Indeed, page architecture is very important in helping readers parse and comprehend the information on a page -- especially on the Web. As Jakob Neilsen's eye

Indeed, page architecture is very important in helping readers parse and comprehend the information on a page -- especially on the Web. As Jakob Neilsen's eye tracking studies have shown, most people don't read the text on a Web page linearly, they scan it in an F pattern (http://www.useit.com/alertbox/reading_pattern.html http://www.useit.com/alertbox/reading_pattern.html). Good info architects design pages to convey the most important information in these spaces and put supplementary information below and to the right in case the user needs more info. Likewise, they use hierarchical headings to show relationship between information, and bullets and ordered lists to show groups of information. With spreeder, the benefit of all this architecture is gone. You may be able to read the words faster, but do you actually comprehend the concepts and do you have good recall later? I'm skeptical. Now, if a future version of spreeder could parse a page and somehow show these hierarchies and groupings in a simplified way (isntead of just showing one sequential word at a time), that might be interesting.