SIGIA-L Mail Archives: Re: [Sigia-l] new research: reformatting online text to improve readability
Re: [Sigia-l] new research: reformatting online text to improve readability
On 5/11/07, Laurie Gray <laurie.gray_at_gmail.com> wrote:
Coming across on Slashdot this morning was this article on a new
technique to break text down into visual/syntactic "chunks" to improve
readability. This is done algorithmically, but if there is anything
that we can glean from it, it could have a strongly positive impact on
user experience. Note - since it was slashdotted, site response times
Hi Laurie, and all.
Although I found the research very interesting, specially because it
focus on syntact and the way it affects our reading proficiency, I
think there are some issues regarding the presentation of the
information that should also be considered.
The way we read and understand a text is not only influenced by it's
content. The visual presentation of the text is crucial - that seems
to be one of the issues this research tries to explore, proposing
another way to format a given text. But it's not only the length of
sentences, and the points were the paragraphs are placed that would
make it easier to read and understand.
The typographic style is crucial to make a given text easier to read.
The letter form, the space between lines, between words and
characters, the length of the text columns, the margins that let the
whole text breath on a page...for centuries these issues have been
studied in printed text - although there is still many books around
with very bad typographic solutions!
Still, it seems to me that on the web (and other devices such as
mobile phones) we have much to do in order to achieve the same
efficiency in composing good reading material.
I've downloaded the paper "Visual-Syntactic Text Formatting:A New
Method to Enhance Online Reading" at
http://www.readingonline.org/articles/r_walker/ where they present
the research, it's methods and it's findings.
I've found it very inspiring. But it deals with only one "layer" of a
complex issue - the visual presentation of a given text. There is
clearly an interesting approach to try to understand the way we "read"
a text. But there is no way to really understand it if one does not
consider the typographic style of the text, as it is part of this
whole "reading and understanding" complex system.
At some point, the paper shows two different text blocks. One of them,
poorly formatted, with a bad typographic solution, and the other one
using the same type but with the "visual-syntactic text formatting"
Comparing the two blocks, it's most likely that everyone would find
the second one better to read. But it's hard to tell if this choice
would be based on the efficiency of the visual-syntactic solution of
the second text block, or based on the poor typographic solution of
the first text block itself, that made it difficult to read anyway!
That's the same critic we can make to the example they show on Venture
The typographic solutions (and the quality of the images compression)
are SO different that they're barely comparable.
I'm not saying that the research is a fraud or anything like this
(although I have doubts about the methods they've used). I've found it
very interesting and promising. But there are some variables in
"reading and understanding" that should be addressed as well, and I
think in this case they have underestimated the importance of
What if we first try to apply all the knowledge that we have from
centuries of type design to the online material we're developing? :-)
prof. mauro pinheiro
universidade federal do espírito santo
centro de artes
depto. de desenho industrial
IA Summit 2008: "Experiencing Information"
April 10-14, 2008, Miami, Florida
When replying, please *trim your post* as much as possible.
*Plain text, please; NO Attachments
Searchable Archive at http://www.info-arch.org/lists/sigia-l/
Sigia-l mailing list -- post to: Sigia-l_at_asis.org
Changes to subscription: http://mail.asis.org/mailman/listinfo/sigia-l
This archive was generated by hypermail 2.1.6
: Sat May 12 2007 - 14:06:02 EDT