You know what else is free too? --> Authoring katas on CW in any spoken language of your choice (The rules prohibit this but not the system ^^)

Done in above fork

I respectfully disagree that this makes the Kata worthless and unsalvageable. What's wrong with authoring a Kata which strikes some "middle ground" between your typical beginner Fibonacci Kata and some ultimate generalized version of a Fibonacci Kata understood mostly by math majors and postgraduates?

By this logic I can author a Kata on the "theory of theories" in mathematics and that would make pretty much every other existing mathematics Kata on this site a duplicate ;-)

If they're immutable, they're also fixed size. I think this is a case where there's no right answer and it's up to preference. Since the types of the inner elements are heterogeneous (for the clues), and that the sequences themselves represent finite sets of data that will never change since the input they're based on will never change, this ticks both major boxes to justify using a tuple. The only gains you get by returning a list of lists is that the function's user will be allowed to modify the returned lists afterwards, but again, those are very subjective API design choices. Again, since the answer is obvious even before runtime, to me it semantically reads like named_tuple('clues', 'clue1, clue2, clue3, ..., clueN')

Some argue that tuples are Python way to have immutable lists. I am not sure I agree, but again, I am not a Python person.

It is accepted, it's just not a system tag, so you don't see the tag in as many screens as such tags. Adding additional system tags wouldn't hurt though ;)

sorry i forgot to press 're-publish'

try again? i don't see why this should happen

It should be fixed now.

yes i do, i guess i'll add the tag

would you consider this a performance challenge?

fair