Theory: text-blocks and separators

[Jorn Barger]

Probably, before the XML-crowd can outgrow their fanatic loyalty to
hierarchies-of-containers, someone will have to come up with an equally
well-ordered alternative model.

So to make a start at that:

- pages can be analysed into blocks of content (text-paragraphs, images,
and forms I guess) visually differentiated by (content-free) separators

- text-blocks normally have variable 'aspect ratio' (you can make them
wider or narrower by adjusting the height correspondingly)

- in print, text-blocks are often split to fit the end of the page, but
on the web this should never be necessary. (it's easy to break one block
into two or more, though-- ie, sentences into paragraphs.)

- separators will normally include whitespace, sometimes with other
visual elements (eg horizontal rule, surrounding border)

- whitespace may provide vertical displacement; horizontal displacement
is usually done as indentation (only the left edge displaced to the
right, the right edge is constant, or displaced _left_ symmetrically)

- font-variations can assist separators in differentiating text-blocks
(face, style, size, color). These variations also work _within_ a
text-block.

- tables and lists can be formatted as 'stacks' (columns) of small
blocks, with identical separators between them. (Such units may be
grouped hierarchically, but this is a convenience, not a structural
prerequisite.)

- side-by-side text-blocks (ie columns) should be used very sparingly on
the web, imho

This vocabulary can probably be extended to include windows and
scrolling, etc...

--
http://www.robotwisdom.com/ "Relentlessly intelligent
yet playful, polymathic in scope of interests, minimalist
but user-friendly design." --Milwaukee Journal-Sentinel