Pull Requests as a Content Management System?

After launching new sites I like to let them sit and then after a few months rethink how I created it to see if maybe I could have done it differently.

• Could I have simplified things?
• Could I have improved it?
• What part of the design failed?

These are all questions I like to think about so the next time I’m tasked with something similar I’ll know which direction I should go in.

A site I launched the middle of last year is Laravel Events and I wrote about my thinking on how I designed here. I purposefully wanted to make the site as a simple as possible. After building it in a very complicated fashion I cam to this decision:

I had authentication, I had different categories of events, I had very specific time selection, and I realized all of those are overkill for this. Why can’t I strip everything down to a single list of upcoming events on the home page, then one form to submit your event. The only auth it needs is for the admin to approve and deny submissions and delete spam. Nothing more.

Now I’m thinking I still overcomplicated it. All the end of users of the site are developers and they are all comfortable with Git so I should have just used a static file system and allowed those that wanted to add their events to just create pull requests. This would have had the bonus of including a full audit log, and much more.

With this in mind I’d like to start today and redo the whole site, but I have a billion other things to work on. So it’ll not happen any time soon, but I think its fun to look back and see what you would have done differently.