Today we took another look at the schema for our back end API to iron out what it was we would be storing. With some insight from our team lead, who is experienced with web development, we identified more things we should store, but also ways to simplify the schema we had originally developed by storing less. For example, we originally planned on storing a “hit” when a visitor loaded a page and then coming back to that hit to add more info when the user did other things. But we’ve now reduced it to a simpler model that acts more like a log. Once a hit is written, it’s written. We don’t muck things up by revisiting its entry. I think the simpler we make the schema from the beginning, the less problems we’ll encounter later on.
I plan on investigating media queries thoroughly. They’re used to find out information about the client’s device and environment. And they can be very powerful. For example, they can tell when a device is low resolution or high resolution, and decide to send either a desktop or mobile version of the website to the client to view. But, they can tell when a client is a mobile phone. So if it’s a modern mobile phone with a high resolution, it wouldn’t send the desktop version because that wouldn’t make sense to view on a phone. We can mine this for our analysis. It’s a wealth of information that can help us learn how Engineering.com’s visitors are viewing the site.