I work at a software company, consulting for the newspaper industry. Newspapers are forever breaking their internal systems. A trait I’m sure is shared with any large organization. For our customers, we will be simultaneously blamed and asked for help.
The lesson I’ve learnt over the years is ignore what the person is telling me, and instead focus in on one specific question:
What changed?
The reason this approach works, is that the system was working prior to us being contacted. Then something happened, something broke, a cascading array of side effects manifested, someone important complains, and it is all our fault.
This sequence of events has been repeated in my travels from Amsterdam to Auckland. In terms of problem diagnosis, asking “What changed?” is something that I find incredibly useful, and am still amazed when others fail to use this technique.
The approach often used by our staff and customers is to assume a particular thing is broken. The assumption is often based on previous experience with the thing breaking, and when they are correct, it re-enforces the assumption. Different people assume different things.
Finding out what it was that has changed is almost an art form. The frequent answer of “Nothing”, is always incorrect.
Something must have changed or the system would still work. This response indicates that you are asking the wrong person, or providing the wrong context. They may not know what has changed, or may assume that a change in a different system, such as network configuration, is unrelated.
Try asking in different terms, or expanding your search and asking other people. You can be safe in the assumption that something changed, even if it is as simple as the day of the week. Y2K bugs were a result of the date changing and the system not coping.
Of course finding a solution may not be simple, but odds are you have found the cause.
