Appendix E: Accessibility Support
This appendix is informative, not normative.
E.1 Accessibility and SVG
The degree to which SVG content can be considered accessible depends on
what SVG is compared against and the manner in which the SVG content is constructed.
When comparing SVG content to images, SVG will be much more accessible
for the following reasons:
- Text strings in SVG are represented as regular XML character data rather than bits in an image.
- At any place in the SVG hierarchy, a drawing can include a long set of
(i.e., the 'desc' element)
and/or a short description
in the form of a title
(i.e., the 'title' element).
Both of these features can be used to help the visually
impaired interpret both the intent and specific content of a drawing. The drawing
can be architected such that there is a single description for the drawing as a whole
or there are multiple descriptions which are distributed within the drawing
and describe each separate component within the drawing.
- SVG has a notion of structure from its grouping constructs such as the
This grouping constructs, when used in conjunction with the
elements, provide information about document structure and semantics.
- Personal style sheets can be used to adjust the color
contrast of graphic elements.
- Because SVG content is scalable, people with partial visual impairment will
be able to magnify the content or zoom in on graphics for easier viewing.
- This specification includes a set of
SVG accessibility guidelines,
which will help to promote the creation of accessible SVG content.
- SVG's Conformance Guidelines:
thereby promoting increased accessibility for both generation and viewing
of SVG content.
On the other hand, when comparing SVG to other markup languages, SVG can
be less accessible. For example, in most cases, an
XHTML [XHTML10] document will be
more accessible as XHTML
than it would be if converted into SVG
since the higher-level structure and semantics will be lost in the translation.
The degree to which SVG content is accessible depends on the degree
to which the document's author follows the
SVG accessibility guidelines.
E.2 Aural style sheets
For the purposes of aural media, SVG represents a
stylable XML grammar. In user agents that support
aural style sheets, aural style properties
can be applied
Aural style properties can be applied to any SVG element that can contain
character data content, including
On user agents that support aural style sheets, the following
can be applied:
For user agents that support aural style sheets and also
the user agent is required to support the
DOM interfaces defined in [DOM2-CSS]
that correspond to aural properties
(See Relationship with DOM2 CSS object model.)
E.3 SVG accessibility guidelines
The definition of a Conforming SVG Generator
requires that it adhere to "Authoring Tool Accessibility Guidelines 1.0" [ACCESS-AUTHOR].
The following are additional SVG-specific accessibility guidelines:
SVG is a language for rich graphical content. For
accessibility reasons, if there is an original source
document containing higher-level structure and
semantics, it is recommended that that
higher-level information be made available somehow.
The latter information is necessary for purposes of
accessibility and intelligent textual processing.
- Provide access to the original source document which
convey's the document's structure and semantics.
- Express the document in another
markup language which conveys the document's structure
- Express the document using a combination SVG with other XML namespaces
where elements and attributes from other namespaces
convey the document's structure
and semantics and the SVG namespace is used to represent
the information graphically.
- Express the document solely using SVG, but utilize SVG's grouping constructs
(e.g., the 'g' element)
to represent document structure and use the
elements to describe the document semantics.
- Wherever supplemental descriptive information is available about part
of the SVG content,
it is strongly recommended that the descriptive information
be included within a
element. This guideline also applies to
elements, as the actual rendered text string might not convey the higher-level
structure and semantics of the character data content.
- Drawing programs often provide some sort of
document structuring capabilities, such as the ability to create and name
layers of graphics, and authors often assign meaningful names to these
structural components for their purposes of managing their own data.
For example, an author might create a drawing which has three layers:
a lake (the background layer), an island (drawn on top of the lake), and a
tree (drawn on top of the island), with the names of the layers being
"lake", "island" and "tree".
It is recommended that SVG content
convey as best as possible both the inherent structure of the
original drawing and any names of the structural components. In particular,
it is recommended that drawing tools which support the concept of named layers
create SVG content which create a distinct
for each layer and that a
element be provided for each layer's
element contains the name of the given layer.
- For SVG viewers, it is recommended that
elements be rendered as popup 'tooltips' when the pointing device is positioned
over the given graphics or container element. For visual rendering environments,
the tooltip might appear within a transient popup window. For aural rendering
environments, the tooltip might be rendered aurally.