I recently wrote about the big problem with Selenium.
This is a guest blog post by Jason Fox - a true Renaissance Man and honorary Arrgonaut. You can read more about his work at his blog and follow him on Twitter.
I get the sense that people think software development and programming is a mythical thing that is fundamentally different from “making” anything else. After all, a bunch of nerds sit in front of black screens and type stuff, and then boom a new start up exists. Very few outsiders actually see the creation, and as a result, attribute special powers to these programmers. As much as I wish I had special powers, I’ll have to be the guy that says it, “there is nothing really that special about programming.”
In fact, I’d argue the programming is like machining a motorcycle part.
It’s no secret that Selenium has problems. Here are some of the more well known offenders…
- It’s slow
- Tests are brittle
- You get false postives & finnicky results
- It’s a real maintenance nightmare
But what most people don’t realize is this: Selenium has a bigger problem, and these are just symptoms of it.
Hi, my name is Dave, and I’m a recovering Selenium user.
I was like you once – new to automated acceptance testing, not sure where to start. You grabbed for Selenium to get a quick fix. But before you knew it – BAM! – you’re in so deep you don’t know how to get out…
NOTE: This post was drafted from a talk given by our Captain (Dave Haeffner) on ‘Risk Management in an Agile World’ at the DC Automated Software Testing (DCAST) Meetup on Feb. 19, 2013. The idea came from Michael Connolly (DCAST’s Organizer), this is merely Arrgyle’s interpretation of it.
There are so many tools to choose from that you don’t know where to start. You’re worried you’ll pick one and find yourself down a rabbit hole with little to show for it.