message proposal for the CSS WG about MathML

[Daniel Glazman]

Hello all,

Here is the proposal of message for the CSS WG I think we should
send. Comments highly welcome! Since all CSS WG technical chats
happen in www-...@w3.org, the message would be posted there.

</Daniel>
----

[EPUB WG Invited Expert hat on]

Hello CSS WG,

The EPUB Working Group has a rather big issue with MathML content
embedded into content documents inside an EPUB package: legacy
readers are unable to process MathML *including* the altimg attribute
on the math element that provides an image fallback. And content
publishers are not ready to embed MathML content if legacy readers are
not able to display a fallback. Added to that situation, JavaScript is
rarely available into such reading systems so fallback cannot be
triggered from embedded JS code. Since those legacy reading systems
can't be modified anyway, there is no available way for us to hide
mathml content and let only the fallback image, whatever the way
it is included, appear.

This is a rather blocking issue for scientific publishers at this time.

After long discussions in the EPUB WG, we came to the conclusion
a feasible way of solving the issue would be through a new Media Query
"detecting" MathML support. We could then assign 'display: none' to the
math element and a non-none display to an image fallback sibling
for legacy reading systems and have that new MQ assign a non-none
display value to the math element and 'display: none' to the image
fallback sibling if MathML is implemented.
Legacy reading systems would of course through away that new Media
Query, allowing the default styles hidding MathML content to work.

We then suggest the following MQ for level 4:

Name: 'mathml'
For: '@media'
Value: yes | no
Type: discrete

We may have to add 'presentational' and 'content' extra values to
query the support for the two flavors of MathML, I'm leaving that to
be discussed.

In substance, it's rather similar to the 'scripting' MQ and is also
based on a capability of the rendering engine.
Looking forward to reading your comments.

</Daniel>

[Romain]

Hi Daniel,

I think we've not reached consensus about the *preferred* solution yet; we still have a couple other solution candidates. Yesterday in our a11y call I was assigned the task to create a gh issue with a writeup of the problem and the various solution candidates (it will be published later today). Someone also suggested a dedicated call for interested people.

So, do you think we should already move forward with the CSS WG anyways, or should we rather wait to reach consensus on our preferred option first?

Romain.

> --
> You received this message because you are subscribed to the Google Groups "EPUB Working Group" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to epub-working-gr...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Ivan Herman]

> On 1 Sep 2016, at 11:00, Romain <rdel...@gmail.com> wrote:
>
> Hi Daniel,
>
> I think we've not reached consensus about the *preferred* solution yet; we still have a couple other solution candidates. Yesterday in our a11y call I was assigned the task to create a gh issue with a writeup of the problem and the various solution candidates (it will be published later today). Someone also suggested a dedicated call for interested people.
>
> So, do you think we should already move forward with the CSS WG anyways, or should we rather wait to reach consensus on our preferred option first?
>

I think starting a discussion with the CSS WG is useful because this type of MQ may become very important.

Ivan

----
Ivan Herman, W3C
Digital Publishing Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
ORCID ID: http://orcid.org/0000-0003-0782-2704

[Ivan Herman]

Hi Daniel,

one question, one comment…

The question: do you think that this would open up flood gates insofar as other, similar, MQ-s could be defined? An obvious one coming to my mind, again important for reading systems, is whether a Javascript interpreter is available or not for the end user…

The comment: alas! the content markup of MathML has been all but ignored by browser implementations, afaik, or the different libraries and, as a consequence, I do not think publishers ever use it:-(

Ivan

[Daniel Glazman]

On 01/09/2016 11:00, Romain wrote:

> Hi Daniel,
>
> I think we've not reached consensus about the *preferred* solution yet; we still have a couple other solution candidates. Yesterday in our a11y call I was assigned the task to create a gh issue with a writeup of the problem and the various solution candidates (it will be published later today). Someone also suggested a dedicated call for interested people.
>
> So, do you think we should already move forward with the CSS WG anyways, or should we rather wait to reach consensus on our preferred option first?

From what I read and heard, my own opinion is that the Media Query
solution is probably the best and simplest one. It will allow us
more flexibility and more possibilities in the future. And it's a
solution that would work in any rendering engine/browser, not only
reading systems. I also find it less hacky than others, given the
constraints publishers have. It's also simple enough to be implemented
and shipped rather fast. One nit: we may need a DOM-based way to set
the MQ to true in the case MathML is implemented through an add-on.

Apart from that, I have no opinion. This is a decision of the Working
Group, to be made by consensus. I was actioned to write a proposal and
submit it to the Group, so I did. I will of course not send the message
without EPUB WG's approval.

That said, I can phrase the message a little bit differently and say
the EPUB WG *considers* such a solution among others and would like to
hear from the CSS WG about it...

</Daniel>

[Avneesh Singh]

Indeed, we have not arrived at consensus yet. It will be an important topic
for us in coming weeks.
If we think that we should move ahead with CSS group, then we may like to
change the following sentence that is indicating conclusion.

"After long discussions in the EPUB WG, we came to the conclusion
a feasible way of solving the issue would be through a new Media Query"

With regards
Avneesh
-----Original Message-----
From: Ivan Herman
Sent: Thursday, September 1, 2016 14:42
To: EPUB Working Group
Subject: Re: message proposal for the CSS WG about MathML

[Daniel Glazman]

On 01/09/2016 11:14, Ivan Herman wrote:
> Hi Daniel,
>
> one question, one comment…
>
> The question: do you think that this would open up flood gates insofar as other, similar, MQ-s could be defined? An obvious one coming to my mind, again important for reading systems, is whether a Javascript interpreter is available or not for the end user…

You mean

https://www.w3.org/TR/mediaqueries-4/#scripting

?-)

In short, yes, there is clearly need AND use cases for MQ checking
the lower-layer capabilities of a rendering engine.

> The comment: alas! the content markup of MathML has been all but ignored by browser implementations, afaik, or the different libraries and, as a consequence, I do not think publishers ever use it:-(

Right. But since we never know what the future holds, it's good to
have the possibility to check for content-mathml support anyway.

</Daniel>

[Ivan Herman]

> On 1 Sep 2016, at 11:19, Daniel Glazman <daniel....@gmail.com> wrote:
>
> On 01/09/2016 11:14, Ivan Herman wrote:
>> Hi Daniel,
>>
>> one question, one comment…
>>
>> The question: do you think that this would open up flood gates insofar as other, similar, MQ-s could be defined? An obvious one coming to my mind, again important for reading systems, is whether a Javascript interpreter is available or not for the end user…
>
> You mean
>
> https://www.w3.org/TR/mediaqueries-4/#scripting
>
> ?-)

Ah, I did not know that! Brilliant...

>
> In short, yes, there is clearly need AND use cases for MQ checking
> the lower-layer capabilities of a rendering engine.

Great

>
>> The comment: alas! the content markup of MathML has been all but ignored by browser implementations, afaik, or the different libraries and, as a consequence, I do not think publishers ever use it:-(
>
> Right. But since we never know what the future holds, it's good to
> have the possibility to check for content-mathml support anyway.
>

We have to see. I have the impression that the content markup is essentially dropped.

But we are not yet there…

Thanks

Ivan

[Daniel Glazman]

On 01/09/2016 11:25, Ivan Herman wrote:

> We have to see. I have the impression that the content markup is essentially dropped.

By implementors, yes. By spec, not yet. And until the spec obsoletes
it...

</Daniel>

[Romain]

> On 01 Sep 2016, at 11:15, Daniel Glazman <daniel....@gmail.com> wrote:
>
> From what I read and heard, my own opinion is that the Media Query
> solution is probably the best and simplest one.

It may be (I am personally still ambivalent about all of these).

> Apart from that, I have no opinion. This is a decision of the Working
> Group, to be made by consensus. I was actioned to write a proposal and
> submit it to the Group, so I did.
>

> That said, I can phrase the message a little bit differently and say
> the EPUB WG *considers* such a solution among others and would like to
> hear from the CSS WG about it...

I agree with Ivan that starting the discussion early-on is useful, so +1 on going ahead if the CSS WG is fine with a "considered" use case!

Romain.

[White, Jason J]

> -----Original Message-----
> From: epub-work...@googlegroups.com [mailto:epub-working-
> I agree with Ivan that starting the discussion early-on is useful, so +1 on going
> ahead if the CSS WG is fine with a "considered" use case!

[Jason] +1, although I would like more effort to be devoted to implementing MathML support rather than to developing work-arounds for its absence or inadequacy.


________________________________

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.

________________________________

[Bill Kasdorf]

I agree. That is ideally a long term solution (but not likely to be available immediately) and in the meantime we can still work on our short-term workaround.--Bill K

-----Original Message-----
From: epub-work...@googlegroups.com [mailto:epub-work...@googlegroups.com] On Behalf Of Ivan Herman
Sent: Thursday, September 01, 2016 5:12 AM
To: EPUB Working Group
Subject: Re: message proposal for the CSS WG about MathML

[Bill Kasdorf]

Confirming that publishers virtually never use content MathML. I don't think most are even aware of it. Specifications rarely specify this: when a spec says "MathML" it is almost always (in the publishing context) taken to mean presentational MathML.--Bill K

-----Original Message-----
From: epub-work...@googlegroups.com [mailto:epub-work...@googlegroups.com] On Behalf Of Ivan Herman
Sent: Thursday, September 01, 2016 5:14 AM
To: EPUB Working Group
Subject: Re: message proposal for the CSS WG about MathML

[Bill Kasdorf]

Can anybody report on the use (or non-use) of content MathML by mathematicians, scientists, engineers, etc.? That who it's for, so that math can actually be interpreted, used, executed, etc. It's not for "showing" a mathematical expression, it's for "using" a mathematical expression. If those folks are using it, then I would hope the spec wouldn't drop it. I wouldn't know. I was an English major.--Bill K

-----Original Message-----
From: epub-work...@googlegroups.com [mailto:epub-work...@googlegroups.com] On Behalf Of Daniel Glazman
Sent: Thursday, September 01, 2016 5:29 AM
To: epub-work...@googlegroups.com
Subject: Re: message proposal for the CSS WG about MathML

[White, Jason J]

> -----Original Message-----
> From: epub-work...@googlegroups.com [mailto:epub-working-
> Can anybody report on the use (or non-use) of content MathML by
> mathematicians, scientists, engineers, etc.?

[Bill Kasdorf]

Excellent, thanks! When they talk about going from "printed math" to Braille it makes you think they would use presentational MathML, but I am delighted to see that they use content MathML. The reason is obvious: content MathML has semantics needed for accessibility that are lacking in presentational MathML. Thanks so much for letting me know about this.--Bill Kasdorf

[Tomoyuki Kudou]

Hi Bill,

# CC: Masakazu

Japanese Mathematician Dr. Masakazu Suzuki created ChattyInfty.

http://www.sciaccess.net/en/ChattyInfty/

This authoring system supports EPUB3 with MathML output.
It can handle not only mathematical expression but also chemical expression.

I called him just now. He mentioned me that many braille conversion system support MathML and braille conversion system can not convert mathematical expression from image, so EPUB3.1 should keep supporting MathML. it is a very important requirement for braille users. He pointed me that JAWS support MathML, it is a world wide used Windows screen reader, braille user can read MathML part with braille which is embedded in HTML.

Best wishes,
Kudou

[Bill Kasdorf]

Excellent, thanks!--Bill

[Daniel Glazman]

Hello,

Dave Cramer asked me to present the potential Media Query solution for
MathML to the joint CSS+DigiPub meeting at W3C TPAC in Lisbon a few
minutes ago.

The result is below:

RESOLVED: Add a MQ that reports MathML support, with some way for
scripts to claim support.
ACTION glazou to report to Epub about the MathML MQ being approved.
ACTION Florian to write the MathML MQ.

It does not say the EPUB WG has to adopt that solution. It only says the
CSS WG acknowledges the issue and offers to add a new Media Query for
detection of builtin or addon-based MathML support.

</Daniel>

[Daniel Weck]

For reference: https://github.com/IDPF/epub-revision/issues/845

/Daniel (Weck)