Don’t bother testing, we’re not going to fix it.
In 2009, I was working at a very small software company that made a game to teach children how to play piano. I had been hired to head up their support department and found out quickly that the department consisted solely of me. On the bright side, support issues were limited, and I found myself with a large amount of free time, and used it to find bugs in the software.
This proved to be a bad decision as I soon discovered that the software was originally coded for Windows 98, and had never been upgraded or patched as the years rolled by because we had stopped paying our programmers. While my list started out with some minor graphical glitches and small play errors, it soon escalated to the point where I could break every facet of the software with a few quick keystrokes.
This included “upgrading” the software to our ultra-deluxe, $400 version by changing a registry key. The software also proved unstable on Windows XP, and with the release of Vista just around the corner, I was tasked with finding ways to fix compatibility issues. They asked me to get a beta copy of the OS, so I could start testing that as well. Since we still were not paying our software developers, there would be no patches or updates. They simply wanted me to have a head start on what our customers would be facing.
During this, they hired a “Creative Consultant” to revitalize the company and start bringing in more revenue. A meeting with the core members of the company was called, and they sat next to my desk and began discussing proposals. The first proposal was to bring our software and keyboard peripheral to the PS2, and how they could have a version out in stores in about a year. This was generally viewed as a good idea. I had to inform them that the PS3 had been out for over two years, and PS2 development wasn’t a good idea. He then suggested the Gamecube, as that was more family oriented. I suggested the Wii, and explained that the Gamecube was no longer manufactured. The idea was tabled.
The consultant continued to work for the company for several more months, at the tune of ten thousand dollars a month, and ultimately produced nothing. Meanwhile, I went six weeks without being paid.
Finally, a new software development house was brought on board, and a joint deal with one of the largest “family-oriented” companies in the world, featuring one of their biggest stars was on the table. We had new software, and new keyboards, and I was tapped again to test them, and report my findings.
While the software was much better, having an entirely new code base developed for an at least semi-recent operating system, the hardware bundled with it was the cheapest, flimsiest manufacturing possible. The early production units arrived with broken keys, malfunctioning speakers, and broken ports. When I tested the one unit that wasn’t broken, I found that it contained buttons and features from an earlier version of the software that were no longer used, and if pressed, crashed the program. When I reported these findings, I was told that these units were the final spec, and would not be changed.
So, we went from having crippled, broken software with semi-decent hardware to semi decent software, with crippled, broken hardware. Isn’t progress great?
Oh, and this was all developed out of our pocket, and the deal eventually fell through. Shortly afterward, I helped hire and train my own replacement.