Some Q&A on WebAnywhere - a screen reader on the go

Last week, I read a post on
Fred’s Head Companion,
about a new and innovative screen reader being developed at the University of Washington.

After reading that, I was interested enough to visit the official
WebAnywhere – a screen reader on the go
site to learn more about this forward-looking web-based application.

Here is the initial information presented on the site:

“WebAnywhere is a web-based screen reader. It requires no special software to be installed on the client machine and, therefore, enables blind people to access the web from any computer they happen to have access to that has a sound card. No $1000 software program required!”

“WebAnywhere runs on any machine regardless of what operating system it is running and regardless of what browsers are installed. This is its advantage over existing products like SA-to-Go. “

The launch date for this ambitious project is late May, 2008. Keep watching the product’s site for the official release.

Additionally, there is a link on the home page to a Youtube video demonstrating the product in use by a blind student. Alternatively, one can also download the file and play it on your own computer.

Okay, as a screen reader user for over 10 years and having a keen interest in assistive technology, I’ll admit that after reading that on the home page and checking out the video, I was even more intrigued by the promise of what they were presenting. I had a few questions I wanted answered and emailed Jeffery Bigham, a graduate student with WebAnywhere who is listed as the project’s contact person to find some answers. Below is our dialogue.


Q: This is a web-based application that serves as a screen reader. Do I understand correctly that it is only in the web browser that this application speaks, not in any other application?
If this is true, are there plans to make it more functional in future generations, so that it might also read email in pop3 clients or documents in .doc, .pdf, or even .txt formats?

A: WebAnywhere speaks only the web, but the web is becoming a platform on which all the other applications on your desktop will be running in the coming months and years. Only having web access would not be sufficient now, but web trends indicate that it may enable access to almost everything users want to do in the near future. Currently, you can already access your email, PDF files, etc. using various available web applications.


Q: This appears to be cross platform, as I do not see a specific web browser mentioned. Is this correct, or is it maximized to work better in one browser over another?


A: Our goal is to make a screen reader available on any computer or device that would provide a sighted person web access. WebAnywhere is entirely web-based and will work on any web-enabled platform using any web browser. That means it will run on the Linux, Windows, Mac operating systems and in Internet Explorer, Firefox, Safari, Opera, etc. And, not only on full computers, but also on mobile devices like smartphones and PDAs.

One current problem with screen readers is that they're always playing catchup. A new physical device is released and screen reader users have to wait until someone comes along to write a screen reader for that device before they can use it. With WebAnywhere, access is immediate on any web-enabled device.


Q: I assume your team is familiar with the recent announcement of the Accessibility Is a Right (AIR) Foundation, which provides the System Access To Go screen reader by Serotek free of charge to any blind computer user with a web-enabled computer. What sets your application apart from the SAToGO software?

A: The SAToGo software is a great development and for some people it may be most appropriate. But, SAToGo will only run on Windows from Internet Explorer, and SAToGo won't run on locked-down public terminals because it requires permission to run new software. SAToGo takes 3-4 minutes to download, compared to almost instantaneous loading for WebAnywhere. WebAnywhere is also open source, which means that anyone out in the community can improve it.

All of that said, in cases where users can run SAToGo in the short term it might be more appropriate for them.

Our initial goal is to provide a base level of accessibility to the web on ANY computer or device that is available. As we move forward, we'll work to improve this interface to make it even better.


Q: I don't recall reading anything regarding cost for using the WebAnywhere application. Will it be offered free to the public or will there be a cost involved?

A: The short answer is that WebAnywhere was designed from the start to be free. We've also released it with a minimally-restrictive open source license, which means that anyone that wants to use it, modify it, or host it, can do so.

But, the devil is in the details, and we're still sorting some of them out. For example, we don't yet know how we'll actually support a public version of WebAnywhere if it becomes popular. The costs of such a system are relatively low per user when compared with other systems, but, depending on how popular the service becomes, it could still require a lot of resources. This isn't to say that it's impractical - popular existing services like YouTube, MySpace, Gmail, etc. also require a lot of resources. They, however, are fortunate enough to be supported by large companies with many servers. We've been actively talking to a number of companies and organizations who have expressed interest in trying to get this off the ground. Nothing is final as of yet.

Q: Will there be any registration process?

A: Initially there will not be a registration process, but, as the system matures, we might move in that direction. Many of the components used in the system could be used for both good and bad purposes. Without a registration process, it's difficult to control which purpose are using it for. One of the traditional ways of getting past this is to use a CAPTCHA - we'll obviously not do that.
Registration also has benefits in preserving users preferences regardless of what computer they're accessing their information from.
Registration would not, however, affect the cost of the system (it would still be free), and we hope to always make some version of the system available for free.

Q: Realizing that WebAnywhere is being built with the trend towards the future use of web apps, will it work on release with any of the already available web-based apps, such as Google docs or Google Spreadsheets?

A: WebAnywhere does not yet support Google documents or Google spreadsheets, but it's definitely one of the things (among many) that we're working hard to support. One of the advantages of WebAnywhere being a web application is that in many ways it's easier for it to interact directly with other web applications.

One of the reasons why we made the project open source is to hopefully attract other developers to the project.

Q: I think the work you are doing is definitely worth sharing. I would like to publish a post about WebAnyWhere and include some information from the answers you've shared on my blog and in email correspondence with both my professional and blind peers. Would that be all right.

A: Feel free to share my responses with whoever you want. To make the project really successful, we need to get the word out and hopefully get some people to contribute to its development!

My sincere thanks to Jeff for taking the time to answer my questions, even when in the middle of a trip to China.

So, do you have any questions I haven’t covered here? In one of my next few posts, I’ll share my initial thoughts on the on the go accessibility that WebAnywhere promises to bring.

Also, please share this with anybody you know interested in assistive technology in general, or screen readers specifically.