QRead

Blog Date: 
Thursday, March 15, 2012

QRead is a new book reading solution being sold by Chris Toth, the blind programmer who is best known for creating the Qwitter  client for Twitter (no longer in development) and the Hope client for Pandora Radio.  It is intended to be a simple solution for reading a fair number of different file formats, and to provide bookmarking and search features within the text.  As a great lover of books, and a devoted reader of eBooks, I sat down with the program to give it a spin.

According to the documentation QRead is capable of reading PDF, TXT, HTML, EPUB, zipped and unzipped Bookshare titles, Microsoft Office documents from Office 2007 or newer  (.DOCX) and Microsoft Help files (.CHM).    It is intended to automatically bookmark a user’s place as they read the book and remember that place every time the book is closed, even if the book is closed because of a crash of the program or the computer.  Other highly useful features mentioned by the documentation include the ability to search forward or backward within a file, a manual bookmark system, which allows a user to have one temporary bookmark, and many “named” bookmarks, and compatibility with the major screen access software packages for Windows. 

In order to give the program a fair trial I collected a large number of different files to test its usability, and ran its major functions against NVDA, Window Eyes, and JAWS (All current versions).  My first file was a PDF document containing links, headings, a captioned picture or two, a table, and a large amount of text.  This is a very usable document in the Adobe Reader software with JAWS, so I expected that it would work well in QRead as well.  When I opened the file in QRead, I discovered that the program essentially stripped all non-textual elements from the file, and presented nothing but the straight text to the user.  In the case of this file, this actually caused a number of different problems.  First, all traditional navigational elements such as links and headings were stripped from the file, which left me forced to use the find dialog to move to the beginning of the desired section of text, instead of the links and heading structure which were available in Acrobat Reader.  Furthermore, QRead broke the structure of the file’s embedded table, so it was not clear in QRead, though it read quite well in Acrobat Reader.  Finally, when I compared the two readers, I found that QRead left out the image descriptions which were found in the original text, so if a user has a well-formed PDF, it may be in their best interest to continue using Adobe’s Reader to consume it. One advantage that I found for QRead which must be mentioned is that it allowed for continuous reading of the document, which in larger files with Adobe is not advisable (or sometimes even possible) since all but the most powerful computers seem to choke in large PDF files when using speech.

The next file I attempted was a much less well-formed PDF.  The text was available to JAWS, but there was no navigation embedded in the document, pictures were not mentioned, and the file was large and unwieldy.  When this file was loaded into QRead, it came up quickly, and it was easier to traverse the document than in Acrobat, because of the bookmark, find, and percentage slider features, and the smaller impact on computer resources.  So, for PDF, the program is a bit of a toss up.  Whether or not QRead is the better option is going to depend on what you want to read, and how you want to read.

EPUB, .docx, DAISY, and HTML files are presented as plain text, just like the PDFs are.  Specifically in the case of the EPUB, DAISY and HTML documents, this presentation strips useful features from the files, and acts as a real detriment to the reading process, though it does once again offer bookmarking, so when reading these types of files it is important to determine whether the ability to easily bookmark the user’s place is more important than to have access to navigational links and headings.  For me, the answer is clear, I want the navigation, but each person has a different reading style, and different needs.

Text files work just as the user would expect.  Since all files are read-only, this is a good solution if a user wants to read and copy bits of text from any of the aforementioned formats.  The bookmarks do give a person a fair bit of power in creating their own navigational structure, and can even be used by a creative soul as a basic system of highlighting important information within a book.  The percentage slider is a useful tool for moving quickly through a long document, or for finding out just how much longer that bit of assigned reading really is.  Finally, the find command works well enough for traditional searches through the text, though some problems were encountered with this feature, which I will outline below.

I want to preface these remarks with the statement that I do understand that this is fairly new software which was programmed by one person, and it has already been updated once since its initial release.  I expect that these problems will be rectified in time, but I discovered several problems with this software in its current form.  While running QRead on a Windows XP SP3 system, I found that the program was prone to some random crashes, particularly when opening files.  If other files were open at the time of a crash, there was about a 50/50 chance that the files would re-open upon launching the program, and whether or not those files would still have their automatic bookmark in place.  Another problem that I found was that although there was an option for using regular expressions (a special set of syntax that allows for complex searches), my tests would cause the search box to not respond when I attempted to use them.   In the same way, when searching backwards in a file, the search box would sometimes hang, and never give any indication of either finding or not finding a result.  Looking specifically at the search box for a moment, it is important to note that this program is actually more accessible to screen access software users than it is to anyone using the screen visually (with or without magnification) because it is not possible to see or manipulate the buttons in that box without speech.  I attempted to open a few files with the program that were not explicitly supported, such as an .rtf file, a .doc file, a zip file containing PDFs, an adobe protected EPUB (.acsm extension) and an EPUB 3 not because I expected them to work, but to simply see how the program handled errors and unsupported document types.  It handled the .doc, .rtf, and the protected EPUB by not offering them in the file list in the open dialog box, which made a lot of sense.  If the program cannot open the file, it’s just as well it not make it available by default.  More interestingly, the EPUB 3 and the .zip file simply dropped me back in the main program interface without warning me that the file was not supported, or that it couldn’t be opened.  It would be nice to have the program mention limitations like this when it cannot handle a document which is passed to it.  Unfortunately, there were two document types which were supported according to the documentation which would also not open on my test machine, the Microsoft help files (.chm), and the zipped Bookshare DAISY files, though as of version 1.11 of the software the uncompressed files now open just as they should.  Most of my testing was performed on version 1.0 of the software, and at that time, all Bookshare books were refusing to open, so as you can see, there has already been improvement in the program, and I only expect that to continue.

The bugs and limitations of the software do mean that it will not have as broad an appeal as it might, but QRead is becoming an interesting option for reading some books in certain file formats.  It is my hope that there will be added support for RTF files in the future, as well as the older .doc format (This is a planned feature according to the website.)  Whether or not this support is added, it will be a useful tool for reading some common file types, though it won’t be the only, or even preferred, tool for reading many of them.