On Oct 4, 2008, at 7:01 AM, Tab Atkins Jr. wrote:

On Fri, Oct 3, 2008 at 8:24 PM, Andy Lyttle <[EMAIL PROTECTED]> wrote:
On Oct 3, 2008, at 4:35 PM, Tab Atkins Jr. wrote:

Man, I could *really* see the "hint" function being viable and quite useful. It offers up a completely new-and-useful semantic, and there's no particular place it should already go. I'd accept this as a new attribute without reservation if it was renamed @hint, so it's absolutely clear what the semantic for it is.


The only reason to use placeholder instead of hint is that Apple already implemented placeholder. Documentation should explain that placeholder is to be used for hints, not for labels (and people can then ignore the documentation and use it for labels anyway, but at least we tried).

Well, we don't really have interop yet, since *only* Webkit implements it currently, and officially only on a single non- standard input type (though it happens to apply to text and similar input types). If we can shift the name over *now*, before FF implements it fully, it would probably be fine.

On the other hand, I don't want to be one of those jerks who tries to block a feature solely because they don't like its name. However, it's a proven fact that most people don't look at documentation *ever*, and so having the name provide a perfectly intuitive hint for what the attribute is supposed to do would probably be best. At the very least it would set up some cognitive dissonance for people using it as a label, hopefully.

I think "placeholder" is as good a name as "hint"; it may sound less "semantic" but it's more clear that it would result in a tooltip like "title" would. That being said, it would not be an excessive burden for us to support "hint" as well as "placeholder" for compatibility.

Regards,
Maciej

Reply via email to