James Graham wrote:
As Anne and Julian have pointed out, that's not a use of data-*
attributes permitted by the spec.
FWIW I think we have a problem in that multiple independent people have
seen data-* and assumed they are for externally readable metadata. If
Right. That is a problem.
the bad effects of people using it for that purpose are worse than the
usefulness of the feature, we need to consider removing the feature. If
the bad effects are not so bad we need to consider legitimizing this as
it is clearly something that it is clear will happen irrespective of
what the spec says.
I think it indicates that many people are looking for this feature.
Introducing something like RDFa (or something similar) would end that
confusion.
BR, Julian