Are frameworks on the decline?
I remember when I started building websites in the late 1990s. It was a simpler time of just Notepad and WS_FTP. We really didn’t have easy access to build software, multiple means of making a website, and especially ready-made frameworks and systems to build on. Web development was very much a “build it from scratch” ethos.
Fast-forward to the last four years, and everything changed. Building a website from scratch became ridiculous in the eyes of many as we had Bootstrap or Foundation to do some of the work, content management systems such as Wordpress to be the back end, and libraries like JQuery to handle the heavy lifting of some animation and interactivity.
I just mainly remember my peers constantly saying “don’t reinvent the wheel”, but now I’m hearing a different tone. After learning and embracing Bootstrap, I had a development team tell me not to design towards it, that they’ll build a grid system from the ground up. They simply advised me that I should do the same from now on.
As surprised as I was by that advice, it made me feel vindicated. However, I’m also not surprised. Probably from the days I learned systems such Bootstrap, Wordpress, and Drupal, I had heard many a developer complain endlessly about one factor or another when using these frameworks.
Sitepoint.com had coined some of this disgust as Framework Fatigue. Their experts really felt that we’re seeing way too many new frameworks launch that seemingly offer little benefit compared to their predecessors. The bigger issue though comes from when a framework gets a massive update (like 2.8 to 3.0), and thus a developer must literally rebuild the entire website in other to utilize it.
It can also be frustrating when a company builds heavily in an existing framework such as Backbone.js, and then suddenly the industry shifts and management wants to use React.js. Now comes another large and expensive rebuild when maybe that old site had seen only a year of life.
Beyond choice, file size and bloat has to be to biggest concern coming from developers. It’s already become common practice when using Bootstrap to craft a custom build using only the functions you want, thus leaving out the added bloat of things you do not. I’m also seeing many speak that developers should pull away from JQuery, as web browsers and CSS have evolved to handle many tasks JQuery was utilized for.
Lastly, as I mentioned in a prior article, security concerns have also made many a developer question if a third-party CMS is necessary for a project or not. Many have since moved back to custom solutions or even static solutions.
Are frameworks on the decline?
It would be easy to say frameworks are on the decline, but I’d like to more think that our usage of them as developers has matured. I recently saw a lecture by Google Chrome Developer Paul Lewis, and I deeply agreed with his conclusion that we as developers have to start questioning “do we need to use this framework?” when we begin a project.
Do we need Wordpress or Drupal for a simple 3-page brochure site?
Do we need Bootstrap if all we want is a simple responsive grid?
Let’s face it, in our quest to speed up development, we have allowed websites to become bloated and slower to load. Now comes the time we question what is necessary and what isn’t.
I for one am going to go back to my roots before the industry pushed me to stop creating and just use already-made solutions. I’ve done it in the past with a custom-built CMS, and I think next I’ll build my own CSS grid system with some improvements I’d love to implement. I’ll post my work on my Github as it happens.
Do you think frameworks are on the decline? Do you still prefer them over custom-building?