[rfc-i] Proposed way forwards on backward compatibility with v2
pkyzivat at alum.mit.edu
Tue Feb 18 08:56:21 PST 2014
On 2/18/14 11:42 AM, Ted Lemon wrote:
> On Feb 18, 2014, at 11:23 AM, Nico Williams <nico at cryptonector.com> wrote:
>> Well, first, there's the possibility of ASCI->SVG conversion (there's
>> a couple of suitable converters out there). There's also the inverse
>> possibility (though that's harder, I think, considering the
>> constraints we have).
> I think SVG->ASCII is easier, although it may produce too-large diagrams. But really SVG and ASCII art are both poor representations for most things that appear in diagrams, because the semantics are not present in the representation, and can only be appreciated by someone who can look at the picture.
While its not good for everything, I find UML to be good for a lot of
things. It has a semantic representation as well as corresponding
diagram representations. (And it has "sequence diagrams" that are
functionally equivalent to call flows.)
But I don't know if that would be helpful to us. While in principle a
layout tool could generate diagrams from the semantic representation, in
practice it still requires manual layout. It would probably be difficult
to get UML tools integrated with xml2rfc toolset.
More information about the rfc-interest