- From: Scott Luebking <phoenixl@netcom.com>
- Date: Wed, 11 Mar 1998 17:21:29 -0800 (PST)
- To: w3c-wai-ui@w3.org
Hi, Here's an example of division of labor between browser and screen reader. Suppose that the user has been navigating through the web page using keyboard navigation. Each time the browser changes the focus, it would be helpful if the screen reader could announce what kind of object was just highlighted. There are a couple of ways that the division of labor could be handled. One is that the brower tells the screen reader where in the HTML object the highlighted object is. The screen reader then has to figure out what the highlighted object is from the HTML object, get the appropriate information about the object, convert the information to speech format and then speak it. The disadvantage to this is that each screen reader has to know about HTML objects. In adition, as versions change, the screen reader has to know what version is being used. A different approach is that the browser tells the screen reader what is highlighted and provides the screen reader with a text string describing the object, e.g. "link - updated sales report". The screen reader just says the words in the text string. This division means that the screen reader has to know much less of HTML. A similar example is that the user wants to know more detail about the highlighted object. The user uses the "Info" menu item or short-cut key. The browser then gathers the appropriate details and pops up an information box. The screen reader then speaks the message/information in the information box. In this way, the screen reader only needs to know how to handle a message box and doesn't need to know about HTML objects. Does this type of division of work between browsers and screen readers make sense? Scott
Received on Wednesday, 11 March 1998 20:21:31 UTC