Add baffling-birthdays exercise

[ErikSchierboom]

This exercise combines two subjects that I feel we are still lacking exercises for:

1. Dates
2. Randomness

I've created a sample C# implementation for reference: exercism/csharp#2402

[tasxatzial]

I feel there may be some confusion between 'birthday' and 'birthdate' among non-native speakers. It might be helpful to clearly explain the difference in the instructions.

[tasxatzial]

I'm wondering if the properties should maybe be renamed at bit:

• matchingBirthday -> matchingBirthdayPair
• hasMatchingBirthdays -> hasMatchingBirthdayPair
• probabilityForMatchingBirthdays -> probabilityForMatchingBirthdayPair

Tough call. I'll share my thoughts.

• The tests with matchingBirthday prop seem to have consistent description and props, and it's only 2 birthdates as input, so i don't find any reason for adding pair anywhere.

• The tests with hasMatchingBirthdays have odd descriptions:

  has matching birthdays with matching birthday pair
has matching birthdays without matching birthday pair

  I think something like this makes more sense:

  has matching birthday pair
does not have matching birthday pair

  Given that the input is an array of birthdates, the word pair might suggest a pair from the array, so birthdates pair. So to avoid any kind of confusion, I would just leave the word pair out and adjust the descriptions to:

  has matching birthdays
does not have matching birthdays

  or something similar.

• The probabilityForMatchingBirthdays also have odd descriptions:

  probability for matching birthday pair for ten people

  This is a bit misleading because it suggests the probability of exactly one matching pair. A better way to phrase this so that it aligns with the paradox would be:

  probability of at least one shared birthday among ten people

  and i would leave anything related to pair out, for sure.

[tasxatzial]

@ErikSchierboom Have you also considered nesting (->), or is it not appropriate for these tests?

[ErikSchierboom]

In general we try to not use nesting, but in this case it might be appropriate due to several completely different properties being tested

[ErikSchierboom]

I've used nesting and tweaked the properties again. Once again, we're proving that naming things is hard :)

[ErikSchierboom]

There are now two variants (of which we need to choose one, or an alternative).

People with the same birthday are referred to as:

• sharing their birthday
• having a matching birthday

Which one do you prefer? Or should it be "same birthday?"

[tasxatzial]

Or should it be "same birthday?"

I think "have the same birthday" sounds more natural. "Sharing their birthday" sounds like they are celebrating together. "Matching birthday" does sound a bit awkward.

You'd have to do a ton of renaming though. Properties, descriptions, instructions...

[ErikSchierboom]

I don't mind as I'm agreeing with you so I think it is a good idea

[tasxatzial]

For the record, the Wikipedia article uses a mix of "share the same birthday" and "have the same birthday". It also uses "shared birthday" a lot.

[ErikSchierboom]

Okay, another iteration :) I've gone for "share" after all, as it seems to be used quite a lot in online material.

[tasxatzial]

I don't know if I'm missing something, but I've noticed that the props sharedBirthday and containsSharedBirthday have very similar tests. The first one is about checking two dates, the other is about checking an array of dates. Would it make sense to remove the sharedBirthday and move those tests to the containsSharedBirthday ?

Edit: By doing so, it will also resolve the issue with the plural in "shared birthdays" (see prev review) as this will allow to change it to "shared birthday" without creating a conflict in parent descriptions and props.

[ErikSchierboom]

That's a great idea. I originally had them separate to sort of force the student to solve it in steps but merging them is a good idea.

[ErikSchierboom]

As suggested, I've merged the sharedBirthday and sharedBirthdays properties.

[tasxatzial]

I think it looks good. I only have one remaining suggestion from a previous review, and the schema error in this one.

[ErikSchierboom]

Fixed both

[ErikSchierboom]

Thanks for the help everybody!