash-based URLs to pushState, and vice-versa. Fixed inconsistency with non-modified attributes being passed to Model#initialize. Reverted a 0.5.0 change that would strip leading hashbangs from routes. Added contains as an alias for includes.

0.5.0July 1, 2011DiffDocs
A large number of tiny tweaks and micro bugfixes, best viewed by looking at the commit diff. HTML5 pushState support, enabled by opting-in with: Backbone.history.start({pushState: true}). Controller was renamed to Router, for clarity. Collection#refresh was renamed to Collection#reset to emphasize its ability to both reset the collection with new models, as well as empty out the collection when used with no parameters. saveLocation was replaced with navigate. RESTful persistence methods (save, fetch, etc.) now return the jQuery deferred object for further success/error chaining and general convenience. Improved XSS escaping for Model#escape. Added a urlRoot option to allow specifying RESTful urls without the use of a collection. An error is thrown if Backbone.history.start is called multiple times. Collection#create now validates before initializing the new model. view.el can now be a jQuery string lookup. Backbone Views can now also take an attributes parameter. Model#defaults can now be a function as well as a literal attributes object.

0.3.3Dec 1, 2010DiffDocs
Backbone.js now supports Zepto, alongside jQuery, as a framework for DOM manipulation and Ajax support. Implemented Model#escape, to efficiently handle attributes intended for HTML interpolation. When trying to persist a model, failed requests will now trigger an "error" event. The ubiquitous options argument is now passed as the final argument to all "change" events.

0.3.2Nov 23, 2010DiffDocs
Bugfix for IE7 + iframe-based "hashchange" events. sync may now be overridden on a per-model, or per-collection basis. Fixed recursion error when calling save with no changed attributes, within a "change" event.

0.3.1Nov 15, 2010DiffDocs
All "add" and "remove" events are now sent through the model, so that views can listen for them without having to know about the collection. Added a remove method to Backbone.View. toJSON is no longer called at all for 'read' and 'delete' requests. Backbone routes are now able to load empty URL fragments.

0.3.0Nov 9, 2010DiffDocs
Backbone now has Controllers and History, for doing client-side routing based on URL fragments. Added emulateHTTP to provide support for legacy servers that don't do PUT and DELETE. Added emulateJSON for servers that can't accept application/json encoded requests. Added Model#clear, which removes all attributes from a model. All Backbone classes may now be seamlessly inherited by CoffeeScript classes.

0.2.0Oct 25, 2010DiffDocs
Instead of requiring server responses to be namespaced under a model key, now you can define your own parse method to convert responses into attributes for Models and Collections. The old handleEvents function is now named delegateEvents, and is automatically called as part of the View's constructor. Added a toJSON function to Collections. Added Underscore's chain to Collections.

0.1.2Oct 19, 2010DiffDocs
Added a Model#fetch method for refreshing the attributes of single model from the server. An error callback may now be passed to set and save as an option, which will be invoked if validation fails, overriding the "error" event. You can now tell backbone to use the _method hack instead of HTTP methods by setting Backbone.emulateHTTP = true. Existing Model and Collection data is no longer sent up unnecessarily with GET and DELETE requests. Added a rake lint task. Backbone is now published as an NPM module.

0.1.1Oct 14, 2010DiffDocs
Added a convention for initialize functions to be called upon instance construction, if defined. Documentation tweaks.

0.1.0Oct 13, 2010Docs
Initial Backbone release.

A DocumentCloud Project