"Goto Fail, Heartbleed, and Unit Testing Culture" Prior Work
These are the prior works that directly contributed to the development of my article Goto Fail, Heartbleed, and Unit Testing Culture, in chronological order. My announcement of the article contains links to the Google+ discussion threads pertaining to each installment.
- » In which I argue how the Apple SSL bug, aka "goto fail", could have been prevented with a unit test, and how code duplication was the most likely programming issue leading to the defect; contains a link to my proof-of-concept unit test; posted on the AutoTest Central blog
- » In which I express my extreme frustration at the reaction to the Apple SSL bug a week after publishing my AutoTest Central "Finding the Worm" post
- » Directly based on the previous two works; examines the "goto fail" bug in detail and a critiques the public response to it; presented to the March 18, 2014 Automated Testing Boston Meetup
- » A Testing on the Toilet-inspired one-page summary of the above presentation
- » An article compiled from the "Finding More than One of the Same Worm in the Apple" presentation; in the publication pipeline of the ACM Queue
- » In which I vent frustration with the Heartbleed bug, its untested fix, and the public excuses people have begun to make for it
- » In which I publish my proof-of-concept unit test for the Heartbleed bug, and take a swing back at anonymous Reddit cowards
- » A Testing on the Toilet-inspired one-page treatment of the Heartbleed bug
- » Based on my earlier TotT-inspired one-page treatment, "Finding More Than One of the Same Worm in the Apple"; published internally only (the link is to the announcement on this blog)
- » In which I openly address cybersecurity expert Dan Geer, and unwittingly inspire a lengthy Google+ discussion that raised challenges to the premise that unit testing could've prevented Heartbleed—and helped recruit several new reviewers for the "Goto Fail, Heartbleed, and Unit Testing Culture" article
- » Based on my earlier TotT-inspired one-page treatment of the same name; published internally only (the link is to the announcement on this blog)
- » Explores how "goto fail" could have been prevented by unit testing, and demonstrates how existing responses to the bug fail to address the root problem of a lack of unit testing culture; this is the published version of the earlier draft, released one day after "Goto Fail, Heartbleed, and Unit Testing Culture"
All of these materials were published under a Creative Commons Attribution 4.0 International License, with the exception of the Google-internal Testing on the Toilet episodes and the ACM Queue article.