Iām collecting āstatementsā for a āTesting Bull$hit Bingoā.
Things like āItās so easy, we donāt need to test itā, āNo user would ever do thatā, āThis is an edge caseā and so on⦠I think you get the idea.
Please reply with your favourite bull$hit phrases, thanks a lot!
āI only changed one line of codeā ā Yeah, but that line of code governed permissions to the entire application and the missing ānotā meant nobody had the permissions they should have.
āItās only cosmeticā ā and if it doesnāt look right users are going to think we donāt care about the big stuff
āItās fully unit testedā ā and? Is it also fully integration tested? Does it do what the customer wants?
āThe UI makes it obvious what to do.ā ā No it doesnāt. It makes it obvious for users like you, who are a minority out there in the non-technical world.
And yes, Iāve run into all of these as well as @christianbaumannās starting list during my career.
(As a side note: Iāve generally found that cases of āNo user would ever do thatā are proved false within days, sometimes hours, of said users receiving the software - and Iāve refuted a few of them with the response āI did. Whatās to stop someone else?ā.)
I modified the test to pass (they were failing because of a bug, the dev didnāt fix the bug, just made the test passā¦)
We canāt test the change in the backend because the UI is not ready (testerās in backend teams who donāt know how to use postman or any other api testing toolā¦)
Or the opposite " We canāt test the UI change because the backend is not ready" , saying this while they have a framework set up to be able to mock the backend to test the UI changes
Boy howdy does all of this bring back some memories.
My most recent gig there was a frequent round of:
āIts just a one line changeā (I fought for extensive testing. was overruled. minor chaos in production ensues. I presented it as a case study that was really just an elaborate āI told you soā nevertheless detailing how small, untested change resulted in $$$ of developer and devops work in front of usersā¦well lets say there was an effect on how testing was evaluated)
āOh thats just a hostile user scenarioā (MSHās Law āAll users are hostile users. If they can do it, they willā Again I presented cases where a defect report was dismissed as āhostileā and yet that very defect was reported by users/support in production)
and yeah. Im bragging. But hey, no one else is going to do it for meā¦
Developer Test Lead, āYouāll only be writing up Sev 3ās and 4ās, we have so many of them, its almost pointless to documentā¦ā - Same Test Lead then goes on to investigate and mark as a Sev 2.
I canāt entirely agree that this stuff is BS There are always two sides to a coin, and it really depends on the context. As a QA engineer, tester, and
QA advocate in a way I have found myself in situations where the things mentioned in the conversation didnāt sound like BS, and I have also said some of those things to my dev and QA team members. However, I do understand why many people consider it to be BS in many cases. I have been there too
Just to be clear: I donāt judge the people doing this.
Both may help getting people wages.
Sadly there is demand for it and a big lack of education on the tester side as well as on the managers.