If I understand what is wanted, when moving to the check box, the person
who the message is from is wanted to be read before anything else.  I'm
not sure if check box checked or unchecked is wanted to be read or
when.  Screen-readers exist in complex environments and there are times
when, for whatever reasons, human intervention is needed to get what you
want read when you want it read.  Sometimes, this isn't possible and
sometimes it is.  In this case, all I have to do in the simple HTML view
is to type the letter x to move to a check box, then immediately down
arrow.  The sender of the message is read immediately.


Implementing a feature where the first line below a check box is read in
general or at least on some web sites may not be a good solution.  It
may do what is wanted on this or that site and not on others where it
may create unwanted verbosity.


I suppose it might be tailored for a specific site but something like
that, I would think, would probably be done by an Add-on, such as a
GMail add-on.

But this concept of helping a screen-reader is important and my
impression is that a lot of users don't do this.


I'll give two examples:

When I use NVDA's browse mode search, if I don't use read current line,
I hear all sorts of unwanted verbosity before what is found is read.  If
I issue the search command, then use the read current navigator
;position, numpad 8 in  the desktop layout, what is wanted is read
immediately.


I don't recall now if I issue a find command, wait for NVDA to start
speaking, then issue the read current line command, or if I issue the
find command, then wait a moment and issue the read current line command
even before speech has begun.  You can experiment and see if both work.


This greatly improves the efficiency of the Browse Mode find command and
it becomes automatic after a little while, which is the reason I don't
recall the exact time when I issue the command.


When you repeat a search, numpad insert, f3 on the main keyboard, what
is wanted is read without doing anything else.  That is the desktop
layout command for repeat search.


Here is the other example:

In Outlook and in Thunderbird, I have seen behavior complained of such
as when you open a message, you hear unwanted information read or read
and repeated.  The solution is to open the message, then issue the read
to end command.  You can experiment.  You may have to wait a moment,
just how long you  can determine before the command is able to work,
probably depending on when the message finishes loading.  You don't have
to wait until speech automatically begins.  Using this command becomes
automatic after a little while and greatly increases efficiency and
causes you to hear little or no unwanted verbosity.


Gene

On 12/3/2020 12:33 AM, Bob Cavanaugh wrote:

I'm not sure if I made my point clear enough on this. I am used to the
way System Access reads things, which is that when tabbing through and
landing on the checkboxes, it reads who the message is from right away
without having to use any extra table commands. Is there some way this
can be implemented in NVDA? It seems like there are situations like
this all over the web.

On 12/2/20, Chris Mullins <cjmullins29@gmail.com> wrote:

Hi Bob
I can’t comment on the Outlook issues but regarding the Gmail basic HTML
view, when you are browsing the list of messages, you are in a table, so
use
the table reading commands rather than just tabbing. These are
Control+alt+left/right/up/down arrows, so, for example you can move up
and
down the subject column to read them. Personally, I hate the gmail
interface and have linked my gmail account to the W10 mail app and find
it
much easier to manage.

Cheers
Chris

From: Bob Cavanaugh
Sent: 02 December 2020 05:32
To: nvda@nvda.groups.io
Subject: [nvda] two different problems with NVDA and reading email

H