On 07/06/05, Phil Boswell phil.boswell@gmail.com wrote:
I am using it semantically.
== Major Header (=== generic explanations) // hidden header ==== quick explanation 1 // which applies to all sub-header sections ==== quick explanation 2 // which applies to all sub-header sections === Sub-header 1 === Sub-header 2
I don't want that first "===" header to show because it breaks up the flow: it's not uniform with the rest of the "===" headers.
Well, logically it *is* "uniform with" those headers, because it's at the same level. I think I agree with those saying that you shouldn't "skip" heading (or rather, section) levels in this way. In general, nothing but the very introductory lead-in of an article should be headerless, assuming headers are going to be used at all.
Consider the following example which has an "implicit" sub-head which I guess you'd call "2.0". Is the text at D on a par with that at A or that at B, C and F? In other words, does it belong to section "2" or section "2.0"? Surely the clearest structure is obtained by simply rearranging the text to make the answer explicit.
== Head 1 A === Sub-head 1.1 B === Sub-head 1.2 C == Head 2 D ==== Sub-sub-head 2.0.1? E === Sub-head 2.1 F
However, if it can be done without too much of a hack, I guess a feature that numbers as though such "implicit" headers were actually there would be reasonable as a better guess at user intention.