Skip to 0 minutes and 0 seconds DOMINIQUE ARCHAMBAULT: Hello. I am Dominique Archambault, professor at University Paris 8. In this short animation, we will try to understand the principles of software accessibility and see who is involved and at what level. Two tips– this animation shows better when displayed in full screen and don’t hesitate to use the pause button. Oh, last thing before we start– this is, of course, a simplification. Things are a bit more complicated inside a computer. The goal here is not to explain in detail how assistive technologies work, but to understand the principle and what is needed for accessibility. I have represented here a computer with two displays connected. Let’s have a minimum of imagination.
Skip to 0 minutes and 42 seconds This large rectangle at the bottom is the central unit of your computer. And on the top right, a screen is connected to it with some text displayed– to first verses of the famous soliloquy of Hamlet. Now you see on the top left, another display connected to the computer. It is supposed to represent a Braille device, which displays a line of text in Braille, usually 30 to 40 characters. Some keys, here with the arrows, a user uses to navigate within the content of the screen. Currently the first line, “to be or not to be,” is displayed here in Braille. When I press on the arrow key, second line, “that is the question,” comes. OK? Let’s come back to the first line.
Skip to 1 minute and 28 seconds Now let’s open the central unit to see what’s inside. You may recognise several parts. Here, I have represented the hard drive where all your data is stored. I have skipped a lot of parts that are not interesting for the demonstration, like the motherboard, the CPU, the network card, et cetera. Here, I want to focus on two other elements– the main memory where the application on the data you are currently working with are loaded on the video card which is driving the screen. The work of the video card is to transform the information to be displayed into a video signal for the screen.
Skip to 2 minutes and 5 seconds In the main memory, there are now two pieces of software– the application that you want to work with, let’s assume it is a browser here but it could be any application so we will call it “the application”– on the operating system of your computer. It may be Windows, Mac OS, or Linux. At the moment, the Braille device is not yet connected. Let’s imagine we have a file stored in the hard drive which contains our text. When we open it in the application, its content is loaded into the memory in the area of our application. The application creates a layout and sends it to the operating system.
Skip to 2 minutes and 43 seconds Then the operating system will calculate the image for display on the screen and send to the video card a table of numbers representing the colours of each pixel of the screen. And finally, the video signal is produced and our text appears on the screen. OK, but, how does it come to the Braille device? The Braille device need a special piece of software that is called a screen reader. Screen readers are also able to talk using speech synthesis but we will only focus on the Braille part. The idea is the same for the speech activity. So the screen reader is also loaded into the memory and will drive the Braille device, more or less, as the video card drives the screen.
Skip to 3 minutes and 24 seconds To better understand the origin of software accessibility, let’s see how first screen readers were functioning back in the ’90s where computer were able to display only the text. You might remember these computers with those ugly green characters. Or maybe you’ve seen some in old movies. What was different inside at this time? Well, the screen was only this green text so the operating system would directly send text to the video card without bothering about pixels. The video card would contain the characters themselves to display. And the card would produce a video signal from it. At this time, the first Braille devices had just been invented.
Skip to 4 minutes and 2 seconds And the first screen reader will directly read the text in the video card memory and send it to the Braille device. It was as simple as that. But in the ’90s appeared graphical screens. This implied a big change in the way a computer works. And finally, software accessibility features started to be needed. As we’ve seen, now the video card contains only numbers corresponding to colours of each pixel of the screen. So the screen reader cannot read any more useful information there.
Skip to 4 minutes and 30 seconds Several solutions have been tried at this time, but very quickly the community realised that the only efficient way will be to have a backdoor in the operating system allowing the assistive software to get directly the information from the system. At the beginnings, the operating system editors were very reluctant. And it has been a long story, that I won’t tell here, to make them implement that by creating an accessibility layer into the operating system itself. Finally, today, all major operating systems have an accessibility layer that allows a screen reader or any other type of assistive technology to ask for the information it needs. Here’s the text of the first line.
Skip to 5 minutes and 7 seconds This also had an implication for the application developers who had to implement an accessibility interface so that the application can send to the operating system what it needs to feed the assistive technology. Now, with this accessibility framework implemented in the operating system and in the application, the text can be displayed on the Braille device. So it seems we have an accessible computer. And the accessibility looks like a very technical thing which was solved by the developers of operating system and of the applications. Then why would orders of documents have anything to do to contribute to accessibility? Actually, it is not as simple. Indeed, now we have a computer with a nice graphical screen.
Skip to 5 minutes and 50 seconds So we have all sorts of possibility to put images on the screen. For instance, I want to illustrate this text with an image– a drawing of a skull that usually the actor playing the role in Hamlet is holding in his hands when he says these verses. The image will be stored in a graphical format on the hard drive and link to the main document. Let’s load this document in memory again. The application creates a layout and sends it to the operating system, which again creates an image to put on the screen. Now our document is properly displayed on the screen. OK, but what should the operating system send to the screen reader? It cannot guess what’s on the image.
Skip to 6 minutes and 28 seconds So now you’ve guessed. It is the author who have to say that. If the author adds an alternative text into document. Here the text is a skull. Then the application will be able to get it and send it to the operating system to be transmitted to the assistive technology. Now the text, a skull, can be displayed in Braille. And of course if I press the next button, the screen reader will ask what comes next to the operating system, gets the text of the first line, and displays to the Braille. The alternative for graphical content is one of the things that author have to provide to contribute to software accessibility, which is easier to illustrate.
Skip to 7 minutes and 7 seconds But a few of those things are necessary. Digital content accessibility is only a question of following a set of simple accessibility guidelines. To conclude, we have seen that software accessibility has three dimensions, corresponding to operating system, to the application and to content. The operating system must have an accessibility infrastructure, able to exchange information with any assistive technology, and must provide an accessibility interface to application developers. This is very technical, but only developers of operating systems have to do that. Then the application has to implement this accessibility interface to send to the operating system what it needs for feeding assistive technology. This is also technical. Easier, but still quite technical. But it’s implemented by the application developers.
Skip to 7 minutes and 55 seconds On the third level is that users, everyone, should create accessible documents. This is not difficult at all, actually. Everybody can do it. It only requires that users will conform to a few simple guidelines. Thank you.
Understanding the principles of software accessibility
Many of you may find the animated video in this page a little too technical for your requirements.
However we encourage those of you who would like to understand more about this topic to watch this video on ‘Understanding the principles of software accessibility’ created by Professor Dominique Archambault, University of Paris 8. The slides are available to download in PDF format from the bottom of this page.
It is important to understand that software accessibility is basically what allows these alternative access methods to a computer to function, in the same way as city accessibility, such as curb drops, allows those using wheelchairs to access pavements with ease.
As you will see in the video, it is rather more complex when thinking about computer technology, if you are not technically minded. But if you divide the process into three dimensions it becomes easier to understand.
The operating system
There is a need for accessibility support in the operating system. It transmits information from mainstream applications to access software (for instance the text on a menu item from your favourite word processor, for the screen reader to read it). It specifies an accessibility interface (an API) so application developers know how to expose this information to the operating system and what to expose. Nowadays, most operating systems have a correct accessibility interface (see links for further information).
The operating system also usually offers some accessibility settings that make it easier for some users to use the computer, such as high contrast themes, magnification and different mouse pointers or steadier control.
Then the application must implement the accessibility interface: for instance put alternative text on image buttons and expose them to the system, so it can be transmitted to the screen reader. This assistive technology will read out the action of the button and many other navigational elements.
Even with an accessible application on an accessible system, user content has to respect some accessibility guidelines. We will look at some of these content guidelines in the step after you have read about Maria’s challenges and tried a screen reader.
There are many links related to this topic available from the bottom of this page.
© This video is created by Dublin Institute of Technology and Université Paris 8 Vincennes Saint-Denis and licensed under CC-BY-SA. Erasmus + MOOCs for Accessibility Partnership.
© This text is a derivative of a work by Dublin Institute of Technology and Université Paris 8 Vincennes Saint-Denis, and licensed under CC-BY BY 4.0 International Licence adapted and used by the University of Southampton. Erasmus + MOOCs for Accessibility Partnership.