I recently posted a question on UI.SE and I initially used some textual representations of common UI elements such as radio buttons and check boxes. Later on I found that there were Unicode characters for representing such elements. So I switched the textual representations from ( ) to ○ and (o) to ◉. Similarly, my check box representation went from [ ] to ☐ and [x] to ☒. It seemed that all answers got the idea of what I was suggesting by using the words "radio button" and "check box" along with my textual representations. I wonder then, if either is correct, or if an image should be used instead to ensure proper display? For example, if a user were on a device that did not have a Unicode font – for whatever reason – these characters would not display.
I wonder then, if either is correct, or if an image should be used instead to ensure proper display?
By all means, use images, or use unicode UI glyphs if it helps communicate your point in a question or answer.
if a user were on a device that did not have a Unicode font – for whatever reason – these characters would not display
I think this should be rare enough that it is not a practical concern.
Except for people still using Windows XP and who do not have admin rights on their box (read corporate wienies). For example, using Firefox I could read the alternates for the radio button, but the alternates for the check box are squares with numbers in them. In IE 7, I can read the alternate for the open radio button but the closed radio button and the checkboxes all look like empty squares.
There is one reason you may want to stick with the simple text versions of the symbols. The fact that questions and answers are likely to keep pouring in with the simpler notation means that:
- We would never be able to standardize on ○, ◉, ☐, ☒.
- We could standardize on ( ), (o), [ ], and [x].
Trying to "upgrade" every post to the special characters will be never-ending work (and, at least for this example, for what seems like little increase in clarity, if any, especially since the special characters seem too small). Also, due to the inherent lag time in doing it, readers will always be exposed to both versions, so none of the benefits of standard representation will ever appear.
Reverting all of the special characters to the simpler notation would probably be a one-time effort (or, even if some users unaware of the desire for standard representation did use the special characters, would mean less never-ending effort...which, by the way, could more safely be made automatic (find-replace) while at the same time mostly eliminating lag time).