We have two machines, lets call them Harold and Albert... Albert and Harold run different versions of the same software... Albert runs version 7, whilst Harold runs version 10.
One day, MrX comes along and uses Harold, and notices something unexpected, a bug! He reports this bug to Developer (Me).
I look at the code, look at logs, and look at the spec, and realise that MrX has seen the system work normally, and seen it operate as per spec, and so the bug is not a bug, but normal operation.
Now at this point one has to ask not "What are they doing reporting bugs which are actual behaviour", but perhaps "Why does the tester think that the software should behave a different way?"... Do they have a different spec? Does the spec have two different possible and plausible explanations? Has the spec changed? Has the machine changed?
So, not wanting to sit guessing I ask, and MrX kindly explains to me - like I'm about five and sucking a wet rusk - "Its always worked this way".
MrX then departs in a funk of smug, and I look at the code, and look at the logs and look at the evidence, which all says it always worked the way it works now, so I set about trying this on Albert, the old machine, the one MrX remembers with rosy tinted Historical hindsight and reckons worked differently.
Click click, chug chug, Albert operates its way through the same software section - and low and behold it works the exact same way....
Old machine, new machine.
Albert, Harold.
Version 7, Version 10.
Both work the same way...
The tester insists otherwise, despite the now overwhelming evidence that it always worked the way it works now, and indeed as I said (three hours ago) I've not changed that part of the system in the new version... No apparently I have to fix it.
Well, I don't I'm waiting on the person who write the specification to decide, but I pretty much think this is going to go poorly for the tester, not least because - well because its always worked the way I see it working - they're just remembering it wrong.
No comments:
Post a Comment