But the rules of program distribution are so ridiculous that I can’t imagine playing by them. I want to give my apps away. And I want to do it without some faceless technician’s approval.

After you’ve done the work to develop your App they can reject it without giving any reason.

This recent post from Riverturn illustrates the problem though I’ve heard of many more cases like this.

Here is the area of setup.py that I modified:

LIBS = ['ibmeci']

HOME = os.environ['HOME']

os.environ['LD_RUN_PATH'] = os.path.join(HOME, 'opt/IBM/ibmtts/lib')

if sys.platform.startswith('linux'):
  INC_DIRS = [os.path.join('../'),
              os.path.join(HOME, 'opt/IBM/ibmtts', 'inc'),
              os.path.join(HOME, 'opt/IBM/ibmtts-devel', 'inc')]
  LIB_DIRS = [os.path.join('../'),
              os.path.join(HOME, 'opt/IBM/ibmtts', 'lib'),
              os.path.join(HOME, 'opt/IBM/ibmtts-devel', 'lib')]
elif sys.platform == 'win32':

I also had to link eci.ini into the folder with my python code in order to properly find it. I bet updateECI is supposed to fix that but I couldn’t figure out how to run it.

Outfox is a Firefox extension that allows in-page JavaScript to access local platform services and devices such as text-to-speech synthesis, sound playback and game controllers. You can read all about it at its Google Code page and discuss it in its Google Group.

Motivation

In our continuing pursuit of the best platform for delivering accessible games and tools to people with disabilities we are now focusing on Firefox to see what we can accomplish from inside the browser. Up until recently, we’ve used Python. Its a great language and we’ve delivered thousands of copies of our apps to people all over the world but I kept thinking we could do better. For example, the ability to author games in Hark the Sound has enabled teachers to do things we never imagined but sharing games is so complicated that is almost never happens. In contrast, the exciting response we’ve seen to Tar Heel Reader with 419 books in 3 languages after just 10 weeks online has convinced me that browser-based apps can have a big place in our work.

Firefox is a nearly perfect platform; it runs everywhere we want to be and it is easy to program but it has a few deficiencies that limit what we can do. Many of our users are blind or visually impaired. For these kids, speech and sound are a must. Speech and sound are important for kids who can see as well. For example, many beginning readers benefit from having text read to them. Unfortunately, sound in the browser gets little respect.

We used SoundManager2 with a web-based text-to-speech server in Tar Heel Reader. It works well enough but the delays associated with fetching speech as mp3s over the web and the requirement to be online to get speech look bad in comparison locally hosted program.

Outfox

Outfox provides a simple programming interface to JavaScript running in the browser so it can access services on the computer that aren’t normally available. In this first release, we enable access to speech and multi-channel sound with callbacks into JavaScript at the beginning and ends of words and sounds. Soon we expect to extend it to include joysticks, DDR pads, and such. With access to these capabilities I believe we can implement browser-based versions of our most popular games such as Hark the Sound, Sonic Zoom, and Descent into Madness. If things work as I hope, teachers will have access to hundreds of curriculum-related games and a simple way to write and share their own. I hope we’ll get cross-over between the amazing community behind Tar Heel Reader and the community of teachers of the visually impaired and orientation and mobility teachers who use our games for blind children. When we start seeing books about games and games about books, we’ll know that synergy is taking hold.

I’m excited about offering Firefox with speech and sound to my students to see what amazing things they’ll think of doing. How about self-voicing web pages and maps? How about books with interactive, switch-accessible games embedded?

This initial version is likely buggy on at least one of the three operating systems we’re supporting. Please give it a try. Or better yet, download the source, and help us make it better.

Also, help us get it out of the sandbox at Mozilla Addons by trying and commenting on it.

So I hacked on CamKeys to add a GUI for configuring all sorts of events. I’ll include a few screen shots below. Now the teachers are recording their own speech and music sounds and hooking them to events on the DDR pad or camera.

The CamKeys roots of this project were funded by the Mozilla foundation.

The source is in CVS at sourceforge uncassist.

These pages are from a typical book at the site.

My friends Karen Erickson and Gretchen Hanser of the Center for Literacy and Disability Studies tell me that children with limited ability to speak or handle traditional books have few opportunities to learn to read and write. While typically developing children have access to dozens or even hundreds of books, severely disabled kids often have access to very few. Imagine being fifteen years old, unable to speak or to reliably use your hands to manipulate a book and getting your first opportunity to read. Books about baby farm animals aren’t going to cut it. You might be interested NASCAR, fashion, or spacecraft but beginner books on those subjects are rare and even more rarely accessible. Could we do something to make creating accessible books easier?

At the Tar Heel Reader site you can select a book and then advance through the pages with just about any of the characters commonly generated by switch interfaces (let me know if you need something not currently supported). You can have each page read out loud by a computer-generated voice if you wish. Also, each book is automatically available for download as a PowerPoint, Impress, or Flash show.

If you don’t find a book to suit your reader’s interest, making a new book is very easy. The book creator wizard guides you through the process of finding pictures on Flickr, and providing a simple sentence to go with each. Then, with the click of a button, you can create a new book that will immediately be available for reading or download.

How it works

I started hacking PHP using the Flickr API and quickly had a primitive book creator wizard working. When I started thinking about all the supporting infrastructure I would need to provide logins, searches, and multiple authors, it hit me that WordPress already had everything I’d need. A blog post by Steve Winton encouraged me to examine the possibilities further. I’m very impressed with the ease of integrating the wizard and switch-accessible reading into WordPress by only hacking on the theme.

I started with the default theme and hacked the css a bit to create more space on the page. Then I created an empty page for the wizard and gave it a custom template containing my PHP code. Working within the page framework in this way gave me access to all the great WordPress functions while leaving me free to program the page however I wished. In the screen shot to the left you can see the wizard in use. The two rows of pictures at the top are the result of a Flickr search for the tag squirrels. The two boxes near the bottom are the pages constructed so far with an area in each for the short sentence for that page. I used the cool Hoverbox Image Gallery css to make the Flickr search images zoom when you move your cursor over them.

The wizard would be much cooler if it used AJAX but the current version is simply a single html form that does a POST to the page URL. At the very top of the code I call the WordPress auth_redirect function to make sure the author is logged in. Then I check to see if the request method is GET or POST. A GET request signals starting to write; all other calls will use method POST. I used the phpFlickr class to manage access to Flickr. When the author has completed the book, pressing the Create my book button uses the WordPress wp_insert_post function to create post representing the book and then redirects the browser to the newly created post.

Each book is simply a multi-page post using the WordPress <!–nextpage–> tag. They have a regular structure that allows me to parse them with a regular expression when invoking the wizard for editing a book. I customized the book display by modifying the single.php file to make a singlebook.php file that is used for posts in the Book category. The singlebook template eliminates the sidebar and other navigation to unclutter the layout. Switch enabling the reading is done with a bit of JavaScript that advances to the next page on a KeyPress event. Speech enabling the books is implemented with a Python CGI script that uses pyibmtts to convert the caption to a Flash movie that is embedded in the page.

Teachers sometimes use slide show tools such as PowerPoint to make accessible books for their disabled students. To support this and to allow reading books offline each book may be downloaded in Microsoft PowerPoint, OpenOffice Impress, and Adobe Flash format. I generate the Impress format XML using the PHP DOM functions and then convert to PowerPoint and/or Flash using soffice running as a server with the PyODConverter as described in another post.

Future directions

The book on Jeff Gordon illustrates a possible future direction for these books. I used the built-in WordPress editor to modify the last page to include a YouTube video instead of a static picture. I think we could extend the wizard to include YouTube searches if there is enough interest.

Karen suggested including sentence frames to encourage authors to create easy to read texts. I didn’t understand her well enough to try to implement something. Hey, I’m just a computer guy!

Gretchen would like a way to make whatever book the class is currently reading available in this format. I think we could do something with a web cam and a bit of uploading assistance.

Developing this prototype was both fun and educational for me. If we get enough interest from teachers perhaps I’ll get a group of students to program an enhanced version this fall.

Then I needed to convert the Impress format to PowerPoint. Thankfully, I found PyODConverter to automate that part of the job. It worked fine when I ran it from the command prompt but I could not get it to find the server when I ran it from PHP. Again the web came to my rescue with a post by Piero. I’ll reproduce the code here so I don’t forget it.

cd /var/www                                      #www-data user home
sudo mkdir .openoffice.org2                      #create the openoffice working dir
sudo chown -R www-data:www-data .openoffice.org2 #set the owner
sudo chmod 700 .openoffice.org2                  #change permissions
sudo -s                                          #you need to be root
su - www-data                                    #if you want to be www-data

#start openoffice headless
xvfb-run -a soffice -accept="socket,host=localhost,port=8100;urp;StarOffice.ServiceManager" -nologo -headless -invisible

Now I need to figure out how to get this into /etc/init.d so that it will automatically start on reboot.

I’m pleased to report that it is really easy to integrate a PHP web-app into Wordpress. I use ordinary blog posts as the books with the <!–nextpage–> tag for pagination and a bit of Javascript for switch accessibility and text to speech. The book creator wizard is simply a set of pages with custom templates. You select some images, provide some captions and a title, and it creates the blog post that is your book. Wordpress provides logins, all the database manipulation, comments, tags, categories, etc. Redirects work fine because Wordpress is silent until you invoke get_header. Query parameters work fine; Wordpress seems to ignore them. Session variables work fine too. So far Wordpress hasn’t gotten in my way once and has helped a ton.

• SHOGUN: SHOGUN Project Documentation
• pychem
• Resources for Discriminant Analysis
• Linear discriminant analysis
• Principal components analysis