#600 add another reason (robustness in sense of type safety) for LSP

[gernotstarke]

please also add your opinion, @kraemerdr

[github-actions]

Build Successful! You can find a link to the downloadable artifacts below.

Name	Link
Commit	5c55100
Logs	https://github.com/isaqb-org/curriculum-foundation/actions/runs/11561892522
Download	https://github.com/isaqb-org/curriculum-foundation/suites/30199561150/artifacts/2114369562

[mikesperber]

No, this is completely wrong.

See #600 (comment)

[gernotstarke]

@mikesperber sorry, I see that differently: type safety is (one) approach to reduce runtime problems, and thus improve robustness (aka reduce the chance of runtime errors).

[mikesperber]

There is a confusion of terminology here:

Type safety reduces the chance of runtime bugs (i.e. problems in the source code within the software), whereas runtime errors are unforeseen circumstances at run time. Type safety does nothing to reduce those.

[gernotstarke]

that's correct, @mikesperber , but I still see "robust" as an attribute that is improved by type safety - as types might be checked both at compile- AND at runtime.

[gernotstarke]

I asked another (outside) opinion and got that result:

IMHO the ISO 25010:2023 qualities indirectly influenced by Liskov would be faultlessness, modularity and testability.

As @danielkr did not directly answer, I close this PR without merging.

[mikesperber]

as types might be checked both at compile- AND at runtime.

Strictly speaking, type checking is a purely compile-time mechanism. But we digress, as a "runtime type error" might also reduce robustness.

[kraemerdr]

As @danielkr did not directly answer, I close this PR without merging.

@gernotstarke Because of vacation and illness, I was not able to respond earlier.

Regarding LSP, I am actually less concerned with type safety (in the sense of data representation) than with compliance to interface contracts (and therefore predictable behavior).

As mentioned in #600, there is no need to drive this discussion any further, though.