GNOME Bugzilla – Bug 561722
Structural Navigation on really huge web pages can be quite slow
Last modified: 2012-12-21 17:14:09 UTC
Given a really huge web page structural navigation, be it via collection or not, can be quite slow due to the examination of the page while searching for objects. We should find a way to speed this up.
While I'm mostly happy with the Orca/FF browsing experience, thiss is definitely one of my biggest frustrations. Out of curiosity, would it be possible to provide a way to abort a particularly long structural navigation? I've already made this mistake several times. I go to the advanced bug form and press insert-tab, thinking it's a form so I'll go to the first field, only I'm stuck waiting a minute or so if I'm on my netbook for Orca to find no form fields and return control to me.
We are late in the 2.28 release cycle and I want to focus on "high impact"/"low risk" items that also fall within the release team's restrictions in place. Regretfully, this bug doesn't fit well within those constraints and we'll review it for the 2.29 release cycle.
When I used to use newsgroup contents in http://members.easynews.com, I would run into very long load times even before I could look through the page contents. I haven't been there for quite some time but I just went back to easynews the other day and spent a fair amount of time there while using Firefox version 3.5.7. I have to say, the load times improved dramatically since my times in the past. I didn't use a lot of structural navigation commands with this particular site but my cursoring around never faltered. So I'm not sure if this but is still a proglem for folks or not. I do notice a considerable lag when using structural navigation keys to go backwards up the page, such as going back to a previous header.
Where this bug is problematic is when you have a really huge table. There are some issues with the Gecko implementation of the accessible collections interface.
A fix for one of the collections issues was made by Mike Gorse to AT-SPI. We should re-evaluate this to see what else can/should be done.
Planning spam. Sorry!
(In reply to comment #5) > A fix for one of the collections issues was made by Mike Gorse to AT-SPI. We > should re-evaluate this to see what else can/should be done. AT-SPI2? can you provided the improvement commit in AT-SPI? Thanks!
The performance has been improved. And there's nothing we can do further within Orca. We will be implementing an AtkCollection interface. Then Gecko (and other toolkits) can in turn implement it.