Enclose selectors so that it's clear what they are

[stasm]

@markusicu wrote in #230 (comment):

The next most confusing thing is figuring out which placeholder-looking item becomes an input to the selection, and how those match up to the variant value tuples. You really need to enclose the selectors so that it's clear which ones they are, and the number of selectors needs to be the same as the number of values in each value tuple. I suggest enclosing both the list of selectors and each value tuple in square brackets. Bracketing both selectors and tuples correlates them visually.

[eemeli]

Quoting @stasm from #263 (comment):

My example also has long variant keys with spaces to demonstrate what they would look like with delimiters, but the EBNF allows bare variant keys too. I think a more realistic example would be this:

"{$foo :func} {$bar :func} one one [Some text...] one few [etc.]"

There's #251 about making the preamble with selectors stand out more from the rest of the message, but otherwise I think this is actually pretty minimal.

The thing that struck me about this example is that when glancing at it, I first see the undelimited one one and one few. In order to register that the actual contents of the message would be Some text... or etc., I need to actively think about how MF2 works.

If instead the delimiters were around the keys:

"{$foo :func} {$bar :func} [one one] Some text... [one few] etc."

it becomes much easier for me to see the contents before I see the markup.

[echeran]

+1 to @markusicu and @eemeli for preferring delimiting selector value tuples. It reduces cognitive load for humans when they mentally parse.

The original comment also mentions delimiting the definitions of the selectors, which sounds consistent and preferable to me.

[echeran]

Issue #253 has our previous discussion on the topic, although it only focused on the part about selector value tuples, not their definitions. The comments there, too, are in support of delimiting, asking why we wouldn't.

[eemeli]

As an expansion of my comment above, I've now filed #275.

Also, I recognise that my comment here should probably have been applied to #253.

[stasm]

This might be one of the issues that are too fine-grained. #253 is so closely related that it might be a good idea to close this one and continue the discussion there.

[eemeli]

Conclusion: This issue is redundant, will continue in #253.