- From: Dominic Mazzoni <dmazzoni@google.com>
- Date: Fri, 8 Feb 2019 09:11:29 -0800
- To: "White, Jason J" <jjwhite@ets.org>
- Cc: Joanmarie Diggs <jdiggs@igalia.com>, Peter Krautzberger <peter@krautzource.com>, "public-aria@w3.org" <public-aria@w3.org>
- Message-ID: <CAFz-FYzamSHyW8FG8PBkCi_RxXtE8zXxR_d+chNMx81HPp4uyQ@mail.gmail.com>
I love this idea. Some questions:
Can we apply this to every ARIA property that exposes localizable text and
not just aria-label and aria-roledescription? I think aria-valuetext and
aria-placeholder are the other two that would make sense.
Perhaps we should explicitly include attributes for speech, as well as
braille? I don't think we want authors to be making the assumption that
aria-label is used for speech by default, even though that's the most
widely-used application of it today. Aria-label could be used for voice
control, the string that somebody should speak in order to activate a
control. So I'd propose something like <button aria-label="Call an Uber"
aria-labelspeech="Call an Oober" aria-labelbraille="⣏⣹"> - in other words,
aria-label should be the default string that could be shown visually or
when spelled out, aria-labelspeech is a pronunciation hint, and
aria-labelbraille is a braille hint.
On Fri, Feb 8, 2019 at 7:18 AM White, Jason J <jjwhite@ets.org> wrote:
> The Unicode braille is also correctly displayed via JAWS and NVDA even
> with Grade II translation enabled. I haven't yet tested Chrome OS/ChromeVox
> or iOS/Voiceover, and I don't currently have access to an Android device
> with BrailleBack or BRLTTY installed. Although I can access a device with
> Amazon's VoiceView, I don't possess a supported braille display (it works
> with HumanWare and Baum protocols only - and my Baum display from circa
> 2005 appears to have hardware problems).
>
> -----Original Message-----
> From: White, Jason J <jjwhite@ets.org>
> Sent: Thursday, February 7, 2019 2:06 PM
> To: Joanmarie Diggs <jdiggs@igalia.com>; Peter Krautzberger <
> peter@krautzource.com>
> Cc: public-aria@w3.org
> Subject: Re: properties for exposing custom Braille descriptions
>
> I wrote the Python code below to create Unicode braille files from arrays
> of integers representing cell contents. I then created a file and accessed
> it in gedit with Orca.
>
> The Unicode characters (encoded as utf-8) were correctly displayed, even
> though UEB translation was enabled. The speech output also announced the
> dot patterns for each character.
>
> On the Mac, I had to turn off contracted braille (i.e., Grade II
> translation) to display the Unicode characters.
>
>
> # Convert braille cell dot specifications to Unicode braille.
> def cellToUnicode(cell):
> """Convert an array of integers (1-8) denoting dots to the
> corresponding
> Unicode braille code point."""
> dots=0x2800
> if len(cell) > 8:
> raise ValueError("Cell too large:", len(cell))
> for dot in cell:
> if dot < 1 or dot > 8:
> raise ValueError("Dot out of range", dot)
> dots |= 1 << (dot-1)
> return chr(dots)
>
> def cellsToBraille(cells):
> """Convert an array of cells to a UTF-8 braille Unicode string."""
> output=""
> for cell in cells:
> output += cellToUnicode(cell)
> return output.encode('utf-8')
>
> def writeUnicodeBraille(filename, cells):
> """Write the specified cells as Unicode braille to the file given."""
> output=cellsToBraille(cells)
> with open(filename, 'wb') as outputFile:
> outputFile.write(output)
>
>
> On 2/7/19, 12:13, "Joanmarie Diggs" <jdiggs@igalia.com> wrote:
>
> Thanks Jason!
>
> We don't need a page per se. It should be sufficient to put some
> braille
> unicode characters in a text editor like gedit or notepad.
>
> --joanie
>
> On 2/7/19 12:07 PM, White, Jason J wrote:
> > I can perform the test on multiple platforms, including
> Linux/GNOME/Orca. Is there a test page already, or should I create one?
> >
> > On 2/7/19, 12:00, "Joanmarie Diggs" <jdiggs@igalia.com> wrote:
> >
> > Indeed, for translation. But Unicode braille characters
> shouldn't be
> > translated. So one question is: Is liblouis smart enough to not
> > translate those? The second question is: Is the library used to
> > communicate with the display (in my case that would be BrlTTY)
> smart
> > enough to handle those characters?
> >
> > I would hope that the answer to both questions is "yes." And I
> went to
> > verify that this morning for my platform, but when I powered on
> my
> > ancient braille display, it made all sorts of noises and gave up
> the
> > ghost. :(
> >
> > A quick test I did suggests liblouis *might* be translating those
> > characters, but I'm not positive. Since you (Jason) happen to
> use Orca,
> > and presumably have a not-dead refreshable braille display,
> could you
> > please do the test for me to answer both the first question and
> the
> > second question?
> >
> > You should be able to answer the second question by disabling
> > translated/contracted braille. That will cause Orca to send the
> > characters directly to BrlTTY unmodified. If what's on the
> display
> > matches the unicode braille dot patterns in the test text, then
> we know
> > BrlTTY is good.
> >
> > To answer the first question, re-enable translated/contracted
> braille.
> > If those unicode braille dot patterns are still preserved, then
> liblouis
> > is also good. If liblouis isn't good, we should fix that in
> liblouis,
> > but that's something I can work around in Orca by looking for
> characters
> > in the braille dot pattern range and not sending them to
> liblouis. As
> > far as I am aware, I don't have control over what BrlTTY does
> with the
> > characters I give it.
> >
> > Thanks in advance for doing these tests. And sorry for having to
> ask
> > you. My braille display picked a fine morning to kick the bucket.
> >
> > --joanie
> >
> >
> > On 2/7/19 9:06 AM, White, Jason J wrote:
> > > Besides Apple, I think everyone is using liblouis at this
> point for translation. I don't know what Apple's Mac OS and iOS
> implementations use. I think it would be reasonable to expect everyone to
> support the Unicode code points for braille rendering.
> > >
> > > On 2/7/19, 09:04, "Joanmarie Diggs" <jdiggs@igalia.com> wrote:
> > >
> > > Do we know if all the braille libraries used by the
> platforms support
> > > Unicode block?
> > >
> > > On 2/7/19 8:57 AM, White, Jason J wrote:
> > > > Braille ASCII varies between countries, hence between
> the braille tables
> > > > loaded by screen readers – so it can’t be used to
> specify a desired
> > > > braille representation reliably.
> > > >
> > > >
> > > >
> > > > For example, if the author assumes North American ASCII
> braille, but the
> > > > user’s system is configured for one of the European
> codes, the result
> > > > won’t be what the author would expect.
> > > >
> > > >
> > > >
> > > > Only the Unicode block is unambiguous.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > *From: *Peter Krautzberger <peter@krautzource.com>
> > > > *Date: *Thursday, February 7, 2019 at 08:43
> > > > *To: *"White, Jason J" <jjwhite@ets.org>
> > > > *Cc: *"public-aria@w3.org" <public-aria@w3.org>
> > > > *Subject: *Re: properties for exposing custom Braille
> descriptions
> > > >
> > > >
> > > >
> > > > Hi Jason,
> > > >
> > > >
> > > >
> > > > This question is mentiond in the list of questions at
> the end of the
> > > > page. My apologies if the page was not accessible enough
> to make it there.
> > > >
> > > >
> > > >
> > > > Speaking for myself, I'd very much hope that Unicode
> Braille would work.
> > > > But Braille ascii might be a useful consideration. As I
> understand it,
> > > > NVDA's prototype back in May pushed the raw value out
> (and Sina's test
> > > > files used grade 1).
> > > >
> > > >
> > > >
> > > > Best regards,
> > > >
> > > > Peter.
> > > >
> > > >
> > > >
> ------------------------------------------------------------------------
> > > >
> > > > This e-mail and any files transmitted with it may
> contain privileged or
> > > > confidential information. It is solely for use by the
> individual for
> > > > whom it is intended, even if addressed incorrectly. If
> you received this
> > > > e-mail in error, please notify the sender; do not
> disclose, copy,
> > > > distribute, or take any action in reliance on the
> contents of this
> > > > information; and delete it from your system. Any other
> use of this
> > > > e-mail is prohibited.
> > > >
> > > >
> > > > Thank you for your compliance.
> > > >
> > > >
> ------------------------------------------------------------------------
> > >
> > >
> > >
> > >
> > > ________________________________
> > >
> > > This e-mail and any files transmitted with it may contain
> privileged or confidential information. It is solely for use by the
> individual for whom it is intended, even if addressed incorrectly. If you
> received this e-mail in error, please notify the sender; do not disclose,
> copy, distribute, or take any action in reliance on the contents of this
> information; and delete it from your system. Any other use of this e-mail
> is prohibited.
> > >
> > >
> > > Thank you for your compliance.
> > >
> > > ________________________________
> > >
> >
> >
> >
> >
> > ________________________________
> >
> > This e-mail and any files transmitted with it may contain privileged
> or confidential information. It is solely for use by the individual for
> whom it is intended, even if addressed incorrectly. If you received this
> e-mail in error, please notify the sender; do not disclose, copy,
> distribute, or take any action in reliance on the contents of this
> information; and delete it from your system. Any other use of this e-mail
> is prohibited.
> >
> >
> > Thank you for your compliance.
> >
> > ________________________________
> >
>
>
>
>
> ________________________________
>
> This e-mail and any files transmitted with it may contain privileged or
> confidential information. It is solely for use by the individual for whom
> it is intended, even if addressed incorrectly. If you received this e-mail
> in error, please notify the sender; do not disclose, copy, distribute, or
> take any action in reliance on the contents of this information; and delete
> it from your system. Any other use of this e-mail is prohibited.
>
>
> Thank you for your compliance.
>
> ________________________________
>
> ________________________________
>
> This e-mail and any files transmitted with it may contain privileged or
> confidential information. It is solely for use by the individual for whom
> it is intended, even if addressed incorrectly. If you received this e-mail
> in error, please notify the sender; do not disclose, copy, distribute, or
> take any action in reliance on the contents of this information; and delete
> it from your system. Any other use of this e-mail is prohibited.
>
>
> Thank you for your compliance.
>
> ________________________________
>
Received on Friday, 8 February 2019 17:12:14 UTC