Thx for the pointer. But the question is, does googlebot/indexer actually use this process to weed out hidden texts? If yes, why do we still see so many people adding keywords in hidden sections? Plain ignorance?
Because they don't necessarily "ignore" or "penalize" for it, especially when it's easily displayed by the end user. If they did some major sites would get hit and that would be bad for their results -- It's a common, accepted practice, so even though they can detect it, it's not an "automatic problem" for rankings.
I use a graphic of fancy text for the header description on a page. I have display:hidden on a H1 equivalent of the graphic.
I was told once to maintain a document outline so that the SE's can figure out what the page is about. Others have said the <title> element would serve that purpose and a H1 saying the same thing would be seen as a duplicate.
On the page with the display:none on the H1 and the graphic I then made the <title> significantly different from the H1 only sharing some of the keywords.
Am I walking a gray line between allowed and gaming the SE? I think I'm leveraging the markup as intended as the H1 matches the graphic. Cloaking = SE's see something different than the visitors.
H1 can say the same as the title, but most people expand on the title in the H1 - that is not "duplicate content" even if title and H1 are the same. The preferred title length is limited and the H1 is not so it isn't unusual to have the same text in both places, but taking advantage of the extra length for H1.
Google prefers that you don't use images in place of text, but that is not an unusual practice either, though it is best if the alt tag is accurate and not used for some other text. The "display:hidden" is more likely to arouse suspicion than an H1 tag that parrots the title. None of this is cloaking.