Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1b) Gecko/20020830 (navigator) ctrl-l h <down> <down> expect magnify to follow the selection

On AIX, magnifier stays focused on the location input bar. The selection that is highlighted shows up in the location bar as well so you can tell where you're at. You just don't know what is before or after the current item assuming that item is off the screen. However, I would think this is the desired behavior since you want focus to remain where the cursor is which is actually in the location bar.

Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8a) Gecko/2004041917 just typing into the location bar does *not* affect what the magnifier sees. mlew@us.ibm.com: what happens if your magnifier is say 100x100 instead of very big. also, move your mouse somewhere interesting. steps: 1. mouse very far from your location bar 1' magnifier follows mouse 2. accel-l (focus location bar) 2' magnifier focuses on cursor 3. move mouse 1px 3' magnifier focuses on mouse 4. type something like "tinderbox" 4' magnifier doesn't focus 5. press right/left/up/down 5' magnifier focuses on where the caret would be in the urlbar (at the end of the selected url) 6. navigate up/down until you reach a url which is considerably longer than the urlbar 6' magnifier focuses on where the caret would be beyond the edge of the urlbar/window i suspect the current behavior is different from the behavior at the time i filed this report. for comparison, when you use the run dialog in explorer: A. if the item is longer than the input area, the right edge of the item is flush with the edge of the box. (in mozilla, the left edge of the location is flush with the left edge of the box.) B. typing as well as left/right arrows cause the magnifier's focus to shift.

1. ZoomText is a very well respected screen magnifer, been around for a very long time. It doesn't follow the autocomplete selection. 2. At least on my system, magnify.exe does not follow the autocomplete selection in Explorer's run dialog. 3. We're firing the correct focus events with the correct bounds. It's up to each assistive technology to decide how it wants to use them. We don't affect that. There's nothing for us to fix since we're already firing the right focus events.