Testing

Testing for accessibility can be complicated for several reasons. It must be performed separately with respect to different user groups with different needs and often it involves the intaraction between an assistive technology (AT) application and a standard application like Firefox. To simplify testing and encourage wider participation we have divided the testing scheme into some simple tests for basic desktop tasks and more in-depth tests of the assistive technology applications like Orca and Onboard.

Generic User Descriptions

The Gnome project has a testing page that defines some useful Generic User Descriptions (GUDs). We have adopted these descriptions for our general testing plan.

v1

Lesser Visual Impairment - The user has a lesser degree of visual impairment and is assisted by an accessibility-enhanced theme, such as High Contrast Large Print. The user can use a standard keyboard and mouse, but requires increased definition, size and clarity of the widgets, icons and fonts on the desktop.

v2

Moderate Visual Impairment - The user has a moderate degree of visual impairment and uses a screen magnifier. This person can use a standard keyboard and mouse, but requires an enlarged view of sections of the desktop. This view is displayed on a small portion of the screen or on a separate display.

v3

Blindness - Requires a screen reader (such as Orca), and/or a braille I/O device. This person cannot use a standard mouse setup and requires a keyboard for program navigation and interaction with a screen reader to parse applications and their widgets.

m1

Minor Motor Difficulties - The user requires some software adaptations such as AccessX to use a standard keyboard or mouse. This can include Sticky Keys to help with using modifier keys, Mouse Keys to emulate a mouse using the keypad.

m2

Severe Motor Difficulties -- pointing devices - Those not able to use a standard keyboard may rely on a range of pointing devices and special software such as Onboard or Dasher for text entry. Operations can be enacted by a switch or by dwell selection. For testing purposes, this level of functionality can be approximated by using only the mouse with and without the ability to click the mouse buttons.

m3

Severe Motor Difficulties -- switch devices - Some people interact with their computers using one or two switches. These may be simple mechanical devices or pressure operated. Interaction with the desktop environment occurs through an on-screen keyboard that allows scanning which starts and stops (on a letter or command) in response to switch signals. A switch may be emulated using a mouse where the ball has been removed. (Though the hardware should also be tested -- are these serial devices?)

General Testing

The GNOME accessibility testing page contains 38 test cases which we will use as inspiration for our general tests. However we will settle for a smaller number of tests and some will be more application-specific, focusing on the key applications in Ubuntu like Firefox, Evolution and OpenOffice.org. The following is a matrix of tests results. More details about each test, comments and bug numbers can be found by clicking on the test name. Since many other things can break accessibility, we need to insure these tests are completed for each release of Ubuntu. Testing for a single release without following up for the next release can leave each item in an unaccessible state.

Task

v1

v2

v3

m1

m2

m3

LoggingIn

SendingEmail

OpenOffice

Warning /!\ (4) HNO

Firefox

Nautilus

UsingIRC

  • Ubuntu - Works perfectly | Warning /!\ - Works somewhat, but there are issues | (X) - Fails
    Requires: (1) Dasher | (2) Orca | (3) Onboard | (4) High visibility theme

How to test

As you perform these tasks for one or more of the Generic User Descriptions, ensure that you can get information from the interface, navigate and enter information. Please note any inconsistencies, like missing icons or actions that are more inconvenient than they should be. When you find problems, please report them in the Ubuntu bugzilla, Malone or an upstream bug tracker and place a link to the bug page on the test result page.

Application Testing for Accessibility

Many non-disable people as well as developers would like to test and make sure their applications are accessible. There is no requirement to actually have a disability to test for that disability. While there are many good tests in Gnome testing, we will provide guidance for a few tests. These are not meant to be the only tests used, but instead as a guide to insure minumum requirements can be met. The following tests are to be used by developers to test their applications for accessibility before release.

  • The audio test is for people with hearing loss or visual impairment.

  • The color and contrast tests are for people with visual impairment including color blindness.

  • Those writing or maintaining any type of documentation, directions, or websites will use the documentation tests.

  • The keyboard navigation tests are for people with motor difficulties.

  • The screen-reader test will be used for people with visual impairments or blindness.

  • Those wishing to insure that their software can be called fully accessible will use the User Interface Checklist

Assistive Application Reviews

More detailed tests and reviews will be done of the specific assistive technology applications (see reviews).

CategoryAccessibility

Accessibility/Testing (last edited 2011-06-27 12:20:01 by t22916)