FWIW, it's now in Firefox 4. I believe we can wait for authors' feedback if any.
Marking this REMIND for now. I'll look at this again in 6-12 months and see what author feedback is like. If UAs move towards this or if authors make good use of this, then it makes sense to add. If UAs do not move towards this or if authors either do not make use of it or abuse it, then it makes sense to keep it out.
mass-move component to LC1
This bug was cloned to create HTML WG bug 19069.
Mounir, have you heard any feedback on this?
Authors usually complain when things do not work as they were expected them to work, not when things work as they expected them to. No one complained but no one said that was a good feature too. In other words, I got no direct feedback.
I don't know what could be the next steps. Maybe check if Webkit had bugs filed? Maybe try to get developers feedback by asking for it? Though, I am not even sure if <output> and forms constraint validation is that much used.
FWIW, I believe that the best thing to do would be to apply that change and maybe revert it if it happens that it is badly used (though, I'm not sure how that could be "badly used").
I don't really understand what it means for <output> to be invalid. There's no way the user can edit it, after all.
I couldn't find any developer chatter or bugs about this anywhere. I don't think there's compelling evidence that authors want or need this.
I'm leaning towards WONTFIXing this unless there's some information that I missed. Anyone know anything relevant that I should look at or consider?
WONTFIX per comment 8, given the lack of further information.