It seems like the OpenGL ES SL 1.0 spec isn't very close to reality. Someone should probably raise an issue with the maintainers of that spec in whichever way is appropriate and then we should spec whatever matches what real shader authors expect and all implementations accept. In testing I found that different implementations would accept but behave differently when given shaders with different strange inputs (see http://www.khronos.org/webgl/public-mailing-list/archives/1012/msg00133.html), so we should be careful that whatever we allow in WebGL behaves identically across all implementations.
On Thu, Jan 13, 2011 at 1:44 PM, Gregg Tavares (wrk) <firstname.lastname@example.org> wrote:
One more point of info,
The OpenGL ES 2.0 Conformance tests have shaders with these quotes in comments.