Our daily got derailed today by a conversation between whether there’s a difference between ‘trust’ and ‘confidence’.
(In the context of automation) You trust the tests. You’ve got confidence in the tests.
(In the context of releasing) You can have confidence to ship, but not trust to ship.
So they’re not quite synonyms, but what’s the difference?
For me, there is a very clear difference between the two, and that is, feelings versus facts.
If I trust something that is an emotional feeling. If I trust someone and they break that trust, I’d feel betrayed.
For releasing, trust on its own wouldn’t be enough. There would be unanswered questions that might have nothing to do with automation and might be about lack of coverage, performance concerns, security, or untested areas. It could even be timing, a lack of a feature, or a capability. E.g. you don’t have all the facts.
If I have confidence in someone or something, it is because I have some kind of evidence or facts to give me that confidence. If that confidence breaks, I’m more likely to be surprised by it.
For automation, the facts might be that 98 or 99% of tests have passed. We have good coverage and a stable environment that is equal to the live one.
Hope that makes sense.
I flunked at English language, so for me the words have similar enough meaning that anything more than 5 minutes debate is akin to the “debate to calculate the number of angels that can fit on a pin head”. If I were to differentiate, for me they lie in trust being more concrete, and confidence a feeling.
Like the way we trust that our banknotes will be honoured by the reserve bank, but have confidence in our companions to bail us out. Or even… the other way? Definitely not worth derailing a meeting, especially if there is no actual measurable goal or anything actionable. Was this in the context of security testing by any chance?
I think it’s the general nature of professional software testers to not trust anything. I wouldn’t use trust on the context of automation, personally.
You can have confidence in your automation, which should hopefully project and give others (non-technical) confidence too, which builds trust between you.
If you have low confidence in your automation, and it shows, that may break trust in your abilities.
I’m often fairly confident that the automation tests should not be the sole trusted oracle for a release decision.
I trust the tests but they do not give me the confidence to ship on their own.
Used in certain ways they can be semantics, what problem were you trying to solve?
Totally, that was just me reaching for a grammatical example relevant to the group
We definitely weren’t trying to solve a problem - it came up in conversation, and became a tangent that we, as a bunch of nerds, had fun kicking around.
This is so insightful!
It also follows that feelings are somewhat binary, and confidence is gradiated (if not quantifiable) - you like or dislike brocolli, but based on experience you have a level of confidence in whether Sainsburys will have any in stock this week.
I love a tangent. My meetings rarely run to schedule. I’m in a globally distributed team, so we only get a short time to work together when we’re all awake, so we’re all happy to take a bit of “team time” when we do.
Close. In this case, it started off with a conversation about reliable infrastructure.
It’s interesting though - when I worked in Security, I remember lots of talk of “trust” (e.g. Zero Trust) but I don’t remember nearly so many instances of confidence. And yet, by @AdyStokes definitions, it should probably have been inverted.
It’s not that we’re inherently untrusting people. But it’s our job to conjure risk questions to ask (aloud, or in the design of an experiment) and so we’re naturally creating a picture of all of the things we don’t know and so can’t empirically rely upon. Yet.
There’s no debate if you are a fan of the American TV show Supernatural. An angel describes themselves as ‘roughly the size of the Chrysler Building’, a skyscraper in New York standing at over 1,000 feet or 319 meters. So, no angels on a pin head, too big 
I fondly recall a sprint retro about 6 or 7 years ago. I was contracting, so I did not really need to join all of the retros. The entire hour for retro and planning was diverted. We even came back after lunchbreak, a lot of baggage was aired, most of it good. I was getting paid per hour, so I was the last to complain obviously. But that one over-running meeting in which emotions hit boiling point too, reminded me of how easy it is to get off track. I have a brain that literally lives in the off-track, so I’m as guilty as the next person. I also love tangents, but I have learned to fear them, ever since that rather fun, but long retro.
I suspect @dancaseley that you have learned something from the derailing. My story was probably very similar, people not using words to mean the same things for everyone. Nerds love accuracy.