I just started a new GWT project for a client and I'm interested in hearing people's experience with various GWT MVC architectures. On a recent project, I used both GXT MVC, as well as a custom messaging solution (based on Appcelerator's MQ). GXT MVC worked OK, but it seemed like overkill for GWT and was hard to make work with browser history. I've heard of PureMVC and GWTiger, but never used them. Our custom MQ solution worked pretty well, but made it difficult to test components with JUnit.
In addition, I've heard that Google Wave (a GWT application) is written using a Model-View-Presenter pattern. A sample MVP application was recently published, but looking at the code, it doesn't seem that intuitive.
If you were building a new GWT application, which architecture would you use? What are the pros and cons of your choice?
Thanks,
Matt
It's worth noting that google has finally written out a tutorial for designing using the mvp architecture. It clarifies a lot of the elements from the google i/o talk listed above. Take a looK: https://developers.google.com/web-toolkit/articles/mvp-architecture
I am glad this question has been asked, because GWT desperatley needs a rails-like way of structuring an application. A simple approach based on best practices that will work for 90 % of all use-cases and enables super easy testability.
In the past years I have been using my own implementation of MVP with a very passive view that enslaves itself to whatever the Presenter tells him to do.
My solution consisted of the following:
an interface per widget defining the methods to control the visual appearance
an implementing class that can be a Composite or use an external widget library
a central Presenter for a screen that hosts N views that are made up of M widgets
a central model per screen that holds the data associated with the current visual appearance
generic listener classes like "SourcesAddEvents[CustomerDTO]" (the editor does not like the real symbols for java generics here, so I used thoe brackets), because otherwise you will have lots of the same interfaces who just differ by the type
The Views get a reference to the presenter as their constructor parameter, so they can initialize their events with the presenter. The presenter will handles those events and notify other widgets/views and or call gwt-rpc that on success puts its result into the model. The model has a typical "Property[List[String]] names = ...." property change listener mechanism that is registered with the presenter so that the update of a model by an gwt-rpc request goes to all views/widgets that are interested.
With this appraoch I have gotten very easy testability with EasyMock for my AsynInterfaces. I also got the ability to easily exchange the implementation of a view/widget, because all I had to rewrite was the code that notified the presenter of some event - regardless of the underlying widget (Button, Links, etc).
Problems with my approach:
My current implementation makes it hard to synchronize data-values between the central models of different screens. Say you have a screen that displays a set of categories and another screen that lets you add/edit those items. Currently it is very hard to propagate those change events across the boundaries of the screens, because the values are cached in those models and it is hard to find our whether some things are dirty (would have been easy in a traditional web1.0-html-dumb-terminal kind of scenario with serverside declarative caching).
The constructor parameters of the views enable super-easy testing, but without a solid Dependency-Injection framework, one will have some UGLY factory/setup code inside "onModuleLoad()". At the time I started this, I was not aware of Google GIN, so when I refactor my app, I will use that to get rid of this boilerplate. An interesting example here is the "HigherLower" game inside the GIN-Trunk.
I did not get History right the first time, so it is hard to navigate from one part of my app to another. My approach is not aware of History, which is a serious downturn.
My Solutions to those problems:
Use GIN to remove the setup boilerplate that is hard to maintain
While moving from Gwt-Ext to GXT, use its MVC framework as an EventBus to attach/detach modular screens, to avoid the caching/synchronization issues
Think of some kind of "Place"-Abstraction like Ray Ryan described in his talk at I/O 09, which bridges the Event-Gap between GXT-MVC and GWTs-Hitory approach
Use MVP for widgets to isolate data access
Summary:
I dont think one can use a single "MVP" approach for an entire app. One definetly needs history for app-navigation, a eventbus like GXT-MVC to attach/detach screens, and MVP to enable easy testing of data access for widgets.
I therefore propose a layered approach that combines these three elements, since I believe that the "one-event-mvp-system"-solution wont work. Navigation/Screen-Attaching/Data-Access are three separate concerns and I will refactor my app (move to GXT) in the following months to utilize all three event-frameworks for each concerns separately (best tool for the job). All three elements need not be aware of each other. I do know that my solution only applies for GXT-projects.
When writing big GWT apps, I feel like I have to reinvent something like Spring-MVC on the client, which really sucks, because it takes a lot of time and brain-power to spit out something elegant as Spring MVC. GWT needs an app framework much more than those tiny little JS-optimizations that the compiler-guys work so hard on.
Here is a recent Google IO presentation on architecting your GWT application.
Enjoy.
-JP
If you're interested in using the MVP architecture, you might want to take a look at GWTP: http://code.google.com/p/gwt-platform/ . It's an open source MVP framework I'm working on, that supports many nice features of GWT, including code splitting and history management, with a simple annotation-based API. It is quite recent, but is already being used in a number of projects.
You should have a look at GWT Portlets. We developed the GWT Portlets Framework while working on a large HR Portal application and it is now free and open source. From the GWT Portlets website (hosted on Google code):
The programming model is somewhat similar to writing JSR168 portlets for a portal server (Liferay, JBoss Portal etc.). The "portal" is your application built using the GWT Portlets framework as a library. Application functionality is developed as loosely coupled Portlets each with an optional server side DataProvider.
Every Portlet knows how to externalize its state into a serializable PortletFactory subclass (momento / DTO / factory pattern) making important functionality possible:
CRUD operations are handled by a single GWT RPC for all Portlets
The layout of Portlets on a "page" can be represented as a tree of WidgetFactory's (an interface implemented by PortletFactory)
Trees of WidgetFactory's can be serialized and marshalled to/from XML on the server, to store GUI layouts (or "pages") in XML page files
Other important features of the framework are listed below:
Pages can be edited in the browser at runtime (by developers and/or users) using the framework layout editor
Portlets are positioned absolutely so can use scrolling regions
Portlets are configurable, indicate when they are busy loading for automatic "loading spinner" display and can be maximized
Themed widgets including a styled dialog box, a CSS styled button replacement, small toolbuttons and a HTML template driven menu
GWT Portlets is implemented in Java code and does not wrap any external Javascript libraries. It does not impose any server side framework (e.g. Spring or J2EE) but is designed to work well in conjunction with such frameworks.
Related
Considering the most popular MVC/MVVM client-side patterns (like Knockout.js, Angular.js, Ember.js, and others), I have one great doubt:
Also considering the modeling redundance in both sides, what is the advantages and disvantages to use those client-side patterns with MVC server-side patterns?
I struggled with how to answer this question... hopefully this helps, even if it is in a round-about way.
While some of the pros/cons have already been stated, I think the best rundown is in this answer.
For me, the biggest advantage to using client-side logic is the rich UI aspect.
But the key part of your question seems to be "model redundancy" (I'd call it duplicated logic, or at least having potential for duplicated logic). In my opinion, that is a problem which may exist independently of the pros/cons in the previous link.
So first of all, I think that the decision of whether or not to use a client-side framework should be made based on the well-documented pros and cons. Once that decision is made, the associated problems can be solved.
Lets assume you are using some sort of server-side framework/platform, as well as a client-side framework to provide a little bit of UI interactivity. Now there is a problem with where to put the model logic: on the client, server, or both.
One way to solve the problem is to define your model logic in only the client or the server. Then you have no code duplication, but it affects some of the higher-level pros/cons.
For example, if your model logic is 100% server-side, you lose some of the interactive part of the UI. Or, you are constantly throwing the model to/from the server, which will have a few cons.
If your model logic is 100% client-side, you could suffer performance problems, depending on the size of your view / model. This is one of the reasons Twitter is moving to a server-side processing model.
Then there is "both"... having model logic exist in both the client and the server. I think this is the best solution, as long as no logic is duplicated.
For example, on a shopping cart page, you may recalculate the cost of an order based on the price of a product, and a user-editable quantity box. I think this logic should only exist on the client. Other model properties that do not change once loaded are probably fine hosted on the server.
There's a lot of gray area here... I struggle with putting all the eggs in one basket. For example, choosing a client-side framework, creating a lot of client-side logic, and then [hypothetically] running into problems with performance, browser support, or something like that. Now you may want to tweak a page or two for performance (like move it server-side, a la Twitter). But I think being smart about how you structure your code will help mitigate that issue. If your code is maintainable and clean, moving logic from client to server won't be difficult.
The advantage is that the client side patterns are applicable at the client where the server has no direct reach. If you're building a rich, interactive HTML UI then use client side MVVM. Server side MVC may still be relevant in that case for delivering appropriate content to the client. For example, the ASP.NET WebAPI is a framework for creating HTTP APIs which has a similar controller architecture to the ASP.NET MVC framework. The API implemented with this framework may be called by client side code resulting in MVC on the server side and MVVM on the client side. Normally, when using MVC server side and MVVM client side, the responsibilities of the respective sides are very different and thus there is no redundancy.
The fact you an incorporate a MVVM model into an already implemented MVC framework is also a great thing, we recently added knockout to some new project pages to fit with in an already outdated MVC framework (old pages, not the framework itself).
I think MVVM is fantastic as the above answer states it provides an exceptional user experience with extremely fast response times, you can hide your validation calls in the backround with out slowing them down and its intuitive.
The pain however is that it is VERY hard to unit test and you can get some extremely LARGE javascript files, also the extra coding we've had to do as our legacy systems still run on IE6 is ridiculous.
But MVVM and MVC don't have to be used exclusively on there own, we use both. But having 3 levels of validation is something that still bugs me.
advantages
This can rock.
disvantages
You can screw it.
Seriously. Making use of transporting part of the frontend logic into the browser can boost your application development why you keep more strict data-processing encapsulated on server-side.
This is basically layering. Two layers, the one above talks with the one below and vice-versa:
[client] <--> [server]
You normally exchange value objects in a lightweight serialization format like Json between the two.
This can fairly well map what users expect in a useful structure while domain objects on server-side could not be that detailed.
However, the real power will be if the server-side is not in written in javascript at some certain point because I think you can not create well domain objects there. Consider Scala (or something similar expressive) then if you run into that issue.
Ten months later after this question, I have used the both patterns inside the same application.
The only problem was the need to map the models twice.
MVC (ASP.NET MVC 4 Web API)
The most important resource was the routes.
Models were created to database interactions and as arguments for
controllers' actions.
Controllers were created to manipulate the API
requisitions and to render the views.
Views were not modeled with
server-side models, but all the resources of Partial Views and
Sections.
MVVM (Knockout.js)
Models were created with the same properties as the server-side models.
Views were binded with models' properties, and decreased a lot of the views' size.
View-models were created with the values provided from API methods.
Overall, the MVC combination with MVVM were very useful, but it needed a big expertise and knowledge. Patience is required too, because you need to think about the responsibilites of each application layer.
My project's domain is of eTendering. And we are planing to use Spring and Hibernate in the architecture and in presentation Spring MVC but we want look and feel like silverlight of .Net or Flex of adobe in short we are planning for Ajax framework in presentaion. So SpringMVC will be worthfull? I have seen wicket and openxava but I am still confused so please suggest correct option in terms of my domain's complexity.
Depends on your needs:
Spring MVC - this option will leave you with writing AJAX on your own (via JQuery for instance) along with HTML and so forth. It might be tedious, but you're controlling everything. It's also up to you what will be the quality of your resulting HTML and how good it will be indexed by SEO.
JSF - this option provides you with a number of components and allows you to create AJAX based forms and handles it out-of-box. But when it comes to writing custom components, it won't that easy as just writing JS/JQuery on your own. Also it generates not that pretty HTML which is not of that good quality and you might be less indexed by SEOs.
GWT - this choice would mean that you don't write JS, instead you write logic in Java and then it transforms to JS. From one hand it will provide you with good-looking AJAX-based app where you don't need to write JS, from the other hand it's a) pretty complicated to write really good-looking apps UI b) it will add another step in your development cycle (it takes pretty much time when you generate that JS) c) it almost won't allow your pages to be indexed. Also, if it comes to such derivatives as SmartGWT, they provide a large set of cool-looking components, but they always require money for support.
Vaadin - this is AJAX based framework that partially generates Java to JS, but it also sends requests go server for logic execution (of course in GWT this happens as well, but not that often, GWT tries to execute logic on client). It takes less time to compile sources to JS and it's also almost impossible to make pages being indexed.
ZK - another AJAX based framework. unlike other frameworks that allow you to work with only one pattern, it can work in MVC, MVP, MVVM modes. It doesn't compile Java sources to JS thus all the requests go to the server (I've heard about internal company's benchmarks that showed it was actually faster than GWT, but I think it depends on your processing logic). SEO won't make it with ZK at all, but it's possible to include ZK components into JSP pages (though this functionality is not free) which makes it possible to kill both birds. It will be not trivial to write your own components with ZK, but it has a wide range of ready-to-use components.
So you should consider several things: SEO, money you can give for the framework, how much AJAX you need and do you want to write it on your own, etc.
Also pay attention to those patterns I mentioned: MVP for instance is suitable for complicated interfaces and is supported by GWT, Vaadin, ZK. MVVM is very cool because of its binding and is supported by JSF and ZK.
I was reading about JSF that it's a UI framework and provides some UI components. But how is it better or different from number of components that are available from jQueryUI, AngularJS, React, Vue.js, Svelte, ExtJS, or even plain HTML, CSS and JavaScript.
Why should someone learn JSF?
JSF to plain JSP/Servlet/HTML/CSS/JS is like as jQuery to plain JS: do more with less code. To take PrimeFaces (jQuery + jQuery UI based) as an example, browse through its showcase to see complete code examples. BootsFaces (jQuery + Bootstrap UI based) has also a showcase with complete code examples. If you study those examples closely, then you'll see that you basically need a simple Javabean class as model and a XHTML file as view.
Note that you should not see JSF as replacement of alone HTML/CSS/JS, you should also take the server side part into account (specifically: JSP/Servlet). JSF removes the need of all the boilerplate of gathering HTTP request parameters, converting/validating them, updating the model values, executing the right Java method to do the business stuff and generating the HTML/CSS/JS boilerplate code. With JSF you basically end up with a XHTML page as view definition and a Javabean class as model definition. This greatly speeds up development.
As with every component based web MVC framework, you have in JSF less fine-grained control over the rendered HTML/CSS/JS. Adding custom JS code isn't that easy as you have to take the JSF view state in the server side into account as well (e.g. enabling a disabled button in JS side won't enable the button in JSF side, which is in turn a huge security advantage). If that is however a major showstopper, then rather look for an action based web MVC framework like Spring MVC. You'll only take into account that you have to write all that HTML/CSS/JS code (and prevention against XSS, CSRF and DOM-manipulation!) yourself. Also if you fall back from Facelets to JSP, you'll miss advanced templating capabilities as well.
On the other hand, if you have a big JSP/Servlet/HTML/CSS/JS/jQuery based website and you'd like to refactor the repeated JSP/Servlet/HTML/CSS/JS/jQuery boilerplate code into reusable components, then one of the solutions would be JSF. Custom templates, tagfiles and components can aid in this. In that perspective, JSF stands above JSP/Servlet/HTML/CSS/JS/jQuery (and that's also why it's pretty important to understand those basics before diving into JSF).
You can find a real world kickoff JSF based project here: Java EE Kickoff App. You'll see that it contains next to JSF as good HTML5, CSS3 and jQuery.
See also:
Difference between Request MVC and Component MVC
Difference between JSP, Servlet and JSF
What are the main disadvantages of JSF 2.0?
Is it possible to use JSF+Facelets with HTML 4/5?
When to use <ui:include>, tag files, composite components and/or custom components?
JSF was created to make it so that java shops didn't have to learn stuff like jQuery and build complex JS but instead focus on a purely Java stack. In a world where time is money and lots of places already focusing on Java development, one less language/piece in the stack makes training and maintaining faster and thus cheaper.
I'll add that JavaScript is easy to become a maintenance nightmare on large teams, especially if some of the developers on the project are not highly web savvy.
With Javascript and frameworks such as jQuery you have full flexibility and full control . With ext's etc you lose much control and must adapt to the framework. With JSF you totally lose control and must totally adapt to the framework. You're invoked in lifecycles etc. and finally you have no control when the call to the server can be made and where not. If you are to do something considered 'special', you're in very hard position. And in JSF world even such basic things as multicolumn table sort or fields where you can type only limited set of characters (such as number field) are considered 'special'.
However, the more flexibility you have, the more errors or bad practices you can made. High flexibility works only with highly intelligent programmers, others will turn the project into unmanagable nightmare.
But, with JSF and its limited flexibility, there's always only a few (or even only one) correct way to do something. You are very limited, you can't make shortcuts, you must write more XML etc. - but when adapting to standard, there's better control on the code the unexperienced or low-skilled programmers will produce. As a result, big corporations love JSF because it is 'safer' for them.
When I moved from GWT to JSF, I was shocked, how many things, that was natural to me, was considered highly untypical and how much simple things were so hard to achieve. What's more, even making the smallest changes, such as adding ':' sign after label, which in GWT/jQuery powered app would be changing one function generating label, required changing dozens of files with localized properties, which wasn't even considered by anyone except me strange...
The benefits of using JSF are not only in generating xhtml + css + js. Sometimes JSF imposes a restriction on the markup you can generate, like any component based framework. But JSF is not just for that, its lifecyle helps greately. After validating the input it can update the model and sync your server side beans without any effort. you just say "whatever the user types here, check if it's a number, if yes then store it in the property YY in object XX" and JSF will do all that.
So yes, you can still use JQuery, JS, etc. But JSF provides many benefits when it comes to writing server side code and saves you from a lot of boiler plate.
I strongly disagree that jsf adds anything. It only adds overhead. Doing ui stuff on the server is the most ridiculous thing ive ever heard. And javascript on large teams works great - its called reusing code.
Just wrap the jquery in some jsp tags, thats all you need and youre done, and dont endure the.shackles and scalability issues with.jsf and richfaces.
Having worked with JSF, Spring MVC, Struts, Grails, JQuery, and ExtJS my opinion is that Grails + ExtJS is one powerful combination.
I would pick Grails over JSF any day. I like the completeness of ExtJS as the client side framework and library, but it comes with a steeper learning curve than JQuery.
Here are the biggest differences between jQuery & JSF:
no MVC architecture
no state control (store date in session or conversation, auto-clean up, etc.)
no (default) validation library
no templating library
no advanced navigation/routing
client side
jQuery was never intended to be used as a full stack webframework. It was more intended for replacing low-level JS code so that writing JS becomes easier and more powerfull in less lines of code.
And it should thus mostly be used to add behaviour on HTML elements.
Having used ExtJS framework for a large web application, I know how easy it is to use. The ExtJS (Schena) is best suited for (Oracle 11g) database interactions in MVC architecture. The View was for the visual / user interactions. The controller specified the 'processing' and the triggers that needed to be used form the PLSQL packages (the API for the CRUD, SQL select queries etc.). The Model and the store files were used to 'map' the data items to the Viewer / inputs.
ExtJS is not suitable for non database intensive web interfaces - where Angular JS may be a better fit.
When you start a new web application, which pattern are you choosing between MVC and MVP and why?
(This answer is specific to web applications. For regular GUIs, see What are MVP and MVC and what is the difference?.)
Traditional MVC for GUI applications
This isn't really relevant to web applications, but here's how MVC traditionally worked in GUI applications:
The model contained the business objects.
The controller responded to UI interactions, and forwarded them to the model.
The view "subscribed" to the model, and updated itself whenever the model changed.
With this approach, you can have (1) multiple ways to update a given piece of data, and (2) multiple ways to view the same data. But you don't have to let every controller know about every view, or vice versa—everybody can just talk to the model.
MVC on the server
Rails, Django and other server-side frameworks all tend to use a particular version of MVC.
The model provides approximately 1 class per database table, and contains most of the business logic.
The view contains the actual HTML for the site, and as little code as possible. Basically, it's just templates.
The controller responds to HTTP requests, processes parameters, looks up model objects, and passes values to the view.
This seems to work very well for server-based web applications, and I've been very happy with it.
MVP on the client
However, if most of your code is written in JavaScript and runs in the web browser, you'll find lots of people using MVP these days. In this case, the roles are a bit different:
The model still contains all the basic entities of your business domain.
The view is a layer of fairly dumb widgets with little logic.
The presenter installs event handlers on the view widgets, it responds to events and it updates the model. In the other direction, the presenter listens for changes to the model, and when those changes occur, it updates the view widgets. So the presenter is a bidirectional pipeline between the model and the view, which never interact directly.
This model is popular because you can easily remove the view layer and write unit tests against the presenter and model. It's also much better suited to interactive applications where everything is updated constantly, as opposed to server applications where you deal with discrete requests and responses.
Here's some background reading:
Martin Fowler's encyclopedic summary of MVC, MVP and related approaches. There's a lot of good history here.
Martin Fowler's description of "Passive View", a variation of MVP.
Google's MVP + event bus
This is a new approach, described in this video from the Google AdWords team. It's designed to work well with caching, offline HTML 5 applications, and sophisticated client-side toolkits like GWT. It's based on the following observations:
Anything might need to happen asynchronously, so design everything to be asynchronous from the very beginning.
Testing browser-based views is much slower than testing models and presenters.
Your real model data lives on the server, but you may have a local cache or an offline HTML 5 database.
In this approach:
The view is very dumb, and you can replace it with mock objects when running unit tests.
The model objects are just simple containers for data, with no real logic. You may have multiple model objects representing the same entity.
The presenter listens to events from the view. Whenever it needs to update or read from the model, it sends an asynchronous message to the server (or to a local caching service). The server responds by sending events to the "event bus". These events contain copies of the model objects. The event bus passes these events back to the various presenters, which update the attached views.
So this architecture is inherently asynchronous, it's easy to test, and it doesn't require major changes if you want to write an HTML 5 offline application. I haven't used it yet, but it's next on my list of things to try. :-)
Both MVP and MVC make sense and allow to separate logic from display.
I would choose MVC because it's widely used in web development these days (Rails, .NET MVC which is used for SO) so my application will be more easily maintainable by someone else. It is also -to me- cleaner (less "power" given to the view), but this is subjective.
Another alternative is MTV, Model-Template-View which Django uses.
Seems like every project I'm on uses a Model View Controller architecture, and that's how I roll my own projects. Is there an alternative? How else would one create an application that has persistent storage and a user interface?
MVC has been around for a while. It's a time tested and proven pattern. Many frameworks leverage the MVC Pattern. Martin Fowler has deconstructed the MVC into: Supervising Presenter and Passive View.
Architect Christopher Alexander said it best:
Each pattern describes a problem which
occurs over and over again in our
environment and then describes the
core of the solution to that problem,
in such a way that you can use this
solution a million times over, without
ever doing it the same way twice.
I'm not sure why you would want to move from MVC. Is there a problem you are encountering that MVC does not eloquently solve? To give you a better answer we need to know more about your problem domain.
Things to take into account when considering patterns/architecture: If you are building something with a Myspace type architecture you'll need a robust architecture (MVC). If you are creating a simple crud interface via the web - almost anything will do.
For .Net Web forms (I am assuming web, since you didn't say thick or web client) which is not MVC, it was a nightmare maintaining them. Web Forms applications that lived more that a couple years tended to become big balls of mud. Even then developers discovered ways to use MVC with web forms.
Ironically, the lack of MVC architecture in ASP.NET web forms was one of the driving complaints that lead to the development of ASP.Net MVC framework.
From experience if you don't use some MVCesk approach, your solutions become hard to maintain and bloated. These applications will die a slow painful death.
If your solutions are small one-off projects, the by all means throw something together. Heck there are tools that will generate everything from the screens to the data access layer. Whatever works to get the job done.
Classic CRUD apps built using tools like VB6 and Delphi have user interfaces, persistent storage and don't use MVC. Most of them used data aware controls linked directly to database fields
Couple of links comparing various MV* patterns which might be useful:WPF patterns : MVC, MVP or MVVM or…? & MVC, MVP and MVVM
Look into MVP model view presenter.
User interface is View and an application will always have a model and the bridge between the two is Controller. The whole MVC is nothing special as this is how the things will be always.
At the most you can get rid of Controller and have your view talk to your model but you loose the flexibility.
I've developed an alternative to ASP.NET MVC. You get the same loose coupling and separation of concerns but the difference is in how you go about building your projects.
I have a couple of videos on my blog, source code for the framework, a sample project and a few VS.NET add-ins (New Project item, New Builder and New View).
Builder for ASP.NET
Some key differentiating Features are
1. Templates are just html - no code mixed with templates
2. Templates are thus reusable across views and Web site designers can design templates in their design tool of choice
3. Strongly typed code (no ViewData and stuff) so you get intillisense, compile time checking, F12 navigation etc.
4. You build pages as compositions of views rather than an inside-out approach
5. View can be treated as "real" classes.
6. Everything is complied so no run-time compilation
Quite a few other differentiating factors as well.
In theory :
MVC is a proved technology and yada-yada-yada, and it is ideal for websites.
But in a real case:
A serious project that use MVC required a framework, hence you are following a framework with all their limiting and restrictions. So, at this point, the specific implementation of MVC is rule and not a simple "guideline".
Also MVC fail miserably for websites when it is about to connect to the model other than simple POST/GET, it fail with xml asynchronism and it fail with ajax. (*)
(*) exist some patch but a design must be clear and functional and if you need to "patch it" then it is neither clear nor functional.
Personally i think that the main problem with MVC is that it put so much effort in the Controller, where most project use not more that 5 lines for the controller part.
ps: Most "well done" MVC projects are 3-tier project.
ps2: MVC excluding some framework, is just a buzzterm, IS NOT A SERIOUS TERMINOLOGY and it is reflexed in the group of "interpretation of mvc".