Building a Better Web
The Web is still the best platform to build functionality for connected consumers. It’s cheap, it’s fast, and it’s extremely robust in scope. However, things are not what they used to be back in 1995 when the Web was just a teeny-weeny baby. They say "it takes a village to raise a child". But, it doesn’t help when there are too many divergent opinions in said village. Unfortunately, the Web has suffered from such an upbringing.
The Web has become this massive, passive-aggressive hunk of potent ability and definitive uncertainty. That’s usually what you get when you try to make everyone happy. Beyond the pseudo-anthropological metaphors though, there are major challenges when we look beyond what the Web has become in structure, and focus mainly on the operational bits.
If you’ve done any kind of Web design consistently over the last 10 years [because I can’t imagine that most sane folks would want to be doing this for more than a decade], you would have noticed a somewhat worrying trend. It’s been the "orange beluga whale in the room" that nobody seems to want to address. Can you guess what it is? I’ll tell you! One word: Complexity.
Building a website today is at least 10x more complicated than it was 10 years ago. One does not simply build a website anymore like in the days of yore. No! One must first climb a mystic mountain [somewhere in the Himalayas maybe] in search of a temple whose patrons must first bestow one with the wisdom to know how to build said website [to modern exacting standards], as well as the actual fortitude to follow through and finish the job.
A modern website has to be Mobile-friendly and Fast; going by the advice of a very influential sage [known in some circles as El-goog] who resides in a place from where mountains can be viewed. All jokes aside though, these recommendations are actually quite valid for two main reasons: (1) mobile Web traffic now makes up almost 50% of the whole, and (2) mobile broadband is not ubiquitous across the world.
Responsive Web Design and Web Performance Optimization are two strategies that have evolved to try and deal decisively with the core issues, but both of them seem to have failed miserably. How do I know this? Take a gander at the following stats:
Based on data from BuiltWith.com, CSS media queries, which happens to be the predominant methodology for making website mobile-friendly, has 0.2% adoption across the entire Internet. What this means is that it is highly unlikely that much more than 0.2% of all websites online are mobile-friendly.
Based on our research for the State of the Mobile Web series, where we reviewed over 8,000 websites considered bellwethers in their fields, we found that less than 25% (on a weighted-average basis) of all websites were considered fast. What this means is that an overwhelming majority of all websites online are simply not performant.
So given these stats, what percentage of all the websites out there do you reckon are both mobile-friendly *and *fast. Your guess is probably as low as mine.
The problem isn’t that the purveyors of these now wonky and slow websites like being wonky and slow. Or that the Web community hasn’t built tools that help to mitigate this. The problem is that the pathway to becoming truly mobile-friendly and fast is ridiculously complicated and haphazard. This has to change.
Back in 2012 when I started looking into responsive Web design, I was incredibly frustrated with the way CSS media queries worked. It just seemed like a glorified hack with no roadmap. It also made CSS harder to use because it defiled the sanctity of the cascade, and virtually doubled the CSS payload. And, because breakpoints weren’t exactly definitive, you had to test your website on virtually every mobile device.
After pushing this to the Web community, I was surprised at the response. So far, the plugin has been downloaded thousands of times and it has been used by Samsung, Nascar, Brookside Chocolate, HDButterCup, and a number of other major brands; as well as some lesser known ones as well. We’ve also used the tool to help small business clients like Dancing Deer Baking Company make their website mobile-friendly.
But, even though Restive.JS worked really well, and was super-simple to set up, it still had some problems. First, it had jQuery as a dependency. Now, I love me some jQuery, but that extra 30-odd kilobytes [gzipped] on top of the core payload had to be trimmed down. Also, the documentation wasn’t so great. And, it wasn’t exactly feature-complete because you couldn’t use it to improve the performance of your Web page.
Long story short, Restive.JS had to be rebuilt from scratch. It’s now called rScript, and it does a whole lot more now, including Web performance optimization. It’s got a critical payload of less than 14 kilobytes [gzipped], and it’s still really easy to setup and use. We’re currently using it on our new website (http://www.restive.io), and it works like charm. Of course, when it’s publicly unleashed, it will remain free and open-source.
At the end of the day, building a better Web [where all websites are natively mobile-friendly and fast] doesn’t have to come with an exponentially larger cognitive load. The new tools and techniques we’ll soon be sharing with the Web community will attest to that. Tough stuff should get progressively easier with time, not more cumbersome, and we have to figure out ways to continually abstract the complexities that get in the way of us living better lives.
I have to give props to all those working on open-source Web technology. Even though it’s not exactly a neat and tidy household, you’re still trying to make the best of it. That’s what parenthood is all about, I think. So to you great men and women raising the Web in your own way, I say "Salut!" Time to kick things up a notch though.
See you on the better side!