- From: Phill Jenkins/Austin/IBM <pjenkins@us.ibm.com>
- Date: Mon, 13 Nov 2000 23:55:08 -0500
- To: w3c-wai-ua@w3.org
- Cc: "Richard Schwerdtfeger/Austin/IBM" <schwer@us.ibm.com>, "Catherine Laws/Austin/IBM" <claws@us.ibm.com>
Checkpoint 1.2 Use the standard input and output APIs of the operating system. Do not bypass the standard output APIs when rendering information. P1. Clarification is need on how this applies to "cross-platform" user agents that may not rely on the operating system API for rendering text. For example: 1. User agents written in Java2, that do not use the operating systems' API, but still provide an excellent accessibility API to assistive technologies to access the information about the rendered text and other information. I would recommend that the phrase "of the operating system" should be deleted from the wording of the checkpoint. 2. Plug-ins, such as Adobe Acrobat and Macromedia Flash, that do or soon will provide published APIs to their player controls and content may not be using standard operating systems APIs, but provide efficient access for assistive technologies. The requirement is that they are published and efficient. 3. The following technique wording needs to be adjusted, perhaps by adding "published APIs" to where it mentions DOM: "Do not render text in the form of a bitmap before transferring to the screen, since some screen readers rely on the user agent's offscreen model. An offscreen model is rendered content created by an assistive technology that is based on the rendered content of another user agent. Assistive technologies that rely on an offscreen model generally construct it by intercepting standard system drawing calls. For example, in the case of display drivers, some screen readers are designed to monitor what is drawn on the screen by hooking drawing calls at different points in the drawing process. While knowing about the user agent's formatting may provide some useful information to assistive technologies, this document emphasizes access to the document object model rather than a particular rendering. For instance, instead of relying on system calls to draw text, assistive technologies should access the text through the document object model. " Regards, Phill Jenkins IBM Research Division - Accessibility Center
Received on Tuesday, 14 November 2000 00:00:09 UTC