Sorry, you're right about the gap being now also used for Flexbox, it's just badly documented.
As for the dynamic title thing, just tested it in NVDA and it works fine. I think the problem you're running into is that you're just expecting the screen reader to read out anything on the page that has changed. That won't happen, you have to make the browser aware that it needs to watch regions for changes. But NVDA has absolutely no problem reading out the new title.
If you have a valid reason to change the title for someone not using a screen reader, then it also applies for someone who is using one.
Sorry, you're right about the
gapbeing now also used for Flexbox, it's just badly documented.As for the dynamic title thing, just tested it in NVDA and it works fine. I think the problem you're running into is that you're just expecting the screen reader to read out anything on the page that has changed. That won't happen, you have to make the browser aware that it needs to watch regions for changes. But NVDA has absolutely no problem reading out the new title.
If you have a valid reason to change the title for someone not using a screen reader, then it also applies for someone who is using one.
I need to get my hands on NVDA - I need virtual machine for that
I'm assuming you're using a Mac then? VoiceOver should behave just the same. Or there's Orca and equivalents on Linux.