BootstrapCDN

This is the fourth Stack Decision of this series. Read the last one to catch up.

My friend and I were driving up to northern California, I got a tweet saying that their corporate anti-virus software was reporting that certain files on BootstrapCDN contained malicious code that installed Java malware on Windows XP workstations. The hackers were clever, using MaxCDN's #API they found a way to update the origin server URL & port (8080) to their server and only updated a few files to fly under the radar.

I thought it was over, who was going to use BootstrapCDN after this? Well, like earlier I was wrong again. After the postmortem was written people left comments and emailed us saying “thanks” which I can’t thank those who did, it kept me going. Which was needed because the next wave of issues are coming in part 5.

AMA below. 👇

This is the third Stack Decision of this series. You can read the last one to catch up (link below).

Josh is a machine. He cranked out the rewrite of the site and as well as the Amazon S3 syncing/permissions/content-type scripts in less than a week (with a full time job + 2 small kids at the time). The new site had/has a section where you could try out all of the Bootswatch themes on our site and while it might sound silly, it showed me that he goes above and beyond and I totally lucked out.

Anyway, Josh decided to go with Node.js, ExpressJS, Jade (now called Pug), and configs with #yaml, all things I have read about but never used in production. I quickly found out Josh was a Vim user (still is), because every Jade file he worked on had //- vim: ft=jade sw=4 sts=4 et: at the bottom.

Everything was running smoothly, I was encouraged by David (co-founder from part 1) to take a vacation. I took him up on that. Unfortunately, my timing couldn’t have been worse. More on that in part 4.

AMA below 👇

This is the second Stack Decision of this series. You can read the last one to catch up (link below). Bootstrap, Jacob Thornton aka @fat tweeted about #BootstrapCDN and according to Google Analytics, that sent 10k uniques to the site in 24 hours. Now I was pumped but I knew I was way over my head and needed help. Fortunately, I met my co-maintainer Josh Mervine at the 2013 O’Reilly Velocity Conference and we hit it off immediately. I showed him the MaxCDN and Amazon S3 stats and his eyebrows went up. When I showed him the code, he was very polite, “well, I mean it works but I really want to try Node.js out so I’m just going to rewrite everything in Node and Ruby for the S3 scripts.

I didn’t know what to expect from Josh, to be honest. In the next decision (part 3), I will go over how he completely transformed the project.

AMA below 👇

The following will be a series of decisions we made that took BootstrapCDN from 0 to over 74 billion requests a month (and growing).

Initially, I didn’t want to do BootstrapCDN. I have attempted a few projects like it before and they always failed to gain any traction. In June of 2012, my boss at the time (and good friend today), David Henzel got a BuzzSumo Alert coming from an #OpenSource project on GitHub called Bootstrap and someone mentioned that MaxCDN was always looking for projects to sponsor. Long story short, David registered the domain and told me to get to work.

The first version of the site was written in PHP. It was quick and dirty but met the scope. We beta tested it for a month then people started to use it after searching for “bootstrap cdn” on Google.

I was still skeptical until, well, that’s for the next decision.

AMA below. 👇