[rfc-i] Accessibility and figures

Joe Hildebrand jhildebr at cisco.com
Mon Apr 30 09:11:18 PDT 2012

Since I'm not one of the people that has accessibility needs (yet!), it
would be easy for me to say I'm ok with some diagrams not having accessible
alternatives, at the discretion of the author; that where it is possible and
reasonable, accessibility is a goal to aspire to.

However, that would seem somewhat insensitive of me, and I wish there were
more folks that had direct experience implementing protocols with
accessibility needs in this conversation.

The safe starting point would be that anything that isn't accessible can't
be normative.

On 4/30/12 6:45 AM, "Paul E. Jones" <paulej at packetizer.com> wrote:

> Joe,
> I'm not sure this would help.  A diagram often exists to express some
> complex concept and text in the form of some data structure not so
> useful as an alternative, IMO.
> That said, explaining a diagram fully is not necessarily reasonable,
> either.  Diagrams exists to aid in understanding and sometimes they
> exist to fully explain.  In either case, a data structure cannot be
> visualized in the same way as a picture.  Pictures are visual and, by
> their nature, not really accessible to blind people.
> I believe explaining most diagrams in text is reasonable, as I think
> most figures should be used to assist in understanding the text.  If a
> diagram exists to fully explain what could have been written in text,
> then it's probably good to encourage authors to explain it in text.
> However, I'll be the first to say that it's almost silly to draw a state
> machine or things like Specification and Description Language (SDL; Rec.
> ITU-T Z.100) diagram, especially a complex diagram, and then to re-state
> in text every relationship.  So, sometimes accessibility needs cannot be
> met without doing the opposite in making the text less accessible by the
> rest of the world.
> Paul
> On 4/28/2012 1:39 AM, Joe Hildebrand wrote:
>> On 4/27/12 8:10 AM, "Paul Hoffman"<paul.hoffman at vpnc.org>  wrote:
>>> On Apr 27, 2012, at 12:33 AM, Larry Masinter wrote:
>>>> for accessibility, every figure's content should also be described in
>>>> accessible text.... that goes for ascii art too, for that matter.
>>> What do you mean by "described"? Assume for a moment that you have a diagram
>>> showing the relationship between entities, such as Figure 1 in
>>> draft-iab-rfc-editor-model-v2. What do you propose that the accessibility
>>> text
>>> for such a diagram say?
>> digraph "Figure 1" {
>>      IAB ->  "Independent Submission Editor";
>>      "Independent Authors" ->  "Independent Submission Editor";
>>      "Community at Large" ->  IAB;
>>      IAB ->  RSOC;
>>      RSOC ->  "RFC Series Editor";
>>      "Independent Submission Editor" ->  "RFC Production Center";
>>      "RFC Publisher" ->  "End Users&  Readers";
>>      IAB ->  IAB ->  "RFC Production Center";
>>      IRTF ->  IRSG ->  "RFC Production Center";
>>      IETF ->  IESG ->  "RFC Production Center";
>>      subgraph cluster_rfc_editor {
>>          "RFC Series Editor" ->  "RFC Production Center";
>>          "RFC Production Center" ->  "RFC Publisher";
>>          label = "RFC Editor";
>>          style = dotted;
>>      }
>> }
>> (see: http://www.graphviz.org/)

Joe Hildebrand

More information about the rfc-interest mailing list