On Tuesday, April 26, 2016 at 6:17:20 PM UTC-4, Daniel Staal wrote:
At the moment, probably the only reason would be compatibility with other
versions of Pandoc - it's the type of thing where a 'random' change could
break people's documents in odd ways.
That said - Pandoc can also read Markdown as CommonMark, and the CommonMark
spec doesn't require the blank line there. So if you don't mind losing
some of Pandoc's extensions to the spec, you can get the more common
behavior. (And it might be worth the thought - if CommonMark is supposed
to be an evolution of the Markdown spec - of: Does Pandoc want to adhere to
CommonMark in this case?)
1. Yes, I do mind losing the pandoc goodness.
2. I tend to agree about following CommonMark, but in this case that also means following the original markdown, which remains a source of confusion to me.
But, if pandoc were to become compatible with this approach, is the backwards compatibility a big deal? How often is someone writing a line of text with a following series of three asterisks or underscores, for example? That is, these hr-makers would seem to be extremely unlikely to occur in other contexts, so such a change would be unlikely to have a big impact on anyone, no?