An “Aha!” moment

I’ve been reading a lot lately (rather than posting, sorry kids) and Chris Coyier asked for people’s “Ah ha” moments with Javascript in a recent article.

Well, I had such just such a moment last night at about 3am, and even though it wasn’t for Javascript, Chris’ solicitation inspired me to share it.
Nope, it wasn’t for Javascript. It was for HTML. Specifically, it was for HTTP requests, and website development in general.

So, the old school website philosophy, which is tried and true, is pretty easy to grasp. A full web application is an “enterprise” level MVC design pattern:

  • You have your database and interface to that database (Model)
  •  You have your web server, which processes requests, reads/writes from the database, and servers web pages (Controller)
  • You have your web pages, which submit information to the server in the form of POST or GET calls (View)

Simple, straightforward, and proven.

Well, with the introduction of AJAX methodologies almost a decade ago (has it really been that long??? Man, I’m getting old…) we now have a significant portion of the processing / logic being done on the web page itself.
This is a good thing! I’m definitely not saying it isn’t. I’m just punching your ticket on my thought train that led me to my “Aha!” moment…

…as I was saying, the “clean” MVC pattern has been enhanced now, making HTML/Javascript pages become applications in their own right, using MVVM or MVP or other design patterns (I promise, I’m coming up on the point here, just bear with me).

These new patterns utilize JSON, AJAX, and Websocket calls, to interact with online services.  Meaning, the webpage itself is now the application, with its own controller/presenter/view model component, and Jquery/Javascript calls to get information from a service is now the model. There is no longer a need for the webpage to make GET/POST calls. In fact, the only thing the web server is needed for now is the downloading of the initial HTML for the web page, and serving up JS and CSS files.


And this is the “Aha!” moment:

HTML is now an application.

(The paradigm of ‘Site’ is dead)


Webpages aren’t webpages anymore. They’re applications. GET/POST should now be looked at as just-another-data-source for the Model of that application, and for “downloading” the application (FKA webpage).

That may not be an “Aha!” moment for anybody else, but the implications of it most definitely were for me:

  • You don’t need Site Maps anymore.
    - What formerly was a separate webpage is now just another view / attribute of the webpage ‘application’.
    - For e.g.: Preferences (it used to be a separate page. Now it’s a panel, just like it would be in a fully-installed app)
  • You can treat a webpage literally as you would a full application:
    - It’s ‘downloaded’ and ‘installed’
    - It has different modes / views (instead of separate pages)
    - It stores information locally (instead of just using cookies)
    - It consumes web services (instead of it being the View of a web service, ala POST/GET calls)
    - It can live on its own, outside a browser (e.g. PhoneGap)

So there you have it- You can start treating HTML as an application. Not a pseudo-application. An actual application. The call to the web server is now the download+installation process, and nothing else.

I’m 100% sure I’m not the first person to think about it this way, but it was definitely a paradigm shift for me, and qualified in my mind as an “Aha!” moment.



Comments are closed.