Re: Some Design Principles

> On 01 Dec 2015, at 13:13, Robin Berjon <robin@berjon.com> wrote:
> 
> So essentially, I propose that SH be entirely comprised of subsets of
> existing standards, with simple extensibility rules that dictate what
> can be guaranteed to interoperate, and what can be added safely but
> might not be universally understood. This is relatively easy to get right.

Here as well, I agree, with a precision.

When we use a subset of existing standards, we must keep in mind that the files may also end up being processed by tools not aware of SH, and that these will understand the superset. So we need to avoid conflicts between the two worlds, and think about this in two directions:

a - When we give specialized semantics to things, they must remain compatible with the original non specialized definition, otherwise generic tools may mishandle SH documents.

b - We need to have clear definitions for what specialized SH processing tools should do when they run into things that are not part of the subset. And preferably, what they should ignore the extra information, not hard failure. That way it is possible to author documents that take advantage of everything a browser can do when viewed in a browser, while still taking advantage of the specialized meaning of SH when processed through SH specific tools. That also gives a much nicer forward compat story.

 - Florian

Received on Tuesday, 1 December 2015 05:31:14 UTC