andyMatthews.net
Saying Goodbye to jQuery Mobile
Are you kidding me? Of course I'm not "really" saying goodbye to jQuery Mobile. I've got far too much invested in it to bid farewell to the best mobile framework out there. What I am doing is making a decision to use it when appropriate. The choice of finding the right tool for the job is one of the hallmarks of an experienced builder, be it a carpenter, a plumber, or a web developer.
We've been using jQuery Mobile at my current company, Goba.mobi for quite some time. It's allowed us to rapidly develop an impressive, beautiful mobile website with bells and whistles with little more than HTML and CSS knowledge. That was when we used our website to mimic our iOS and Android apps as closely as possible. Recently we stripped down our mobile website and rather than attempt to offer all of the functionality a true mobile app provides we decided to limit what was available. In the past a user could create events and groups, add friends, a discover businesses and services around them. We decided to go the route of Path and display to the user a single interaction, a moment in time as it were.
So how does jQuery Mobile come into this? A typical page load for our old mobile website hovered somewhere around the 300k range. This included web fonts, jQuery, jQuery Mobile, a Theme.css file, and an overrides.css file and a smattering of JavaScript plugins. This was acceptable when the user actually interacted with multiple pages. As each of the files would be cached the download cost would become less with each additional page the user viewed. In this new model though we expect that each user will only view 1 page. Download speeds for mobile devices are improving all the time but 300k for a single page for a mobile device is far too excessive.
To that end I've undertaken a rewrite of the mobile website. I started from scratch and attempted to mimic exactly the existing design and functionality while putting the CSS, HTML and JS on a diet. The resulting website weighs in at a svelte 30k-40k with near identical features. As this new site will be primarily mobile I focused heavily on webkit support with fallbacks for other browser who might happen across one of our pages. I know that I'm not done yet, but it's a great start. Click the image up top there to take a peek at the static version of the build. Not every link works, but you get the point.
jQuery Mobile, I'll always love you, but you're not the best fit for every situation...and I think you know that, and are okay with it. Love, Andy!