Problem-specifications: README of atbash-ciper mentions group size, but there are no tests for it

[dvberkel]

The atbash-ciper exercise describe the _group size_, i.e. the size of the group that characters are grouped by.

When I ported this exercise I interpreted it as to provide a means to alter the group size, but the canonical data does not provide tests for this scenario.

It this something to address? Either by

1. Changing the description in the README.
2. Provide tests for group size

Or are we happy, just the way we are?

[jtigger]

Personally, I'm happy the way we are. The group size of 5 is given which matches the canonical data.

Perhaps strengthen the suggestion to use 5?

[jtigger]

@dvberkel and @stevejb71 included a test for different group sizes in exercism/xocaml#72. With that example in hand, @exercism/track-maintainers should we update the exercise across the tracks?

[Insti]

should we [update the exercise to test for group sizes other than 5]

No.

The main point of this exercise is the encoding. This is what should be tested for.
If grouping is an interesting problem it should have it's own group-size exercise.

[dvberkel]

I am happy not porting this to all language tracks. We could close this issue, if no other opinions emerge.

[kytrinyx]

Or are we happy, just the way we are?

I think we're happy just the way we are :heart: