Several months ago I was searching for resources on how to write a decent – “good” – Alternative Description, also known as an “ALT description.” The use of the ALT description has been around for a number of years and is one of the most important coding adaptations needed to make a digital document accessible to people with disabilities. While this type of coding evolved out of the HTML world, it is important to remember that alternative descriptions should be used for images in all digital documents including word processed, PDF or desk top published documents.
Fortunately, the use of the ALT description has gotten wide notice. I’d like to say that it has become well known (and well implemented), but sadly most digital documents are still prepared and published without the inclusion of any alternative descriptions for images.
You probably already know the definition and purpose of the ALT description: it is underlying code added to images that have been inserted into a digital document. The code is used by assistive technologies called screen readers to provide users who are blind or have visual impairments with a textual description of the inserted image.
For years, the debate of “when to use the ALT description” and “when to use the Null ALT” (i.e., ATL=”” in HTML) has been discussed within the accessibility community. My own personal – and completely unscientific – research revealed that there has been no universal agreement on this debate. My observations among screen reader users has been that those users who had become disabled later in life, or who still had some limited vision, preferred that every and all image have an extensive and detailed alternative description. On the other hand, those born blind seemed to me to care less about the “pretty pictures” – as one friend called them – except in situations where the image was a “text graphic” and contained information needed by the user.
For years, I have operated with the philosophy that states “less is best.” I have determined that nearly all of the images added to digital documents are in fact “pretty pictures” and that since they do not provide any important meaning to the document, it is best to use the Null ALT. It should be noted that an equivalent method for adding a Null ALT to digital documents, other than HTML/XHTML documents, does not exist. For word processed, PDFs and the like, you will still need to add some text to create any alternative description. See my series of articles about accessible documents.
My brevity philosophy apparently has some support. A new resource by Australian, Dey Alexander, which takes a similar position as mine, has recently gotten some attention in the accessibility world. Alexander suggests that for purely “decorative” images, and for those “informational” images that repeat information already in the body of the document, the Null ALT is perfectly acceptable. When I was a graduate student, my professors always preached that a well written research document describes everything the person needs to know in the text of the paper and that the charts, graphs and images are only there to support the written content – not the other way around.
At this point I am not ready to discuss the ALT descriptions for completely visual content like graphs and charts. You’ll have to stay tuned for my opinions on that. Suffice it to say that these purely visual elements present a great challenge for people concerned about accessibility and for screen reader users as well.
But getting back to the point of this commentary…I have not (yet) found the definitive explanation for how to write decent/good alternative descriptions. I suspect that any offering in this regard would be something no one could agree upon and result in more cantankerous debate.
A few months ago, fellow accessibility guy, Cliff Tyllick from Texas offered this suggestion on what a good ALT would/should be. Cliff wrote that a good ALT description would be “…a succinct statement of the meaning you expect a sighted person to get from that image.”
Sounds reasonable to me, but I can see the hair-splitters fighting over the word “succinct.”
For a final thought, I’ll turn back to another article by Dey Alexander called “The 5Es of Content Usability” which speaks to the need for Effective, Efficient, Engaging, Error tolerant and Easy to learn writing. I know that “a picture is worth a thousand words,” but maybe it is safe to assume that you should use less than a thousand words in an ALT description for that picture.
Happy New Year.