The user interacts with the View, but the View must communicate the actions to the Controller. The Controller may update the Model, but it isn't required with every/any change.
The description I am providing is based on my personal experience with the .NET implementation of MVC. Your implementation can be different.
The Controller is where actions are processed, basically a business layer. A simple controller will do nothing more than get the data from the Model to feed to the View. A complicated Controller will perform all sorts of actions, up to security management, authentication, authorization, registration, and possibly many other things.
The View should only be responsible for displaying the information in a fashion that the user can understand. There can be some cross over here with both the Controller and the Model as things like Single Page Applications (SPAs) will have data validation feedback for the user. Any other cross overs are heavily frowned upon.
The Model deals with data. This includes validation of data (where applicable). Data storage and retrieval is also handled in this layer.
UPDATE
There seems to be some confusion surrounding who does what when. I included two different overviews of the MVC architectures because they are similar, but not the same. There is room for either interpretation. Possibly, many more. The descriptions above are my interpretation of MVC from multiple sources, including my own experience building applications using this methodology. Hopefully, this update will help to clear up some of this confusion.
MVC is an attempt to build a Separation of Concerns design pattern for software development. It has primarily been implemented in web based applications (to my knowledge).
The View handles all of the user interaction. If your user clicks on a button, the View determines if the click is a user interface interaction or something that is beyond its concern (a Controller interaction). If the button does something like copy values from one field to another, your implementation will determine if that is a View concern or a Controller concern. You will most likely only have this blurring of concerns when dealing with a Single Page Application (SPA).
The Controller is where your actions are processed. The View has communicated the user decided to change values for some fields. The Controller may perform validation on that data or it may be handled by the Model. Again this is implementation dependent. If the Controller has security features, it may determine that the user doesn't have sufficient privileges to perform the action. It would reject the changes and update the View accordingly. The Controller also determines what data to retrieve from the Model, how to package it, and update the View with that data.
The Model determines how and where to store data. It may also perform validation of that data before storing it (it should do this because people will bypass the View on occasion).
Wikipedia has an article on MVC.
- A model notifies its associated view/views and controllers when there has been a change in its state. This notification allows views to update their presentation, and the controllers to change the available set of commands. In some cases an MVC implementation might instead be "passive," so that other components must poll the model for updates rather than being notified.
- A view is told by the controller all the information it needs for generating an output representation to the user. It can also provide generic mechanisms to inform the controller of user input.
- A controller can send commands to the model to update the model's state (e.g., editing a document). It can also send commands to its associated view to change the view's presentation of the model (e.g., by scrolling through a document).
From Microsoft's Overview of MVC.
Models. Model objects are the parts of the application that implement the logic for the application's data domain. Often, model objects retrieve and store model state in a database. For example, a Product object might retrieve information from a database, operate on it, and then write updated information back to a Products table in a SQL Server database.
In small applications, the model is often a conceptual separation instead of a physical one. For example, if the application only reads a dataset and sends it to the view, the application does not have a physical model layer and associated classes. In that case, the dataset takes on the role of a model object.
Views. Views are the components that display the application's user interface (UI). Typically, this UI is created from the model data. An example would be an edit view of a Products table that displays text boxes, drop-down lists, and check boxes based on the current state of a Product object.
Controllers. Controllers are the components that handle user interaction, work with the model, and ultimately select a view to render that displays UI. In an MVC application, the view only displays information; the controller handles and responds to user input and interaction. For example, the controller handles query-string values, and passes these values to the model, which in turn might use these values to query the database.
The ViewModel object is not what gets stored in a database table, generally. It's the individual items in the ViewModel object that get stored. Each of those items already has an ID.
For example:
public class InvoiceViewModel
{
public Customer Customer { get; set; }
public Address BillingAddress { get; set; }
public Address ShippingAddress { get; set; }
public List<InvoiceLineItem> Items { get; set; }
}
Since there's no single table in the database that corresponds to an InvoiceViewModel, there is no ID for an InvoiceViewModel object.
Of course, you can always use the InvoiceID as the id for this particular ViewModel. InvoiceID is handy, because that's what this object ultimately represents. But I could see having a ViewModel object that doesn't correspond to any particular ID in the database.
Best Answer
I'll speak specifically to iOS as it has its own concepts and some language features make its way of doing things a good fit. Often in iOS the Controller - Model layer interactions are based on some variant of Observer pattern (key-value observing (KVO) but also
NSNotification). I find the model layer often doesn't engage in the same type of interactive back and forth that happens in the View - Controller layer interface (think the complexity of a dynamic, fully-featured table view), so delegation is not the most natural fit. Usually something (user input, a network request, etc) changes the model, and the controller responds accordingly by updating the view. But delegation is still used in long-running, dynamic interactions such as a table view that displays the contents of a frequently updating data set (seeNSFetchedResultsController).As to your question of having the controller be the delegate of your view and your model, there's no inherent problem other than you'll degrade the cohesion of your controller, and it sometimes becomes a good choice to split the controller layer into view controllers and data controllers. Using KVO or NSNotification to observe changes in the model layer is usually straightforward enough that you won't find a pressing need to extract a new class.