Eyjafjallajö-what?
As a native English speaker, the pronunciation of Iceland’s volcano has eluded me. But I knew Icelandic was a tough language from when I visited there a year ago (see my post: Six Testing Lessons From Iceland). What I also know is that travel in Europe has turned into a mess, and that travelers from around the world have suffered from the wrath of Eyjafjwhatever (including uTest’s very own product manager who is currently stuck in the United Kingdom).
Of course, getting stuck on a vacation to a foreign country sounds awful (I have terrible nightmares of one day getting stuck in Tahiti), but there are some valuable lessons we can learn from all this.
1. Expect the unexpected.
In my testing lessons from Iceland, I invented a few Icelandic testing tours in the spirit of James Whitaker. One of them was the Heimaey tour:
Heimaey – The Icelandic island of Heimaey is known for two things: fishing and its little problem with volcanoes. In 1973, the entire island was evacuated after massive cracks formed in the ground spewing lava and ash everywhere – eventually forming the volcano Eldfell. Take a tour of your users’ worst nightmare problems and make sure everything works correctly.
2. Be ruthlessly devious.
Figuring out a worst nightmare scenario requires a little luck and creativity. Most people didn’t think of Icelandic volcanoes as a threat to European air travel, partly because one like Eyjafjallajökull hasn’t erupted in recent memory. Testers can’t think this way. A tester must be devious and look for all those nightmare possibilities that others miss. What happens to your app if something really catastrophic happens, even if it’s unlikely?
3. Plan for catastrophe.
Pop quiz! Which of these two things would have put airplanes back in the air sooner:
- Eyjafjallajökull stops erupting and all the ash falls into the North Sea
- Airplane manufacturers had developed good understandings of how engines are affected by different concentrations and compositions of volcanic ash
The right answer is the second one. Knowing that information would have made it easier to route planes around dangerous parts of the ash cloud while still operating in safer areas – versus not flying at all. Why would anyone even answer a question like that in the first place? Because a ruthlessly devious person thought it would be a good idea to find out.
4. Be reasonable.
The biggest downside to all this catastrophe planning is that it’s expensive, time consuming, and easy to dismiss. So be reasonable with your thinking. Volcanic ash in Europe is a reasonable thing for which to plan. For software, that could include planning for cyberattacks or datacenter meltdowns. These things could happen, and planning for them is sensible.
Not sensible: asteroid strikes, nuclear apocalypse, and Godzilla.
Have you been impacted by Eyjafahrvergnügen? Tell us in the comments where you’re stranded and how things are going.








I’m new to the community, glad I found this site. Great article.
I was struck at first by the apparent contradiction between points 3 and 4. As a tester, my definition of what is reasonable when planning for catastrophe tends to differ from development and project management. I think the key to being reasonable is partially knowing what is sensible to test for, and partially knowing when to put up a fight or back down.
Skinny –
I couldn’t agree more. If I had thought about it, I would have added another lesson:
5. Know how to communicate your thinking and persuade others.