WCAG 2.0 Comment Submission

Name: Chris Lilley
Email: chris@w3.org
Affiliation: W3C
Document: W2
Item Number: Make it easy to distinguish foreground information from its background 
Part of Item: 
Comment Type: TE
Comment (Including rationale for any proposed change):
The term \"luminosity\" is incorrect here (it applies only to certain broadcast video signals). Relative luninance is the correct term. When used as a ratio, the difference between absolute and relative luminance can be dropped, but the term luminance rather than luminosity should be used.



http://www.w3.org/TR/2006/WD-WCAG20-20060427/appendixA.html#luminosity-contrastdef

defines the coefficients for calculating luminance, and its good to see that the Rec.709 chromaticities are used (rather than, for example, the NTSC ones which do not apply to modern computer monitors at all). (the coeeficients are correct, see the \"luminanceToAlpha\" section of http://www.w3.org/TR/SVG11/filters.html#feColorMatrix but the specification does not explain why this is so.



However the text as it stands implies that the formula given is universally applicable. it is not. This is particularly importnt when printing Web materials.

Proposed Change:
change \"luminosity\" to luminance throughout. 



change \"The luminosity of a color is defined as\" to \"the luminance of an sRGB color is defined as\".



change \"blue RGB values\" to \"blue sRGB values\" (the equation does not apply to other color spaces).



Remove the exponentiation operator and the 2.2 gamma approximation. Instead, use the correct sRGB transfer curve.



Reference the sRGB specification. See the SVG 1.2Tiny specification for an example of how to reference it.



You may want to note that the equation given is only correct for typicalcolor vision (the ISO standard observer). For atypical color vision, often incorrectly termed color blindness, different equations apply depending on the type of  atypical color vision and the degree of severity.

For more information, please see

http://www.w3.org/Graphics/atypical-color-response

Received on Monday, 22 May 2006 15:44:36 UTC