Today all attributes of an html-element is service's
(applied information is between its openning and closing tags).
At the same time, data are saved in database as records, which is
as xml-elements, attributes of which contains applied information.
Traditional way of visualization of these tags
is bad, because it does not allow to specify styles for visualization of these attributes.
I offer to specify attributes in css-file with sign "ž" before their names,
that withdraw attribute from tag and transform it into enclosed tag.
Let's name these tags as virtual tags.
Thus the following construction
So we see, that michanism, based on property content,
is essentially narrower, than michanism of virtual tags.
I don't want to reduce importance of property content:
it is necessary to change already existing content of html-element,
or to specify content of pseudo-element.
But when we try to reduce all cases to this property,
it acts as limiter,
and thus data are subdivided into two groups depending on their forms:
if data is between an open and close tags,
then it is data of full value (it may has style);
if data is in an attribute, it is inferior.