WebAIM - Web Accessibility In Mind

Assistive Technology Experiment: ZoomText

As mentioned in a previous post, I recently decided to spend some time becoming more familiar with a few common assistive technologies, starting with the screen magnification software ZoomText. I have shared a few of my experiences with ZoomText below.

My preferred configuration

I am quite nearsighted, so I decided the best way to evaluate a screen magnifier would be to remove my glasses. With my glasses off, I found that my preferred configuration included the following settings:

  • Power: 4X magnification. I would really like a 3.5X magnification option.
  • Type (part of the screen that is enlarged): Usually full. During the first few days, I spent a great deal of time trying out these different views, including dual monitors, but I found I like to devote the entire screen to magnification.
  • Mouse Pointer: Large yellow.
  • Color scheme: Normal. I find the inverse color scheme easier on the eyes, but images, especially images of people, look too strange.
  • Cursor: None.
  • Focus: None. While doing keyboard-only testing I find red outline helpful.
  • Reader: On, surprisingly. I had not planned on using the reader feature, but I find the constant narration actually makes things more usable.

Accessibility thoughts and recommendations

I am well aware that a few weeks experimenting with a screen magnifier does not make me an expert, and that some of the difficulties I experienced were due to my lack of familiarity. Still, I feel I gained valuable insights. I have listed a few web accessibility recommendations below.

  • Ensure adequate contrast. As expected, it is harder to read text with low contrast. Instances of light gray text are very difficult to read. Sometimes I have to increase the magnification above 4X to read text with poor contrast.
  • Use true text when possible. True text is definitely easier to read than text in images, but I am surprised to find that in most cases where text is used in images, the images remain fairly readable, even at high magnification. I think this is because I am not able to see the pixels as clearly (no glasses), so the otherwise jagged text is still smooth and readable. When the default color scheme is changed, text in images sometimes becomes very difficult to read, especially when the background included a gradient. I had not previously considered this situation. It provides an additional reason to avoid text in images when possible.
  • Use a proper heading structure. This can definitely enhance accessibility, especially on pages that are content-heavy.
  • Use good content separators such as whitespace or color change. Whitespace is helpful, but I sometimes get "lost" if a page has too much whitespace between sections. I assume that a page or section has come to an end when it has not. I find it easier to navigate through pages that use color and other non-whitespace separation between page sections.
  • Provide a flexible page layout. I usually keep my window at about 800-900 pixels wide. This offers a good readable line length without breaking too many page layouts. While most pages display content well, some pages display a horizontal scroll bar (which I can’t see because it is not onscreen) or have content sections that overlap at this width. It seems that many of these problematic pages have a three-column layout (sidebars on the right and left side).
  • Use succinct, descriptive link text. Long links with filler words (e.g., "Click here for more information on…") often require panning or scrolling.
  • Be careful with rollover menus. These are kind of a mixed bag for accessibility. On one hand, they occupy less visual space, making it easier to see content on the page. On the other hand, the menu itself is often very difficult to navigate. When creating rollover menus, avoid menus that are too long or that contain more than one level of links.

Other insights/thoughts

  • While I have typically used the term "screen enlarger" to refer to this type of software, it seems that "screen magnifier" is the more common term. I will try to use this term from now on.
  • 2X magnification (a WCAG recommendation for enlarging all content as well as text) is actually quite a bit bigger than it might seem. Many (probably most) sites on the web do not support text resizing to this level. I cannot image a situation where a user would double text size without using a screen magnifier to enlarge other content on the page as well (text in images would probably be unreadable to this person). What I can imagine is a user who resizes text to a more moderate size (say 1.5X) in conjunction with a screen magnifier. While the WCAG recommendation for text resizing is a good goal, I think that an otherwise highly accessible site that supports 1.5X test resizing is in a pretty good place.
  • After returning to my normal resolution the screen seems small and difficult to read for five minutes or so.
  • Screen magnifiers and projectors don’t play nice. The last time I tried to demo ZoomText in a presentation, it blue-screened my computer.
  • I actually find using a screen magnifier kind of enjoyable at times, especially at the end of a long day. It is nice to remove my glasses and "relax" my eyes a bit. This probably suggests that I need to improve my screen location and resolution.
  • It took me weeks to realize that the "logo" in the ZoomText window was actually a button that would allow you to quickly enable and disable ZoomText. I guess it pays to read the manual.
  • I have a very difficult time typing more than a few words with ZoomText enabled. There is something about only being able to read a few lines of text that makes it very difficult to compose my thoughts. This is the one area where I couldn’t keep from cheating during my "final exam" (a full day using ZoomText).

Importance of screen magnifier testing

This was a very worthwhile experience and I feel that I learned quite a bit. I plan to continue to use ZoomText and demonstrate it in trainings (assuming I ensure it works with the projector beforehand). However, as valuable as this experience has been personally, I still think the built-in browser zoom and text resizing functions are adequate for most accessibility testing. These provide a comparable experience with no cost and little-to-no training. If I were to add an additional recommendation, it would be to test inverted color schemes, but this is something that can be done at the level of the operating system.

Next up: Dragon

Next on my list is Dragon NaturallySpeaking. Look for a similar post on Dragon in the next month or two.

Comments

  1. Shawn Warren

    Nice report. Please give us (Ai Squared) a call when you are ready to try to use it with a projector. 802 362 3612. We demonstrate ZoomText this way frequently and I’m confident our support staff can isolate and fix the problem.

  2. jill

    I just wanted to agree that the reverse video/ reverse brightness option are nice but I have issues dealing with images as well. This is really unfortunate as dealing with browsers colour settings is annoying (ie and chrome) I have tried just the replace colour to swap just white/black and leave all other colours as is but it won’t let you do that.

  3. Joanna Jakubiak

    Did you try holding “Ctrl” key and scrolling your mouse? That should work in most browsers and in many common programs (e.g., Word, Excel, Adobe, etc.) — the simplest solution, literally, at your fingertips.

  4. Cynthia Jones

    This information is well and good however your services fail to identify the fact that asking one to select radio buttons, check boxes and/or text boxes to complete information is not alwasy viable and may offer inaccurate selections.

    Not all text to speech screen reading software acts as JAWS which is not affordable to the majority of most who are visually impaired. There needs to be ammended actions to remove radio buttons, check boxes and to list in the text box require action rather than list before to ensure the user can clearly identify and provide accurate information. Also noting that screen readers do not read Capital letters, Punctuation marks, and other image files such as astrik, hash mark and or other graphical images without proper alt tags.

    The fact is technology is continually advancing but many attempt to overlook and take forgranted that all system work alike as a measure of trying to eliminate human interface.

  5. Abe

    Thanks for sharing your experiment.
    As ‘Joanna Jakubiak’ has pointed out,
    holding “Ctrl” key and scrolling the mouse, is a handy tool.