The Reluctant Communist

We've been making books and we would like to hear how accessible they are from your perspective, and what we can do differently to make them better.

The Reluctant Communist

Post by
farrah
»
Wed Feb 07, 2018 3:59 pm
Download the e-text, EPUB 3 (media overlays) and DAISY files for The Reluctant Communist:
https://nnels.ca/items/reluctant-commun ... orth-korea

We would appreciate your feedback about what works and what doesn't work.

Here are some questions we have:
  • Can you navigate through all of the headings? Does the structure make sense?
  • What do you think about the quality of the image descriptions? How can we make them better?
  • Are there any words/phrases/sentences in the book that do not translate correctly using synthetic speech?
  • Are there any other accessibility issues that you notice in each of the different formats?
  • Which format do you prefer to read (e-text, EPUB, or DAISY) and why?
farrah
Posts: 42
Joined: Tue Feb 06, 2018 10:50 am
Contact:

Re: The Reluctant Communist

Post by
ka.li
»
Thu Feb 08, 2018 4:59 pm
It was a fascinating story so I finished reading the entire book today. I read this using the EPub file with Voice Dream reader.

The heading structure makes sense. I'm not sure if it is a Voice Dream Reader issue or an issue with the file but I couldn't get Voice Dream to read the text and the alt-text in the photograph section. It would skip over it entirely. It would just say "Photo section" before moving on to the next chapter.

The photo descriptions were good. Even though there's text for Figures 12 and 13, I don't see any alt-text or even a graphic element for those figures. Are they missing?

The only weird pronounciation I noticed was when "6'4?," was read out. The voice I used to read the book said "double prime."
I tested this passage out with some other voices/synthesizers and they handled that phrase just fine so I wouldn't change it and let it be handled by the different synthesizers. I'm not sure if Alex the media overlay or with the DAISY file says that phrase properly.

The only other issue that I've noticed is that the table of contents is read at the end.

I prefer EPub because I can read the book with my app and voice of choice. Also, if I wanted to, I could read the book with my braille display.
ka.li
Posts: 72
Joined: Tue Feb 06, 2018 2:36 pm
Contact:

Re: The Reluctant Communist

Post by
Sabina
»
Fri Feb 09, 2018 4:16 am
That "double prime" problem sounds brief but confusing. Is it worth finding the error so that the sentence has meaning?

Would you be willing to test the photo section with another reader?

Table of Contents at the end sounds impractical.

Thank you for noticing missing image descriptions.

I wonder what the Ace Accessibility Checker would have to say about this book.

Hearing a reader say a book is fascinating is music to a librarian's ears, and Farrah, especially, will be very happy to read this news.
Sabina Iseli-Otto at NNELS
1-855-383-5761, ext. 1031
sabina@nnels.ca
Sabina
Posts: 35
Joined: Tue Jan 30, 2018 3:41 pm
Contact:

Re: The Reluctant Communist

Post by
ka.li
»
Fri Feb 09, 2018 10:34 am
The "double prime" thing is a quirk of the synthesizer I'm using but many other synthesizers will interpret the punctuation correctly. If it were changed, it might confuse a braille reader so I don't think that phrase needs to be changed.

I looked at the image section with Microsoft Edge and sometimes I found that I would loose focus in that section. I'm not sure if it's Edge, NVDA or the file. I can take a look at that section again in a few other EPub readers.

The ACE checker found 6 issues although it looks like most of them are minor issues such as missing the HTML Language element. The only major issue it found is missing Accessibility Metadata. A couple of the issues it found i believe are related to that.
ka.li
Posts: 72
Joined: Tue Feb 06, 2018 2:36 pm
Contact:

Re: The Reluctant Communist

Post by
farrah
»
Fri Feb 09, 2018 11:30 am
Thanks for checking out that book (and glad you enjoyed the read! I want to read that one too).

That's really interesting that Voice Dream throws out the photos and image descriptions. I think this is probably intentional on Voice Dream's part - I'll make a note to ask them about this. I'd like to know how it reads using other software.

As for the ACE checker, yes, we don't include accessibility metadata in our files yet, but we will be adding that metadata in going forward. Interesting about the missing HTML language element, I wonder if this is not being converted from DAISY correctly as our DTBook XML files always identify the languages in the file <head>. For example:

Code: Select all

<meta name="dc:Language" content="en" />
I'd be curious to unzip the EPUB and check the .opf file and html files to see what's missing, and find out what we need to add into the EPUB code to make the ACE checker validate that.

ka.li wrote: Fri Feb 09, 2018 10:34 am The "double prime" thing is a quirk of the synthesizer I'm using but many other synthesizers will interpret the punctuation correctly. If it were changed, it might confuse a braille reader so I don't think that phrase needs to be changed.

I looked at the image section with Microsoft Edge and sometimes I found that I would loose focus in that section. I'm not sure if it's Edge, NVDA or the file. I can take a look at that section again in a few other EPub readers.

The ACE checker found 6 issues although it looks like most of them are minor issues such as missing the HTML Language element. The only major issue it found is missing Accessibility Metadata. A couple of the issues it found i believe are related to that.
Content Coordinator, NNELS
farrah@nnels.ca
Toll Free 1-855-383-5761 ext 1033
farrah
Posts: 42
Joined: Tue Feb 06, 2018 10:50 am
Contact:

Re: The Reluctant Communist

Post by
Daniella.LP
»
Tue Feb 13, 2018 3:25 pm
I first began reading and exploring the E-text version using Word, before moving to the DAISY version, which I read with Read2Go in my iPhone. I could not download the .epub version, I got the same error several times ("This XML file does not appear to have any style information associated with it. The document tree is shown below."

Photo descriptions

The descriptions in the DAISY format using the Read2Go app are read well and in their entirety.

In the E-text version, it took me some exploring to make JAWS read the descriptions properly:

In the Photo Section (level 1 heading), if I start reading continuously, JAWS read the descriptions but if I stopped it, it did not continue reading them; I could not find the descriptions moving line by line. I tried to find the descriptions by looking for specific words but could not move to them. I assume that the descriptions are inserted as objects.

I could make JAWS read the descriptions by moving up and down using the read previous or next paragraph command. Nevertheless, at first JAWS was not reading the entire description. It appeared to skip part of the descriptions of the photographs; it did not finish reading them, even when I moved to them using the previous or next paragraph commands.

A couple of examples of JAWS not reading the end of the descriptions:

Description for figure 1: "... uniform. He has a st..."

The description for figure 3 stops with "pin depicting the torso o..."

Description for figure 4: "... floral dress that reaches just pass..."

Would it be possible to insert photo descriptions in the text, rather than as objects?

Using JAWS in Word, I can move to the graphics (the description objects) as follows:
Activate quick keys by pressing insert+z
Then type the letter "g". Shift+g will find the last graphic).

Note: JAWS detects text layer objects if the Graphics option under General Options is on in the Quick Settings dialog. Press insert+v, then type graph, scroll down to find "include graphical dimensions" and activate it. This fixes the issue of JAWS reading incomplete descriptions (in most instances). It is still necessary to move to the descriptions using the previous or next paragraphs, and the letter g when quick keys is on in Word.

It may also be useful to include some instructions in the production notes section at the beginning of the book (the E-text version) for readers to know that there are photo descriptions. It may also be useful to add some instructions on how to find/read descriptions; perhaps add this information with other tips on the website.

I saved the .docx file as a pdf to see how the figures and descriptions were read by JAWS, but I could not find any descriptions searching for text (I looked for specific words like floral and dress). Also, it is interesting that after converting the Word file to PDF, there appear to be only three headings: about this digital talking book; Additional acknowledgement; copyright and publishing information.

Regarding format preferences, I usually prefer E-text versions if they are available; the exception is if the DAISY format is narrated by a human rather than synthetic speach. I prefer E-text for two main reasons:

1. They are much smaller than DAISY formats and are quicker to download.

2. It is easier to search for text. If I want to read E-text versions away from my computer, I save them into a different format rather than .docx (I find that .docx files do not work seemlessly with the Victor Stream.
Daniella.LP
Posts: 77
Joined: Sat Feb 03, 2018 6:20 am
Contact:

Re: The Reluctant Communist

Post by
farrah
»
Wed Feb 14, 2018 9:23 am
Thank you Daniella. This is really useful feedback. I'm going to take a closer look at the e-text version this week to see why the photo descriptions are not reading with JAWS. I'll let you know what I find.

Also, about the EPUB file - could you test if it works in other software? Read2Go only supports DAISY 2.02 and DAISY 3.

Regarding your preferred format, I wonder if it would be better if NNELS were to produce text-only DAISY books by default? Right now we produce full-text, full-audio DAISY books only (synthetic speech). This would result in a much smaller file size, and it seems like most (if not all?) users have devices or software with built-in text-to-speech anyways. I wonder what others think?



Daniellalpl wrote: Tue Feb 13, 2018 3:25 pm I first began reading and exploring the E-text version using Word, before moving to the DAISY version, which I read with Read2Go in my iPhone. I could not download the .epub version, I got the same error several times ("This XML file does not appear to have any style information associated with it. The document tree is shown below."

Photo descriptions

The descriptions in the DAISY format using the Read2Go app are read well and in their entirety.

In the E-text version, it took me some exploring to make JAWS read the descriptions properly:

In the Photo Section (level 1 heading), if I start reading continuously, JAWS read the descriptions but if I stopped it, it did not continue reading them; I could not find the descriptions moving line by line. I tried to find the descriptions by looking for specific words but could not move to them. I assume that the descriptions are inserted as objects.

I could make JAWS read the descriptions by moving up and down using the read previous or next paragraph command. Nevertheless, at first JAWS was not reading the entire description. It appeared to skip part of the descriptions of the photographs; it did not finish reading them, even when I moved to them using the previous or next paragraph commands.

A couple of examples of JAWS not reading the end of the descriptions:

Description for figure 1: "... uniform. He has a st..."

The description for figure 3 stops with "pin depicting the torso o..."

Description for figure 4: "... floral dress that reaches just pass..."

Would it be possible to insert photo descriptions in the text, rather than as objects?

Using JAWS in Word, I can move to the graphics (the description objects) as follows:
Activate quick keys by pressing insert+z
Then type the letter "g". Shift+g will find the last graphic).

Note: JAWS detects text layer objects if the Graphics option under General Options is on in the Quick Settings dialog. Press insert+v, then type graph, scroll down to find "include graphical dimensions" and activate it. This fixes the issue of JAWS reading incomplete descriptions (in most instances). It is still necessary to move to the descriptions using the previous or next paragraphs, and the letter g when quick keys is on in Word.

It may also be useful to include some instructions in the production notes section at the beginning of the book (the E-text version) for readers to know that there are photo descriptions. It may also be useful to add some instructions on how to find/read descriptions; perhaps add this information with other tips on the website.

I saved the .docx file as a pdf to see how the figures and descriptions were read by JAWS, but I could not find any descriptions searching for text (I looked for specific words like floral and dress). Also, it is interesting that after converting the Word file to PDF, there appear to be only three headings: about this digital talking book; Additional acknowledgement; copyright and publishing information.

Regarding format preferences, I usually prefer E-text versions if they are available; the exception is if the DAISY format is narrated by a human rather than synthetic speach. I prefer E-text for two main reasons:

1. They are much smaller than DAISY formats and are quicker to download.

2. It is easier to search for text. If I want to read E-text versions away from my computer, I save them into a different format rather than .docx (I find that .docx files do not work seemlessly with the Victor Stream.
farrah
Posts: 42
Joined: Tue Feb 06, 2018 10:50 am
Contact:

Re: The Reluctant Communist

Post by
rmarion
»
Wed Feb 14, 2018 4:48 pm
I think the idea of producing daisy text only could work for the most part. Since most of the books NNELS produces are synthesized speech anyhow, it would not be much different for those who listen to synthasized speech on a regular bases. However, I personally prefer human read books when I am listening for pleasure for the most part.
rmarion
Posts: 66
Joined: Fri Feb 02, 2018 1:08 pm
Contact:

Re: The Reluctant Communist

Post by
farrah
»
Thu Feb 15, 2018 12:58 pm
Thanks Richard. That's the direction we're leaning in, DAISY text-only. Synchronized human narration is definitely ideal as synthetic speech still has some way to go. We work with our volunteer narrators to produce MP3 books upon request. We definitely need more volunteers though - the demand is high for narrated audiobooks!

rmarion wrote: Wed Feb 14, 2018 4:48 pm I think the idea of producing daisy text only could work for the most part. Since most of the books NNELS produces are synthesized speech anyhow, it would not be much different for those who listen to synthasized speech on a regular bases. However, I personally prefer human read books when I am listening for pleasure for the most part.
farrah
Posts: 42
Joined: Tue Feb 06, 2018 10:50 am
Contact:

Re: The Reluctant Communist

Post by
farrah
»
Thu Feb 15, 2018 3:00 pm
@ka.li
I checked with Voice Dream and they automatically throw away images and the associated alt text descriptions; however the captions are retained.
As for Figure 12 and 13, both images have a caption and alt text, as do all the images in this book.
The Table of Contents at the end is a result of our DAISY to EPUB conversion process, we use DAISY Pipeline and, for some reason, they decided to put the inline table of contents at the end of every EPUB. As this is an open-source tool, there is probably a way to fix this if you're a good programmer and confident in altering the DAISY scripts... I would be happy to point you to the script if you're feeling adventurous.

ka.li wrote: Thu Feb 08, 2018 4:59 pm It was a fascinating story so I finished reading the entire book today. I read this using the EPub file with Voice Dream reader.

The heading structure makes sense. I'm not sure if it is a Voice Dream Reader issue or an issue with the file but I couldn't get Voice Dream to read the text and the alt-text in the photograph section. It would skip over it entirely. It would just say "Photo section" before moving on to the next chapter.

The photo descriptions were good. Even though there's text for Figures 12 and 13, I don't see any alt-text or even a graphic element for those figures. Are they missing?

The only weird pronounciation I noticed was when "6'4?," was read out. The voice I used to read the book said "double prime."
I tested this passage out with some other voices/synthesizers and they handled that phrase just fine so I wouldn't change it and let it be handled by the different synthesizers. I'm not sure if Alex the media overlay or with the DAISY file says that phrase properly.

The only other issue that I've noticed is that the table of contents is read at the end.

I prefer EPub because I can read the book with my app and voice of choice. Also, if I wanted to, I could read the book with my braille display.
farrah
Posts: 42
Joined: Tue Feb 06, 2018 10:50 am
Contact:

Post Reply

Who is online

Users browsing this forum: No registered users and 15 guests