| Section | Commenter | Comment | Response |
|---|---|---|---|
| General | INSTITUTO NACIONAL PARA CIEGOS-INCI | We have read this information , and we think this development is quite important for all visually impaired people. At the moment we do not have any comments about the draft. | |
| General | Swedish Library of Talking Books and Braille | In general I think there is a bit confusion on what it really means
to be MathML-unaware. I believe the players need to be divided into three
groups: * Players that are MathML-unaware. They know nothing about MathML and have no idea they need to apply a stylesheet to transform the math to an image group. * Players that are MathML-aware but do not support MathML. This category of players will apply the XSLT. * Players that are MathML-aware and support MathML. |
Will add these groups to the Purpose and Scope section where we already talk about the issues. |
| General | Design Science | "braille" is capitalized inconsistently in the document. | Capitalize only when referring to Louis Braille. |
| Foreward | Universität Karlsruhe | (Enable simple playback...): That does only cover the audio part of DAISY books. As it would be technically possible to have a player rendering MathML to Speech (first serializing, then using a TTS system) that should be noted. | Add another bullet point about simple playback devices that can synthesize the audio from the math/alttext. |
| Foreword | Universität Karlsruhe | (Enable software players...): Here only mathematics is mentioned, but perhaps the spec should say that a lot of more extensions would have the same problem, so instead of mathematics one could say "some extension, i.e. mathematics" after "need to be aware of". I know that the spec deals only with math, but I strongly expect that future specs will take the math spec as basis for their work. | Add a sentence or two to Strategy to elaborate on the existing bullet point "Any solution should be able to be generalized to other extensions".. |
| Foreword | Swedish Library of Talking Books and Braille | The last paragraphs of the foreword describe the problems in trying to emulate intelligent reading system. This is a bit unclear at first, which I think could be helped by concatenating the paragraph that ends with "...for several reasons" with the following one, "The working group believes that" | Should accept suggestion. |
| Section 1.1 | Universität Karlsruhe | At the end of that section the spec says: "Such a player could allow...". In my opinion the ability to explore math content in non-audio-parts should be mentioned, as future players could/will allow users to explore math content on the braille display and on the screen. | Change 'Such a player..." to "A MathML-aware player could
allow a user to explore the structure of the expression tactilely using
a refreshable braille display and/or with audio without having to listen
to the expression in its entirety.".
We should also add: "A future version of this specification may add finer SMIL granularity within the math audio stream." |
| Section 1.2 | Design Science | Links point off to DAISY Spec. Should be local. | Agree. |
| Section 1.2 | Universität Karlsruhe | I think that the definition of "Global navigation" is misleading. What the spec describes is in fact a local navigation, only at a rougher granularity as what is described in "Local navigation" in the spec. I would like to have "Global navigation" defined in a way that real links to external DAISY books could be possible. Just think of an exercise of a math teacher linking to chapters of the math book for students! Of course, some resource locater definition would have to be found, just like the URL of http. Perhaps it could be in the for of "daisy://ISBN1234567/..." where the ISBN-number is being used. How to find such a reference would then be part of the players capabilities (for example, take all local DAISY books and index /use them. If such a link cannot be used locally, try to access the DAISY book online at a digital library). | Although good comments, the definitions of global and local navigation are taken from the DAISY spec and the comments are out of scope for the MathML extension. |
| Section 1.3 | Universität Karlsruhe | At "The third principle" it should be noted to which mode (audio, braille, screen) "a simple prepared presentation" refers. One could use that in all three modes, with different options of course. | Added to the end of "...to present a simple prepared presentation" so that it becomes "to present a simple prepared presentation with an image, alt-text, and/or an audio file". |
| Section 1.3 | Universität Karlsruhe | I also do not understand what you mean by "display MathML visually". Do you mean a form that is used for sighted people (like in a math book), do you mean some serialized form (like LAMBDA or Nemeth) or could the player do what it/the user wants? | To make this clearer, added "(i.e., as traditional mathematical notation)" . |
| Section 1.3 | Universität Karlsruhe | Additionally, in the last line of that section the spec could note "...can create different braille math notations such as LAMBDA, Nemeth, AMS and LaTeX from the MathML" as that section is informative and the single reference to braille math codes could also refer to the differences between German, US, ... braille codes as well as 6-dot and 8-dot codes. I think that the expression braille math notation is more clear. | Adopt the suggestion so that the sentence "...can create braille math codes from the MathML" becomes "...can create braille math codes such as LAMBDA, Nemeth, Marburg, and LaTeX from the MathML" |
| Section 1.3 | Universität Karlsruhe | At the end of that section the spec could add a remark that players could i.e. take care of different braille notations for different countries. | With the above addition, the point is accounted for. |
| Section 2 | Dedicon Netherlands | XSLT file is not the same as the others | Remove as a list item and make into a paragraph. |
| Section 3 | Dedicon Netherlands | Does ".xslt" need to be the suffix? | Change to be 'content="xslt-file-name"' and change
text afterwards to:
< "'xslt-file-name'" must be replaced by the name of the name of the xslt file being used. All other attributes values must be as listed above. |
| Section 3.1.2 | Universität Karlsruhe | (Note): Text and graphic presentation are mentioned, but what happens with audio content? I still think about someone making a player that does TTS in real time. | Change "...but can present the altimg and the alttext attributes
given in the MathML" to "...but can display the altimg and/or
synthesize speech from the alttext attributes given in the
MathML." |
| Section 3.2 | Swedish Library of Talking Books and Braille | What is the mime-type of the XSLT in the OPF file? | Add:
The MIME type of theXSLT must be "application/xml". |
| Section 4.2 | Swedish Library of Talking Books and Braille | Since a new public ID is declared for DTBook+MathML, why should the
system ID still point to the plain DTBook DTD and then include the extension
in the internal subset? It seems more logical to either: A. Use the plain DTBook public ID and include the extension in the internal subset, or B. Use the new public ID but let the MATHML extension be declared in a new system ID. |
Markus to talk to Linus about suggestions. |
| Section 4.3 | Swedish Library of Talking Books and Braille | The example is missing a namespace declaration for the dtbook:smilref attribute. | Fixed. |
| Section 5.4 | Swedish Library of Talking Books and Braille | The short SMIL example is followed by a description of the full SMIL example found in the appendix. That is a bit confusing. | The description is for the current example (which is a short version of what is the appendix). It is correct as is. |
| Section 6 | Universität Karlsruhe | A dot is missing at the end. | Added the "." |
| Section 8.1 | Universität Karlsruhe | There is an additional line break. | Removed the line break. |
| Section 9 | Swedish Library of Talking Books and Braille | First paragraph, "Furthermore, cultural differences and users preference...": s/users/user/ |
Fixed. |
| Appendix 3 | gh, LLC | Note, however, that the Appendix 3 text states, "This Appendix contains
a small but complete example of a MathML-extended DTBook." but the WAV
amd MP3 files are not included. So this is NOT a complete example. We
should either remove the use of the audio files in the example or supply
the files. Of course, the files would be included in the zip file but
not the content not shown in the Appendix so we would change the sentence
(just prior to the Appendix 3.1 heading): "A zip file containing these
files can be found at ..." to "A zip file containing the complete example
can be found at ...". We could also give an incomplete example by informing the reader that the audio files are referenced but not included. |
Modified the intro to add: "Image and sound files are not included in this appendix but are included in the zip file referenced below." |
| Appendix 3.1 | gh, LLC | Package File: The SMIL file refer to two audio files: nativemathml0001.mp3 and nativemathml0002.mp3. The Resource file refers to two other audio files: formula.wav and matte.wav. These file names should be included in the manifest of the OPF file. |
Added the reference to the sound file. |
| Appendix 3.2 | Swedish Library of Talking Books and Braille | First paragraph, "The MathML is typically contained in this file...": Is there another option? | Dropped "typically". The MathML can not be in a separate file in this proposal. |
| Appendix 3.2 | Swedish Library of Talking Books and Braille | DTBook example: Doctype system ID references a local DTD, should probably point to the DTD at daisy.org. Otherwise the DTD is missing from the package file. |
It (and others) point to NISO: "<!DOCTYPE dtbook PUBLIC "-//NISO//DTD dtbook 2005-2+mathml//EN" "dtbook-2005-2.dtd" ...> Agree -- should use http reference to dtbook. There is also a problem with Appendix 3.3's use. |
| Appendix 3.2 | Swedish Library of Talking Books and Braille | DTBook example: On the root element, the xml:lang attribute is declared twice. |
Remove the extra declaration. |
| Appendix 3.2 | Swedish Library of Talking Books and Braille | DTBook example: The dtb:uid meta data item is missing its 'dtb' prefix. |
Fixed. |
| Appendix 3.4 | Swedish Library of Talking Books and Braille | NCX example: The dtb:uid value is different from the one in the OPF, DTBook and SMIL examples. |
Fixed. |
| Appendix 3.4 | Swedish Library of Talking Books and Braille | NCX example: Shouldn't the dtb:depth meta data have value="1"? |
Fixed. |
| Appendix 3.4 | Swedish Library of Talking Books and Braille | NCX example: The navList with id="pagenum" should be a pageList. |
Fixed. |
| Appendix 3.4 | Swedish Library of Talking Books and Braille | NCX example: The playOrder attributes do not follow the normal playback sequence of the book (2005 spec section 8.4.3) |
Fixed. |
| Appendix 3.6 | Swedish Library of Talking Books and Braille | XSLT file: Why does the doctype declaration in the output document declare the DTBook+MathML public ID? The output is plain DTBook. |
Fixed to point to normal DAISY DTD. |
| Appendix 3.6 | Swedish Library of Talking Books and Braille | XSLT file: The output doctype declaration could be created using the doctype-public and doctype-system attributes on the xsl:output element. |
Fixed. |
| Appendix 3.6 | Swedish Library of Talking Books and Braille | XSLT file: Shouldn't the smilref attribute on the m:math be namespaced? |
I think he is right, but this is really out of my technical
depth. Dennis to test to see if dtbook: namespace
needs to be on selector. |
| Appendix 3.6 | Design Science | Should break the first line into several lines. | Fixed. |