So I decided after 10,00000 posts of me reflagging abstract art and cute animals I would start using tumblr as more of a platform to express all the different ideas I had on Webdesign/Type/UI/Code/Etc. I am going to use it to power the Journal section of my new portfolio site and I will be attaching a new theme I have been working in the next few days to match. THanks to everyone that has followed me so far and I hope that the nerdom coming in the next few weeks doesn’t run everyone off. Also shout out to @bencallahan for telling me to write more. I promise I will.
I read a little rant the other day by Joshua Gross on his blog highlighting being sick and tired of Twitter Bootsrap. And Yes, I am sick of Helvetica Nue and those little gradients just as much as Josh and yes I think its very bloated if you just need some simple UI and a grid to work with. But I think there are several other reasons people should stop developing sites in the framework. Let me start with a little story.
I work for a large non-profit that receives millions of dollars in online donations a day. The day I got my new job, the first thing i did was go on the website and try to make a donation. It was a terrifying experience. A table based form and 10px type made completing the 50+ fields without erroring almost impossible. I teamed up with a co-worker and went to work. We only had a few weeks before the holiday donation rush started and needed a quick/responsive way to improve the UI. Twitter Bootstrap was perfect. It offered tabs and modals and giant buttons and great type. We got to work hacking it together and immediately saw a drastic improvement in our abandonment and even improvements in donation amount. I was hooked. I immediately went out and started to whip together my next web app in Bootstrap.
What I didn’t realize in my haste was that its a lazy mans out. Figuring out a sustainable system for growth and UI on a enterprise level site needs to be a custom experience. Soon after the donation form launched we quickly moved into updates in the user flow. We realized there was a mass of custom hacks we would have to make in order to allow the form to work the way we wanted. Also there was a boatload of things we were not using in the framework adding bloat and killing load-time. On top of that the look and feel of bootstrap was great compared to the previous form but it felt alien and wonky with our current brand.
So here we are at version 2.0 of this form, that I like to call the “Million Dollar Baby” and we are now stripping out Bootstrap and replacing non-sematic markup with lovely custom HTML5 and CSS that reduces load on that page by 300k.
The thing that was so great about Bootstrap was that I could have a grid, nice typography, and UI with little to no setup. It allowed me to get straight to work with no hassle. It gave me more time to do what I wanted to be doing which was solving UI problems and building things. But what I have learned since is that It takes the same amount of time to get started if I just code more efficiently. Tools that I have now (specifically SASS and Emmet) are making up for that time and saving me lots of hassle trying to customize Bootstrap to feel like our site. Its also building a sustainable project and code base around just the features we need in the application. I’ll stop before this becomes a rant. Comment and let me know what you think.