Brian Rice, in his keynote speech at StarEAST 2006, made a great point: He said “Talking to developers is like talking to monkeys.” What he meant — at least, what I thought he meant — was that software testers and software developers have trouble communicating. Although he was only trying to be funny, I think he was making a much more direct point.

I don’t think software developers and software testers are speaking the same language. Both developers and testers have the same goal — that is, they are trying to engineer good software — however, I think there are many cases where they don’t understand each other. Brian, mentioned above, was blaming the developers for the communication problem; he was jokingly saying that developers have a lower cognitive ability, and that testers are, in fact, the better, evolved human form. Although some testers may believe they care more for the software than the developers do, I don’t think this problem has only one side to blame.

Take this example for instance: Two people are asked to communicate; one person in English, and the other in Chinese. They must only communicate through text, they must never use an interpreter, and they cannot — by any means — meet in person to judge facial expressions. Do you think these people would have communication problems? If I had to guess, I’d say so.

Although this is an oversimplification of the problem, I think it is parallel with what companies are asking developers and testers to do. Developers and testers must, by all means, talk about a bug through a bug tracking system. These systems are great for management, in that they provide ways to audit bugs, analyze metrics, and promote insight into the status of a software project; however, I don’t think they are that great for the people actually doing the work. Could these bug reporting systems, with all their benefits, be getting in the way of tester-developer communication? Could these systems — for the sake of auditability, metrics, and insight — be causing unwanted side effects to those actually communicating? Although I have no definitive answers to these questions, I’d like to cite this example as some proof that bug reporting systems may be doing more (or less!) for communication than we actually think they’re doing. We might — if I’m at all on to something — need to rethink the way we do this.

What do you think? At the end of the day, I’m simply one man — and in fact, I’m just a student. I’d love to hear from people who have different — if not contrasting — views on the topic, who’ve actually spent some time getting work done. If there is anybody out there who’d like to comment on their experience, I’d love to hear it.