- From: Ryan Freebern <rfreebern@unionstmedia.com>
- Date: Fri, 23 Nov 2012 09:27:31 -0500
- To: Dave Pawson <dave.pawson@gmail.com>
- Cc: Markdown List <public-markdown@w3.org>
On Nov 23, 2012, at 7:19 AM, Dave Pawson <dave.pawson@gmail.com> wrote: > Do you think headers well defined? I don't. Too many variants, too > many unclear and open to interpretation scenarios. Yes. There is atx style, which is # ... #*\n and setext style for h1 and h2, which is ...\n =+\n and ...\n -+\n The number of underline characters for setext style headers may need to be clarified further to avoid ambiguity, but if setext is widely used, a spec that invalidates it won't be welcomed. I think an idea that is falling by the wayside in these discussions is that markdown has two goals: first, it should be a syntax that can be transformed easily into presentational formats, and second, it should "look nice" in its native format. Our spec needs to make sure the format is well-defined in order to facilitate the first, but also flexible enough to fulfill the second, which is more subjective. > This is for the core/ baseline remember. Additional syntax can be > added with other profiles. If we plan to define a core that invalidates a significant number of existing documents, then I think we need to define the other profiles necessary to keep those documents valid simultaneously, or, again, risk rejection of our spec. > That would be my preference, starting from simple, clearly defined > syntax and semantic, valid with current implementations, adding > complexity as we feel is justified. "Valid with current implementations" is only one half; "encompasses expected usage" is the other. Ryan
Received on Friday, 23 November 2012 14:28:05 UTC