Device testing. It's something that all Web developers should be doing. It's the process of literally testing everything we build on real world devices.

We're all too fast to assume that we're all on the latest model of the latest phone, but something my morning commute has taught me is that there are always people use the Internet in unconventional was. Have you have uses a browser on a blackberry? I used to own the Bold, I don't remember using the native browser much, it was too clunky. Googling was as far as that went.

I once saw @quiffboy taking about user testing, and that the optimal testing group size is 5 people, by which he explains in most cases, it takes 5 users to catch around 85% of anomalies on a website.

So applying that to device testing, if we have 10 devices that cover a spectrum (or at least a snapshot) of real world conditions, give two devices to each of the 5 of us in the studio, we should get a pretty good idea of what our site will look like in the wild.

I've now found that the best way to test devices is not to sit in the studio and test, but to take a handful out to the nearest coffee shop offering free WiFi and start there. Luckily, Mansfield itself has free WiFi, which is ideal because it's not fast at all. (becoming the grand equaliser of Internet speeds)

I'm thinking of running a small workshop on testing on browser for a few that want to learn more or just have access to some old Internet phones, if you're interested, shoot me an email:

Oh, and If you've got any old devices you no longer need, tweet me @yratof and I'll be glad to give them a better home.

Note: - It's great to do a search for the website we've built, (This is after its gone live). Seeing how most people will access the website is a good way to see how what i've built impacts the sites ranking – An upcoming vlog on micro data explains more about this and how we can expand on the content of a site to favour how search engines will dissect it.