Building 16 Sites in 2 Hours

Our program (Boulder Digital Works) recently hosted the Making Digital Work conference in Boulder. Jesse Weaver and I were requested to help with the conference. The idea was to split the conference attendees into teams and develop an idea. They would come up with two variations of that idea and create some copy/keyword combinations to test between. Then Jesse and I code all the variations for each team into web pages that we could test with Google’s web optimizer. Tim Malbon developed the logos and Jesse and I coded. It turned out well and the teams had some great ideas! I have  a few takeaways:

Simple Works….

The template, the process and the testing platform were all immediately understandable and easy to work with.

Start with a Clear Explanation…

Start with a clear explanation of the intended outcome. Tim explained the idea in one short email. The concept wasn’t hard to grasp because it was clearly explained and thus we were able to put most of our energy into building it.

Don’t Agonize…

When you’re rapidly prototyping an idea it doesn’t need to be perfectly polished, the key word here is “rapid.” I noticed several small errors while were presenting the websites, but it didn’t matter because the groups understood that this was an exercise in speed not quality. Most importantly they got the overall gist.

Make Backups…

About 4AM on the day of our presentation, I had the thought to make backups. Like an idiot, I thought “nah, my server hasn’t gone down in 8 months…what are the chances?” Famous last words. About 7:30AM I woke up to a text message telling me my server had gone down. I checked several ports including SSH and found confirmed my server was unavailable. I contacted my hosting provider who informed me their whole network was out due a (d)Dos attack from LA, California. I assumed it would only take an hour to clear up, but three hours later we getting a little worried. I had an encrypted backup, but couldn’t open it because I didn’t have my encryption key. It’s been a while since I’ve used those backups and forgot to get the key when I left my last job. A couple hours before the presentation the network came back online. We got lucky. Had my server been out during that time, we would have been screwed! Always make a backup, test the backup works and have  a backup plan to easily switch to another server just in case.

Designate a Zero Hour…

Decide collectively on a time when all production must be finished and changes cannot be made after that point. There is always an urge to “tweak” things to make you project just a little bit better. This is dangerous territory, in programming there are always unforeseen consequences to changes and they may not be immediately apparent. The reason I was up at 4AM (day of) was because I was “tweaking” things. However, when you’re tired you make mistakes and I failed to realize that some other tweaks I was made obfuscated some of the text on the pages. This made things difficult to read and became pejorative rather than helping.

All in all I think the project was successful, things could have gone smoother but considering we built 16 websites in a day, things went as well as they probably could have. Thank again to Jesse Weaver and Tim Malbon for making this a success.

Leave a Reply

Your email address will not be published. Required fields are marked *