-2

I am reading the ISO 25023, and I am not sure if I understand the concept of Functional Completeness correctly, but I think it is useless in comparison to Functional Correctness.

Functional Completeness measures what proportion of the specified functions has been implemented. A missing function is detected when the system or software product does not have the ability to perform a specified function.

Functional Correctness measures what proportion of functions provides the correct results. An incorrect function is one that does not provide a reasonable and acceptable outcome to achieve the specific intended objective.

Could someone please explain to me how a function can be complete without being correct? How do you measure functional completeness? Do you just ask developers, or check if a function is defined in code?

2
  • 1
    Not everyone has a direct link to ISO 25023. Can you at least tell us what that ISO standard has on those terms? Commented Apr 22, 2020 at 9:08
  • @BartvanIngenSchenau thank you for pointing that out. I have added the definitions as given in the document.
    – lkarus
    Commented Apr 22, 2020 at 23:43

1 Answer 1

1

Suppose you have a website selling products. Then, by law, you are required to charge VAT (Value Added Tax) on the products you sell, at least for the customers living in your country/state.

Functional Completeness is affected by the question, do you charge VAT on your sales, yes or no. If you charge VAT, then the functionality is there, so Functional Completeness increases.

Functional Correctness is affected by questions like do you charge the correct percentage and do you charge it to the correct group of customers. If you charge 25% VAT, while the official rate is 18%, then you are not Functionally Correct.

So, you verify Functional Completeness by checking if the functionality exists at all, and Functional Correctness by testing that it gives the correct results. Most testing effort is spent on verifying Functional Correctness of a software product.

Software can be Functionally Complete but not Functionally Correct if it gives incorrect results. The feature exists but doesn't work.

2
  • thank you for your answer. how do you usually measure functional completeness? What I have in mind about measuring this metric is that you just ask dev (or someone in charge) if a function is implemented. but such a verbal way of testing and measuring makes the metric seem unreliable.
    – lkarus
    Commented Apr 23, 2020 at 6:51
  • There probably is a list of desired functions/features somewhere. Based on that list, you can ask the developers/project manager to tick off what they implemented, or you can try the product to see what is there. The information from the developers/project manager can even be supported by evidence, like progress reports or closed stories/tickets. Commented Apr 23, 2020 at 8:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.