On Tue, Dec 8, 2009 at 3:48 PM, Brian Kuhn <bnk...@gmail.com> wrote: >> but it's invalid to set it to anything other than the two values I just >> mentioned. > > That's the part I'd like to see changed. I understand that if it's present, > it's on. So, why can't async="true" be valid? I think all browser vendors > will implement it that way anyway. They'd be crazy not to.
Indeed, async="true" does set it to be on. However, so does async="false", async="off", async="no", and any other string you can think of that might imply that it's turned off. Setting it to *any* value turns it on, so it's best to avoid values that would lead to confusing results. The two values that are currently valid are the minimum necessary, and neither have antonyms that would confuse people. ~TJ
