- From: marbux <marbux@gmail.com>
- Date: Sun, 25 Nov 2012 06:24:47 -0800
- To: Karl Dubost <karld@opera.com>
- Cc: Dave Pawson <dave.pawson@gmail.com>, Markdown List <public-markdown@w3.org>
On Sun, Nov 25, 2012 at 2:43 AM, Karl Dubost <karld@opera.com> wrote: > > Le 23 nov. 2012 à 16:32, Dave Pawson a écrit : >> If we followed that logic we would do nothing, i.e. accept the current >> syntax definition and implementations? > > close to that. yes, but with an additional turn by being clearer and describing exactly what is happening so any new implementers follow the spec. The goal is to increase the interoperability surface. > > If *implementers* want to evolve the language then it can be done pushed by them. Note that they have little room too. They have to support the previous content of their users. That's life. Agreed, but with a caveat that they will have to add support for any construct we specify that differs from their present implementation in order to claim conformance. For example, if we require that they be capable of processing headings without trailing hash marks and their implementation presently requires trailing hash marks, they will have to add support for headings without hash marks in order to claim conformance. It would be up to them how they handle legacy docs. E.g., support both or add a filter to strip trailing hash marks when documents are opened. I assume that you intended something akin to this when said the " goal is to increase the interoperability surface" but think that to aid understanding that more detail is required. Paul
Received on Sunday, 25 November 2012 14:25:15 UTC