Default Place-Holding Characters

Gez Lemon asks whether Default Place-Holding Characters in form controls are still an accessibility requirement. I’ve always felt that this checkpoint was weird and could cause more problems than it solves. Apparently it is no longer something you should do, according to the WCAG 2.0 working draft: Place-holding characters in empty controls (deprecated):

This is a negative technique, an example of something you shouldn’t do. This technique corresponds to a WCAG 1.0 checkpoint. There is a proposal for an erratum that states the until user agents clause is met and this checkpoint (and thus technique) are no longer necessary.

Posted on August 23, 2005 in Accessibility, Quicklinks

Comments

  1. August 23, 2005 by paul haine

    I cheated on that requirement; glad it’s being removed.

  2. August 24, 2005 by Jens Meiert

    In a short correspondence last year, Joe Clark and me also agreed on this. Personally, I never used placeholders for input elements since testing never revealed any problems, assuming that you label your forms and write sufficient microcontent, among others.

  3. August 24, 2005 by Small Paul

    The WCAG 1.0 checklist (http://www.w3.org/TR/WCAG10/full-checklist.html) says “until user agents handle empty controls correctly”.

    As far as I know, the only user agent that had a problem with empty form control was Netscape 2, which I think we can safely consider obselete.

    Thus I think leaving out placeholding text entirely is fine, as far as meeting that guideline goes. That’s the approach I’ve always taken.

  4. August 25, 2005 by Steve Williams

    WebXACT flags empty form controls as a Priority 3 Error and so far I have acquiesced and put them in, but I’ve always felt placeholding text gets in the way and I never quite understood why it was needed…

    What SHOULD we use (besides common sense of course) to show clients how accessible their site is or isn’t, if the online tools do not reflect current thinking?

    I can’t envisage most clients sitting through a rendition of WCAG 2.0, but they seem able to get their heads around an online report that says good / bad…

Comments are disabled for this post (read why), but if you have spotted an error or have additional info that you think should be in this post, feel free to contact me.