Enforce the use of only one keyword between else and otherwise

[tontonialberto]

• Conditional steps, where both then and else clauses are specified in the same step, make an inconsistent use of else and otherwise keywords
• In addition:
  • In some cases otherwise is followed by a comma, and in some others it's not
  • In some cases we can encounter Otherwise (ie. capital 'O') which follows a dot instead of a semicolon

I suggest enforcing the use of else over otherwise. This is because of the many different ways the latter is expressed throughout the spec, and also because it appears overall less frequently than else.

Here are the detailed reports for each case:

• If-else
• If-otherwise
• If-otherwise (with comma)
• If-Otherwise
• If-Otherwise (with comma)

Additional information is also available here.

[michaelficarra]

For these single-line conditionals, I have a slight preference for if-otherwise, even if if-else is a bit more common today. From a quick browse of the existing cases, there doesn't seem to be a reason to have both forms, so I support trying to unify them.

Something else I'd like to make sure we keep consistent is that both branches "do" the same thing: both define a particular alias, both set a particular alias's value, both return, both throw, etc. It appears this is already the case, so I'd like to include it in our editorial conventions along with whatever convention comes from this issue.

[gibson042]

I think this is a duplicate of #2919.

[michaelficarra]

#2919 got a little bit off track. Maybe we should rename #2919 to track the more significant restructuring proposal and we use this one to track the simple phrasing consistency.

[tontonialberto]

Probably the low hanging fruit would be to unify the different forms of if-otherwise. I suggest to move towards if cond, s1; otherwise s2 (ie. lowercase, no comma) as this is exactly the case with if-else forms.