On 30.04.2010 16:11, Boris Zbarsky wrote:
On 4/30/10 8:02 AM, Nikita Popov wrote:
I personally prefer using <h1-6> and do not see, why always using <h1>
may be better.
If you're the only author, sure (maybe; see below).
If you're one of several co-authors on a document, with sectioning
structure in flux, and you're supposed to produce one of the sections
but don't know how deeply nested your section will be in the final
document, then how are you supposed to use <h1-6> correctly?
Yeah, I see.
What would make much more sense, is to omit the rank fully, so only only
has to write <h>.
Yes, and I'm fairly certain that's been discussed in the past. But
<h1> already exists, and people already use it and the new uses will
degrade gracefully in UAs not supporting the new spec text much better
than uses of <h> would.
I don't know whether I would be happy, if all headings in my document
were shown *BIG*, 'cause I use h1 everywhere. I would much more
appreciate them to be unstyled. (But this is only personal opinion.)
Beyond that, using <h> instead of <h1> would even be more backwards
compatible to the HTML 4 use of headings.
But not to existing HTML4 UAs...
As said above, I think it increases backwards compatibility by omitting
all styles. But it depends on the case whether it's better to have only
big titles or only unstyled titles.
I easily think that using h1 everywhere isn't semantically correct.
Especially if the subsections (with their h1s) cannot be redistributed
solely it does not make any sense.
<h> would mark up a heading, <h1> does mark up an heading of highest rank.
But maybe you are right. The html5 spec is already blown up with stuff
nobody will ever use (keygen?) enough. In this case less is probably more.
