W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 1998

Re: Vector Graphics

From: Al Gilman <asgilman@access.digex.net>
Date: Wed, 29 Jul 1998 16:08:32 -0400 (EDT)
Message-Id: <199807292008.QAA06935@access5.digex.net>
To: burke@ucla.edu (Patrick Burke)
Cc: w3c-wai-ig@w3.org
to follow up on what Patrick Burke said:

> I just heard of vector graphics for the first time ever. Apparently a
> graphic format that could replace bitmaps, and also apparently being
> proposed as a specification to the W3C.
> Does anyone know more about these and particularly any positive/negative
> impact on accessibility?

Vector Graphics is actually older than raster graphics, in many
application areas.  Vector graphics work like a plotter whith a
pen that moves in two dimensions as it draws.  The "vector" in
vector graphics is "where to move the pen next," which can be
anywhere on the page.

Raster graphics work like an old WirePhoto machine.  The pen
sweeps straight across the page and is either up or down to make
black and white pixels in that row.  Then the paper inches
forward and another row of black and white dots is laid down.
These horizontal rows of pixels are the "rasters" this kind of
format is named after.

Current laser and ink jet printers are a raster format where the
rubber meets the road.  FAX (tiff) is a raster format and CGM and
PostScript are vector formats.

As Daniel explained, vector graphics formats are friendlier to
zooming and retain more structure than raster formats.

For example, if you have a scene from a virtual furniture
showroom, and there is a chair for sale, there is a set of draw
instructions which is the representation of the chair.  In a
raster format, there is no subset of the commands that you can
lay your hands on easily that is "the chair."  So while the chair
as represented in the vector format may be inaccessible if it
only contains the geometry of the draw instructions, at least the
vector graphics format gives you a logical place to hang the
verbal attribute "chair" and maybe a make and model.

Even better, from an understanding point of view, are VR formats
such as VRML and Java3D where the system knows the 3D shape of
the chair and generates what goes on the the screen by rendering
the chair's solid shape in terms of a user-controlled point of

This VR style of format would let you push a virtual grocery cart
through the virtual grocery store and when you run it into the
shelves you hear the sound coming from the point of impact.  So
you know which way to steer away from on the next push.

Since aural perception gives you a sense of three dimensions, if
user interfaces migrate to three dimensions there is some
potential for accessibility gains and richer eyes-free
computation and browsing.  This is why Raman put spatial stuff in
ACSS so that the sound can come from different places.


Received on Wednesday, 29 July 1998 16:08:22 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:01 UTC