Final Report

OverDrive Adobe “Read Aloud” Text-to-Speech Study

Phase 1 (January through June 2004)

September 2004

Prepared for the

Mid-Illinois Talking Book Center (http://www.mitbc.org/)

and

OverDrive, Inc. (http://www.overdrive.com/)

Tom Peters of TAP Information Services (http://www.tapinformation.com/)

Executive Summary

The purpose of the OverDrive Adobe “Read Aloud” Text-to-Speech Project was to gain real world experience and feedback from blind and visually impaired individuals concerning the accessibility and general usability of DRM-protected Adobe ebooks distributed by OverDrive. In particular, the project intended to ensure that the entire contents of actual ebooks in Adobe format were “read aloud verified” (RAV). Over 60 blind and visually impaired individuals participated in the study, which ran from January through June 2004.

The project was conceived, supported, and coordinated by the Mid-Illinois Talking Book Center (http://www.mitbc.org/), a sub-regional library serving the blind and physically handicapped in central Illinois. The Center is part of a state network administered by the Illinois State Library, a division of the Office of the Illinois Secretary of State, and a national network under the administration of the National Library Service for the Blind and Physically Handicapped (NLS), a division of the Library of Congress. The Center is locally administered by the Alliance Library System (ALS). Various staff team members from OverDrive (http://www.overdrive.com/) and Tom Peters from TAP Information Services (http://www.tapinformation.com/) also served on the project team.

The collective experiences of the volunteers revealed several challenges. Many blind and visually impaired users experienced problems downloading and activating the Adobe Reader software. Most of the volunteers in this project use screen reader software programs such as JAWS for Windows and WindowEyes. Accessing Adobe ebooks on a PC running a screen reader program presented some initial challenges Nearly all of the volunteers preferred using their screen reader software to access and read aloud Adobe ebooks, rather than using the embedded text-to-speech engine within the Adobe Reader software. Many of the volunteers asked for better documentation and keyboard maps. To that end, Richard (Louie) Most constructed a tip sheet for screen reader software users who want to access and enjoy Adobe ebooks. The volunteers also experienced difficulties returning a book early, prior to the expiration of the circulation period.

The blind and visually impaired volunteers are glad that a group of interested persons are seriously addressing the issue of how to make ebooks more accessible for the print-impaired. They emphasized that everyone involved in the ebook and digital audiobook movements needs to recognize that digital rights management (DRM) systems affect accessibility. Often security measures can thwart the use of screen reading software by authorized customers and users. The distinction between accessible and inaccessible ebook needs to be made more self-evident to the potential purchaser or user before the point of sale or use. The PDF format remains a frustrating format for many print-impaired readers, in part due to lack of available information about whether or not a specific file is or is not accessible using Adobe’s embedded TTS engine, third party text-to-speech software, or screen reader software.

Several conclusions and recommendations emerged during the course of the study. Although the quality of the TTS system is important to blind and visually impaired users, the accessibility and ease of use of the entire software program and the quality and variety of content are at least as important. More documentation and training are needed. Several suggestions to improve the process of using and the accessibility of OverDrive’s Content Reserve website were made.

Several volunteers commented that, once they jumped over all the hurdles and mastered the nuances of using screen reader software to interact with Adobe ebooks, they generally were pleased with the experience, especially the access to so much content. For blind and visually impaired individuals, as with sighted individuals, easy access to the content one needs or wants when one needs or wants it is the ultimate payoff.

Purpose of the Study

The purpose of the OverDrive Adobe Text-to-Speech Project was to gain real world experience and feedback from blind and visually impaired individuals concerning the accessibility and general usability of DRM-protected Adobe ebooks distributed by OverDrive. In particular, the project intended to ensure that the entire contents of actual ebooks in Adobe format were “read aloud verified” (RAV). Over 60 blind and visually impaired individuals participated in the study, which ran from January through June 2004. The project was conceived, supported, and coordinated by the Mid-Illinois Talking Book Center (http://www.mitbc.org/). Various staff team members from OverDrive (http://www.overdrive.com/) and Tom Peters from TAP Information Services (http://www.tapinformation.com/) also served on the project team.

One goal of the study was to raise awareness both among the vendors (Adobe, OverDrive, and others) and among blind and visually impaired individuals about the potential and challenges associated with ebooks. There has been longstanding interested among the blind and visually impaired population in improving the accessibility of PDF files and Adobe ebooks. As one volunteer commented, “...accessibility to PDF files has been a real concern of mine since many of my web viewers are blind and vision impaired. That is one reason I feel so strongly that this project is so meaningful. As great as I feel Adobe is, they must somehow recognize that a segment of their targeted readers, the blind and visually impaired, are being left out. I'm convinced they do care or they wouldn't be involved in this project.”

About the Volunteers

Over 60 volunteers participated in this six-month study. They self-selected in response to an invitation to participate that was sent to various email discussion groups. Although the project team did not request demographic information from the volunteers, nearly all of them were blind or had visual impairments that required the use of screen reader software, such as JAWS for Windows or WindowEyes. The volunteers came from a variety of states. Their levels of computer expertise ranged from expert to novice.

Training, Orientation, and Documentation

The Project Team intentionally decided to provide minimal training and orientation to the volunteers in this project, in part because we wanted to determine how easy or difficult it was to access DRM-protected Adobe content from scratch.

After several weeks, many of the volunteers asked for better documentation and keyboard maps. To that end, Richard (Louie) Most constructed a tip sheet for screen reader software users who want to access and enjoy Adobe ebooks.

Collection Development

For the initial collection the project team selected approximately 45 titles that would appeal to popular tastes. After approximately 50 volunteers had begun using the collection we started to receive some mild complaints about the lack of selections and variety within the collection, so in January the project team purchased an additional 37 titles.

Circulation Procedures

Each volunteer was assigned an OverDrive “library card number” and PIN that enabled him or her to find and check out Adobe ebooks on their own. Volunteers were limited to a maximum of three checked out items at any given time. The circulation period was set at two weeks.

Experiences, Challenges, and Problems

Activating the Adobe Reader Software

Numerous volunteers reported problems activating their Adobe Reader software program. Activation is required in order to receive and access DRM-protected content through OverDrive and other vendors of Adobe ebooks. Activation is a process that registers your copy of Adobe Reader 6.0 to your device, enabling your device to open and use eBooks that have been copyright-protected using Adobe DRM (Digital Rights Management). This process is necessary because many publishers require strong copy protection to prevent unlawful redistribution of their eBooks.

A .NET passport is a service from Microsoft that enables the user to use his or her email address to sign into websites and programs and enables the user to be authenticated automatically to use the service the program or website provides. The procedures involved in obtaining a .NET passport were very confusing to many of the volunteers. The process of obtaining a .NET passport involved identifying a graphic which screen reader software cannot interpret well. This presents a major accessibility issue for blind and visually impaired individuals. One stymied and frustrated volunteer reported that he eventually sent an e-mail to Microsoft using their response form. It took Microsoft two weeks to respond and help him set up his .NET Passport by e-mail.

Near the beginning date of this project Adobe streamlined the procedures in subsequent releases of the software. The new process, called EZ Activation, defused the issue of Microsoft passport activation. The new version of Adobe Reader automatically activated when the user attempted to access DRM-protected content.

Using Screen Reader Software in Conjunction with Adobe Reader

Initially several volunteers were confused about how their screen reader software would interact with an ebook running in Adobe Reader. They wondered if they should have their screen reader software turned off. After trying the embedded TTS engine within Adobe Reader, however, many volunteers reported that they preferred using their screen reader software, with TTS voices familiar to them, to access the content of the Adobe ebooks. As one volunteer noted, “I'm not exactly certain why we need to confine ourselves to the voices supplied by Adobe. The overwhelming majority of folks using this software will also have their own synthesized speech to which they've become accustomed. Unless the ultimate goal is to transfer this digitized book format to a portable player of some sort, there's no real reason to use the supplied voices. Another volunteer wrote, “I suspect that in the real world of blind computer users, few would choose the "read aloud" function as they have more familiar means of voicing their documents. That feature in Acrobat was built-in, I suspect, not for blind users but a user who needed voicing but didn't happen to have screen reading software.

One volunteer succinctly articulated his preferences, “…it seems to me that most folks who would ultimately make use of the adobe e-book format would be those who are already at least somewhat familiar and skilled with their computers and text-to-speech. If that's the case, it isn't going to be quality of synthetic speech which becomes the salient factor regarding whether or not people grab on to this modality. I believe it's going to be other factors...

1. Ease of use

2. Availability of a broad spectrum of titles. In other words...
Choice.

As for synthetic speech, most of us who've been using it have our
preferences, and most will not want to stray from those preferences, unless someone comes out with a stand-alone e-book reader in which our preferences aren't available. If that's the case, I'd be very willing to get myself used to other speech engines in exchange for the portability of an accessible e-book reader I could carry around with me.”

Learning the Functions of Adobe Reader

Several of the volunteers expressed problems learning the basic functions of the Adobe Reader software. Some of them admitted that most of the problems they encountered are part of the normal process of learning any software program from scratch. Other problems, however, seemed to be caused by the lack of keyboard alternatives for clickable actions.

Problems with the Embedded Read Aloud Function Itself

For those volunteers who did try the embedded read aloud feature within Adobe Reader, they did report some problems. For example, the header or footer text and page number tends to be read aloud, which detracts from the flow of the reading of the main body of the text itself. One volunteer articulated this generally felt annoyance, “To have to hear so often the author and title is more than disruptive. I don't really have an adequate word for it! Very distracting, and sometimes so redundant and at the sime time abiguous as to make the context of a passage almost crazy, and make one have to go back and figure out if the number was indeed page number or text or, something else.”

Other volunteers lamented the limited functionality of the embedded TTS engine. As one volunteer put it, “Hence, if one wants to check spelling, closely review a table or text, hear more or less punctuation, correct pronunciations for future readin in an exceptions dictionary, they should be using their screen reader, not the Read Out Loud function. I would view the "read out loud" function much as I do the Narrator program in XP. It is nice to have there if a screen reader isn't available; but neither is a substitute for a first class screen reader such as Window-Eyes or JFW. And as Adobe closely works with GWMICRO and Freedom Scientific on accessability issues, they will not compete no more than Microsoft competes with screen reader manufacturers.”

There were a few features of the embedded TTS engine that the volunteers did like. For example, one volunteer noted, “The one thing I've found where Adobe's read aloud functionality is superior to that of WindowEyes is in the area of continuous reading. Since WindowEyes treats .pdf files like html documents, it uses MSAA. In order to read the full document, you have to load it into the MSAA buffer. This can take forever. The Adobe read aloud function allows for continuous reading, what we know of as "read-to-end," without waiting for the entire doc to load into MSAA mode.”

Returning the Ebooks

Numerous volunteers wanted to return the ebooks they had checked out prior to the end of the circulation period, but they experienced problems completing the task. When a user is in the “my bookshelf” area of Adobe Reader, if he right clicks on the image of the dust jacket of the book he wishes to return, he then clicks on the “return to sender” option. For a sighted individual, the process is easy. Blind and visually impaired individuals, however, experienced major frustration. Evidently there is no keyboard alternative for completing this task, and screen reader software cannot easily locate these thumbnail images. If the user switches from the thumbnail view to the list view in “my bookshelf” evidently the ability to right c

Changing Attitudes About Purchasing Digital Books and Audiobooks

Many of the volunteers reported that they are becoming more interested in purchasing e-books and digital audiobooks. The overall marketplace for ebooks and digital audiobooks is becoming quite diversified, and blind and visually impaired users are realizing that they have many options now. They are accepting the fact that in certain situations there are advantages to paying out of pocket for digital content, software, and hardware. Many of these individuals are willing to try both mainstream services (such as Adobe ebooks) and services designed specifically for this sub-population (such as Bookshare.org).

One volunteer summarized the evolving situation and attitudes well, “While we as blind readers have traditionally looked for free sources for our reading material, the changing technology of book distribution may force some of us to strongly consider direct purchase and paid subscriptions. After all, and we forget this, sighted readers, even if they are library patrons, often buy books so they can read them when they want, to have for their permanent collections, et al. Now, with Audible, BookShare, these Acrobat encrypted Ebooks, and who knows what else, change my mindset towards reading.”

General Challenges

Many of the volunteers were pleased that this project was attempting to create a dialogue between e-content suppliers and the blind and visually impaired population. Through dialogue both sides can share information, ideas, and concerns, as well as raise awareness about the opportunities and challenges.

One volunteer summarized the general challenges well, “My very first impression re this project is that it is great to find people seriously addressing the issue of how digital rights managment affects accessibility, and how to make the distinction between an accessible ebook and an inaccessible ebook in the Adobe Reader 6.0 format (or any ebook format for that matter) more clear. …this issue has proven to be one of the biggest problems in making ebooks accessible. PDF in particular remains a thoroughly frustrating format for many readers, as there often seems to be a lack of information regarding whether the ebook will or will not be accessible with a screen reader until you have actually bought the ebook and tried it, and of course there are not only no refunds if it is not accessible but usually no one willing to discuss the issue with you if you wish to determine the reason. In closing, I think there is a real need for everyone involved in ebook design and distribution to be more aware that security settings on ebooks often can lock out screen readers and that the blind consumer may wish to speak with a knowledgable person about potential issues with accessibility.”

Conclusions and Recommendations

Remember: Software and Content Are as Important as TTS

Several volunteers noted that, while the TTS features and functionality of Adobe Reader may be important to the general population, for blind and visually impaired readers the accessibility and ease of use of the software and the quality and variety of available Adobe ebooks are at least as important.

Improve and Expand Documentation and Training

General Recommendations about Documentation and Training

Many of the volunteers wanted more training and documentation. They valued training and documentation as an integral part of the adoption and diffusion of Adobe ebooks among the blind and visually impaired sub-population. They realize that designing, delivering, and maintaining high quality training and documentation are not easy. They suggested a partnership between Adobe, talking book centers, and peer-to-peer information sharing.

As one volunteer noted, “I fully believe that these problems are surmountable. I also firmly believe that many of the problems new users to the technology face are not a fault of the Ebook as such but giving the new user the necessary handholding and guidance allowing them to become more comfortable and familiar with their screen reader and ebook software. This may mean that, over time, libraries for the blind may need to employ or have access to access technology trainers who can explain and help configure software to fit the books and then instruct users in how to read.”

Redesign the feedback form so that it focuses less on the quality of the TTS within Adobe Reader and more on what Adobe ebooks can do for experienced, print-impaired PC users.

The version of the feedback form that was distributed to the volunteers is contained in the appendix to this report.

Create a general primer on how to find, download, and access ebooks using screen reader software.

Numerous volunteers recommended a general introduction or primer for users of screen reader software. One volunteer commented, “I concur that even a very slight introduction to basic concepts of ebook reading with a screen reader would prove beneficial to the new user. I still run into many blind readers who firmly believe no ebooks at all are accessible, or that one experience with an inaccessible ebook means that all ebooks are inaccessible. Once you have some tech experience and some computer savvy you can distinguish between formats and jargon like secured and unsecured, but until you feel comfortable with the tech it is difficult to catch those clue words.”

How to download and install the software

How to activate the Net Passport

This problem and source of frustration should diminish as people upgrade to the new version with EZ activation.

Some volunteers had difficulty using JAWS to activate their Net Passport. The audio files that provides a spoken representation of the graphic that needs to be identified is intentionally muffled to prevent speech recognition software from automatically obtaining passports.

How to use Adobe Reader with Screen Reader Software

Create a list of suggested Adobe Reader configuration settings for print-impaired users.

Create a tip sheet of suggested Acrobat configuration settings as well as a list of keyboard shortcuts for Adobe Reader functions. These would make the reading process much easier and smoother for blind and visually impaired users.

Create a list of keyboard shortcuts for print-impaired users.

In general, keyboard sequences are much more accessible to blind and visually impaired users than clickable buttons, hotlinks, and dropdown menus.

Create a set of instructions on how to navigate within a book, add bookmarks, check spelling, etc.

Clearly indicate if an ebook is accessible to the print-impaired.

Prior to purchasing an Adobe ebook print-impaired readers need to know if the “read aloud” capabilities have been enabled on that particular title. Evidently, if the read aloud capabilities have not been enabled, not even screen reader software can access the content.

As one volunteer noted, “PDF in particular remains a thoroughly frustrating format for many [print-impaired] readers, as there often seems to be a lack of information regarding whether the ebook will or will not be accessible with a screen reader until you have actually bought the ebook and tried it….”

Focus on Interoperability Between Software and Hardware

Make it easier to listen to and navigate through Adobe ebooks using other software, such as screen reader software.

Make it possible to load and listen to Adobe ebooks on the various portable playback devices on the market designed for the print-impaired (e.g., Victor Reader Vibe, Telex Scholar, Book Port, and BookCourier).

Improve the System of Technical Support

Have someone (who knows Adobe Reader, all the various screen reader software programs, and how print-impaired users typically use a computer) be available to provide technical support.

Encourage more peer-to-peer tech support.

Make Software Enhancements

More and Better Text-to-Speech Voices

Adobe should consider purchasing several good voices from AT&T Natural Voices, Rhetorical, etc. Perhaps Adobe could market this as an enhanced version of Adobe Reader and charge a nominal fee for it.

As one volunteer noted, “Last night I looked at a lot of the read aloud
programs with higher quality voices than Microsoft Mary and Sam. … All of the possible programs I ran across which provide this service such as ReadPlease, TextAloud, Natural Reader, etc. have some free reading but the free software always utilizes Microsoft Sam and Mary. To get the "premium" AT&T voices like Mike and Crystal with these programs, you have to pay $25-$30 for the program. AT&T also licenses their voices to application developers.

I wonder if this takes off if Adobe would consider licensing the higher quality voices for read aloud or if readers will want to purchase programs like Read Aloud and Natural Reader for access to these voices.
Although there are some other voice providers out there such as Rhetorical found by Tom Peters, it seems as if the major player "premium" voices are produced by AT&T.”

Another volunteer expressed the need in this way, “I wish there were a way to get the AT&T Natural Voices for use with these eBooks at a reasonable cost, both in terms of money and CPU usage.”

Change key sequences for basic commands

Some of the “shift-control” options to read to end of page, read to end of document, pause, etc. do not work when the end-user also has screen reader software running (e.g., ZoomText Speech).

Process Enhancements

Checking Out a Book

Returning a Book Prior to the Due Date

Make it easier for print-impaired users to purchase, download, and listen to ebooks.

Improvements to OverDrive’s Content Reserve Website

Use more headings, which are easier for screen reader software programs to navigate.

One volunteer suggested, “…use more headings, so that when using the numberic keypad in Jaws you can utilize the insert plus enter conbination to go from section to section. As it is, the page is not divided up into easy-to-swallow clusters of text, i.e., one section on information, one on browsing ebooks, one for your account status, etc. This made navigating through the entire page time-consuming.”

Use smaller, clearly divided chunks of text.

Communication Enhancements

Hold monthly iVocalize meeting for volunteers.

Conclusion

The functionality of text-to-speech is highly valued by many sub-populations, including the print-impaired (i.e., anyone who for any reason cannot comfortably read printed material), persons learning to read a language (either as a first or subsequent language), commuters, and exercise enthusiasts. Most users currently prefer naturally narrated audiobooks over synthetic TTS DTBs. Unfortunately, the economics and time delays inherent in the production of naturally narrated audiobooks mean that only a small percentage of all new and existing texts will ever have naturally narrated versions. TTS DTBs have some distinct advantages, such as the speed and economical efficiency with which they can be created. This in turn means that a vocalized rendition can be produced on demand of almost any digital text.

The quality and accessibility of TTS features need to be improved so that the inherent advantages of TTS over naturally narrated audiobooks can shine forth. There are at least four development routes that could be pursued.

First, screen reader software could continue to be improved. As the diffusion of screen reader software continues, perhaps the purchase price may decline. Adobe and other producers and distributors of e-content would then concentrate on ensuring that their software and content interact well with at least the leading screen reader software programs. Based on the feedback from the volunteers involved in this study, this development route appears to be preferred by the blind and visually impaired population. The problem with this route, however, is that other sub-populations that could benefit from an outstanding TTS system, such as language learners, commuters, and exercise enthusiasts, do not really need screen reader software.

A second route would be to rely on third party plug-ins. For example, in the case of Adobe Reader, the PDFaloud plug-in would be an option that falls into this category. A plug-in such as PDFaloud appears to be geared more toward the language learning sub-population, not the blind and visually impaired. This route, however, could develop into a logistical nightmare for end-users as they would need to discover and install third-party plug-ins for all the e-text software they use. Also, the relationship between mainstream producers of e-text software and producers of third-party plug-ins is not always amicable and symbiotic. Sometimes the functionality of either the main software program or the third-party plug-in is thwarted by the other.

A third route would be to improve the TTS functionality of mainstream software such as Adobe Reader. This will involve increased commitment and investment by Adobe. To recoup that investment, Adobe may wish to offer the enhanced TTS functionality in some sort of premium edition of Adobe Reader. The basic edition could continue to be distributed free of charge, but a nominal fee could be charged for the premium edition. One limitation of this route is that end-users use several software programs to access e-content. To truly benefit the end-user, each major producer of mainstream e-content software access would need to concomitantly increase their commitment to and investment in quality TTS functionality.

In general, the volunteers who participated in this study tended to see this third route as ultimately unworkable. As one volunteer noted, “I honestly can't imagine, unless Adobe starts using a clearly superior speech engine, that anyone (blind or sighted) is going to use the built-in read aloud function. It just isn't going to happen. Sighted people who want a book read to them are never going to put up with this software synth long enough to get used to, the synthetic speech Adobe is currently supplying. And, blind people who are sophisticated enough to use a system with Adobe on it will already have had at least some moderate experience with a screen reader and speech synth of superior performance and sound quality.”

A fourth route would be to improve the TTS functionality at the operating system level. Evidently, the current “embedded” TTS functionality within Adobe Reader actually draws heavily from the resident operating system. One advantage of this route is that there are a smaller number of mainstream producers of operating systems. With a good TTS system at the operating system level, developers of e-text software could exploit that basic system at the application level. This fourth route would involve increase commitment and investment by software producers.

The blind and visually impaired volunteers in this small study seem to prefer and advocate a version of the first route. They would like to use the major screen reader software programs, such as JAWS for Windows and WindowEyes, to access e-textual content from Adobe and other producers and distributors of e-texts. They encourage Adobe and other software companies to continue to improve the accessibility of their software and content, especially the interface between their systems and screen reader software. While Adobe and other companies should heed the ideas and wishes of the growing blind and visually-impaired sub-population, they also need to be mindful of all the sub-populations that stand to benefit from improved and more accessible TTS functionalities.

Appendix

Quality Assessment Form

Thank you for participating in the first phase of a joint project between the Mid-Illinois Talking Book Center (www.mitbc.org) and OverDrive (www.overdrive.com).

The goal of the first phase is to have the participants in this pilot project test the “read out loud” (text-to-speech or TTS) feature of the e-books in the collection, using the current version of Adobe Reader software, then rate the quality of the text-to-speech content and the overall experience. We ask that, if at all possible, you listen to the entire book using the text-to-speech functionality within the Adobe Reader software.

This is a quality assessment form, not a technical information sheet or a set of detailed instructions on how to download content. For more information about technical details of the Adobe Reader software and its text-to-speech functionality, please email info@mitbc.org

Please answer as many questions as possible. If you do not understand a question or do not know the answer, just skip it. Your comments are very valuable. If you need more space, please use the end of the document.

When you have completed this text-to-speech quality assessment form, please return the form to Tom Peters, preferably as an email attachment. Thank you.

You can also fill this form out online at:

http://ebooks.mitbc.org/Feedback.htm

Tom Peters
1000 SW 23^rd Street
Blue Springs, Missouri 64015

Phone: 816-228-6406
Email: tapinformation@yahoo.com

About the Book:

Author: _____________________________________________________

Title: _______________________________________________________

Publisher: ___________________________________________________

About You:

First Name: __________________________________________________

Last Name: __________________________________________________

Email: ______________________________________________________

About Your “Read Out Loud” Experience:

Did you listen to the entire book?

Yes _____
No _____
If no, why not? ___________________________________________

________________________________________________________

What “voice” did you use for your text-to-speech experience?

Microsoft Sam _______________
LH Michelle _________________
LH Michael __________________
Other (please specify) _____________________________________
Comments about the voice you listened to: _____________________

________________________________________________________

At what speed did you play back the content?

Normal speed ____________________
Somewhat faster than normal __________________
Somewhat slower than normal _________________
Much faster than normal ______________________
Much slower than normal _____________________
Comments about playback speed: ____________________________

________________________________________________________

What pitch did you use for playback (1-10): _______

Comments about the pitch you used: __________________________

________________________________________________________

What type of listening apparatus did you use?

Speakers _____________
Headphones __________
Ear buds _____________
Other (please specify) ____________________________________
Comments about your listening apparatus: _____________________

________________________________________________________

How well did the TTS feature handle the various parts of the book?

Cover