NVDA should speak the TITLE of iframes and frames

[nvaccessAuto]

Reported by vtsaran on 2010-05-21 05:56
It seems that currently NVDA ignores TITLE attributes on HTML iframes and frames.
Blocking #5432

[nvaccessAuto]

Comment 1 by jteh on 2010-05-21 06:19
How do sighted users see these titles? Ideally, how would you like NVDA to report them?

[nvaccessAuto]

Comment 2 by jeronimo on 2010-05-21 13:41
Changes:
Milestone changed from None to 2010.2

[nvaccessAuto]

Comment 3 by vtsaran (in reply to comment 1) on 2010-05-25 04:53
Replying to jteh:

How do sighted users see these titles?

Sited users do not see those titles at all. The TITLE attribute for an iframe works conceptually similar to an ALT attribute on an image.

Ideally, how would you like NVDA to report them?

Could be something like:

• <title> iframe or
• iframe <title>

[nvaccessAuto]

Comment 4 by mdcurran on 2010-05-25 05:37
Should the title appear as content in the virtualBuffer, (i.e. you would be able to arrow through it), or should it just be announced along with "iFrame", i.e. its meta information related to the field.

[nvaccessAuto]

Comment 5 by jteh (in reply to comment 3) on 2010-05-25 06:18
Replying to vtsaran:

Sited users do not see those titles at all. The TITLE attribute for an iframe works conceptually similar to an ALT attribute on an image.

An alt attribute is only provided because the content in question (the image) cannot be described in a textual (and thus non-visually accessible) form. IN the case of an iframe, the content of the iframe is already accessible, so there shouldn't be any need for any additional meta information. I can't see a good justification for exposing this.

[nvaccessAuto]

Comment 6 by katsutoshi on 2010-05-25 07:06

IN the case of an iframe, the content of the iframe is already accessible, so there > shouldn't be any need for any additional meta information. I can't

see a good justification for exposing this.

Title of the each frames or iframes helps us deciding which frame should we read carefully.
For instance, if we don't want to read frames contains advertisement and frame title is set to"advertisement" allready, we are able to skip it.
In that case, we need to know each frame title to decide which frame to be skip.

==== Possible related links ====

• H64: Using the title attribute of the frame and iframe elements | Techniques for WCAG 2.0
• H70: Using frame elements to group blocks of repeated material | Techniques for WCAG 2.0

[nvaccessAuto]

Comment 7 by jteh (in reply to comment 6) on 2010-05-25 07:56
Replying to katsutoshi:

Title of the each frames or iframes helps us deciding which frame should we read carefully.

For instance, if we don't want to read frames contains advertisement and frame title is set to"advertisement" allready, we are able to skip it.

So why don't sighted users need a similar paradigm? I'd suggest that this is possibly better suited to landmarks. That is, main content should be marked as main, or perhaps we could have a new landmark for advertisement.

[nvaccessAuto]

Comment 8 by jteh on 2010-07-14 07:15
Changes:
Milestone changed from 2010.2 to 2010.3

[nvaccessAuto]

Comment 9 by mdcurran on 2010-12-01 06:33
Changes:
Milestone changed from 2011.1 to None

[nvaccessAuto]

Comment 10 by MABennett3 on 2014-06-20 17:15
I recently encountered NVDA not announcing the title attribute placed on an iframe element while testing a page using v2014.2 and Firefox 30. A quick search directed me to this existing ticket and, based on the conversation in the comments, it appears that there was a decision back in 2010 that this was working as intended and not a bug; however, the issue is not marked as 'Closed'.

As an accessibility consultant with a US federal government client responsible for complying with Section 508 1194.22(i), I wanted to follow up on this ticket to see if the NVDA team has reconsidered their stance and has plans to implement this functionality or if this is still considered working as intended and the bug just needs to be closed out.

[nvaccessAuto]

Comment 11 by jteh on 2014-12-02 06:02
We're happy for this to be implemented, but it's not a priority right now.

[feerrenrut]

P3. There seems to be some demand for this, and its possibly fairly quick to implement.

[seabizquik]

As an accessibility analyst for the UC, this comes up again and again. NVDA users very often use firefox, which still lets users tab to an iframe. Currently, NVDA announces nothing when focus is on an iframe in Firefox. Seems like this should be implemented.

[Adriani90]

Does anyone have a minimal test case in Codepen or so for this?

Based on the comments above, it seems especially Firefox is impacted by this. But I wonder whether it is common understanding in the web design and web developers communities to use clear titles for frames and iFrames? Or do people in practice rather use landmarks, grupings and separators?

cc: @jcsteh