The way they presented it actually makes it harder to read, I think, because the line breaks are way too frequent. Jumping from the end of one line to the start of the next is quite tiring when you have to do it several times to read a single sentence. It breaks up any rhythm for immersive reading.
Believe it or not, I do think there is some potential in this technology, specifically the algorithm for breaking up sentences into meaningful bits. It would help for reading a text in a language you are learning, because parsing sentences in a foreign language can be difficult. Also, this might be helpful in determining more logical places for line-breaks in composing some forms of paragraphs, perhaps ragged text with massive degrees of freedom in line lengths.
Maybe. They did well in choosing Melville as a proof of concept. No force on earth would make me read Moby Dick, but I read through the whole sample they provided without hating it at all. Though that's equal to, what, the first page?
The medical abstract, however, belied the algorithm's weaknesses. Some sentences were presented in a way that did not logically match their true sentence structure, causing confusion; for instance,
"RAPA was developed to provide an [easily administered and interpreted] means..."
seems to become
"RAPA was developed [to provide an easily administered (what?)] and interpreted..."
I'm sure this was discussed on Typophile a couple of years ago, probably when it first appeared (note the 2007 copyright date on the site). Obviously, it's caught on like wildfire.
I don’t enjoy the appearance of the very frequent line breaks, but the key technology that they are showing here is that they are marking off all the phrases within a sentence. Understanding the syntax of a sentence is a major component of comprehension, and the line breaking is helping with that. Eye tracking research shows that we fixate on the last word of a sentence or phrase longer than expected. We seem to try to process one phrase at a time as we read. There have been studies as early as the 1950s showing that indicating phrase boundaries improves reading speed and comprehension. The effect is more pronounced with normal, adult readers on the poorer end of the comprehension spectrum and with children.
Has this been tried with increased horizontal spacing instead of line breaks? (With stacked lines I suppose you'd run into distracting holes in the paragraph.)
Phrases have been marked with extra horizontal space, and horizontal space also works. Tom Bever did a lot of work on this in the late 80s and early 90s using extra horizontal space. His work was primarily about understanding which phrases had the largest impact on reading speed and comprehension. Unfortunately extra horizontal space doesn’t look fantastic either.
Not. But I'm sure someone was awarded a Masters or a PhD based on the research, and it appears that some clever business types were able to get startup money from a venture capitalist.
I found it hard to read due to bad breaks and short line lengths. But we are all a fairly literate group here (only 1/50th of the spelling/grammar errors I see on other boards). Perhaps it becomes more effective for poor readers.
If so, we will soon see it adopted by the public school systems, where they have no compunction against slowing down the top 80% if it might be effective to the bottom 5%.
for me, as a person who reads english as a second language,
even long sentences such as here in typophile can be hard to read.
sometimes the meaning is lost in a sea of words, and i have to re-read the sentences.
so this gadget should be helpful, but the short sentence breaks the syntax of the text and diminishes the original art of typesetting. suddenly the text has no meaning and i have to re-read it and try to re-create what was originaly intended.
Oh, absolutely,
because our
sentence structure
is such that
it would
for example
allow
for very long
and contorted
and hard to understand
and kind of backwards
(and complex
– if you know what I mean
and are actually reading
these poetic lines –)
elephants
(-:
Seriously, that's a good point that (I think) demonstrates how futile the attempt is to visually group text into those little bites that make sense. Because language doesn't necessarily proceed in a linear fashion. Often the opposite is true; as Yaronimus said, this breaks the syntax.
Now German does often have horrid sentence structure in that regard, like putting the verb in the end of subordinate clauses. That can make for strange surprises a bit akin to what I tried to do above; and I don't think this model would help comprehension of such sentences, more like the opposite. :-\
Citation for "…Eye tracking research shows that we fixate on the last word of a sentence or phrase longer than expected. We seem to try to process one phrase at a time as we read…"?
Stiff did an interesting study on semantic line breaking. Eye tracking on top of this could show some correlational data, but in my opinion "easier to read" should be measured by things such as comprehension, time, recall, visual search &c… Not saccades and fixations. But I have to admit, the eye tracking equipment is pretty fun to use.
Stiff, P. (1996). The end of the line: a survey of unjustified typography. Information Design Journal, 8(2), 125–152.
27 Apr 2009 — 1:30pm
Uhm,
not.
:o\
I think
. . . that
. . is a
. . . . lot
of hooey.
And a sly plot to turn everything into free-form poetry.
-=®=-
27 Apr 2009 — 2:11pm
I vote "not," too.
27 Apr 2009 — 2:29pm
The way they presented it actually makes it harder to read, I think, because the line breaks are way too frequent. Jumping from the end of one line to the start of the next is quite tiring when you have to do it several times to read a single sentence. It breaks up any rhythm for immersive reading.
Believe it or not, I do think there is some potential in this technology, specifically the algorithm for breaking up sentences into meaningful bits. It would help for reading a text in a language you are learning, because parsing sentences in a foreign language can be difficult. Also, this might be helpful in determining more logical places for line-breaks in composing some forms of paragraphs, perhaps ragged text with massive degrees of freedom in line lengths.
27 Apr 2009 — 2:31pm
A
n
n
o
y
i
n
g
;
-
)
27 Apr 2009 — 2:45pm
ditto
27 Apr 2009 — 2:45pm
I don't like it offhand, and I don't really expect it to work, but I would hold off judgment in the absence of actual testing. It *might* work.
I'd be curious to hear what Kevin Larson thinks of this.
Cheers,
T
27 Apr 2009 — 3:21pm
Maybe. They did well in choosing Melville as a proof of concept. No force on earth would make me read Moby Dick, but I read through the whole sample they provided without hating it at all. Though that's equal to, what, the first page?
The medical abstract, however, belied the algorithm's weaknesses. Some sentences were presented in a way that did not logically match their true sentence structure, causing confusion; for instance,
"RAPA was developed to provide an [easily administered and interpreted] means..."
seems to become
"RAPA was developed [to provide an easily administered (what?)] and interpreted..."
27 Apr 2009 — 3:32pm
Interesting because the em-space indentations make me unavoidably pause slightly before reading each line.
The effect is dramatic staccato reading that destroys mood, rhythm and flow.
If I was scanning for a general understanding of a passage, however, it might be useful?
27 Apr 2009 — 5:25pm
I'm sure this was discussed on Typophile a couple of years ago, probably when it first appeared (note the 2007 copyright date on the site). Obviously, it's caught on like wildfire.
27 Apr 2009 — 7:01pm
Good memory, Mark - and hard to find, but yes Yves mentioned it here... http://typophile.com/node/34343
28 Apr 2009 — 6:15am
I don’t enjoy the appearance of the very frequent line breaks, but the key technology that they are showing here is that they are marking off all the phrases within a sentence. Understanding the syntax of a sentence is a major component of comprehension, and the line breaking is helping with that. Eye tracking research shows that we fixate on the last word of a sentence or phrase longer than expected. We seem to try to process one phrase at a time as we read. There have been studies as early as the 1950s showing that indicating phrase boundaries improves reading speed and comprehension. The effect is more pronounced with normal, adult readers on the poorer end of the comprehension spectrum and with children.
Cheers, Kevin
28 Apr 2009 — 6:25am
Has this been tried with increased horizontal spacing instead of line breaks? (With stacked lines I suppose you'd run into distracting holes in the paragraph.)
28 Apr 2009 — 8:10am
No, bowerbird...your posts are still painful to read. ;o)
28 Apr 2009 — 9:31am
Phrases have been marked with extra horizontal space, and horizontal space also works. Tom Bever did a lot of work on this in the late 80s and early 90s using extra horizontal space. His work was primarily about understanding which phrases had the largest impact on reading speed and comprehension. Unfortunately extra horizontal space doesn’t look fantastic either.
28 Apr 2009 — 10:15am
Not. But I'm sure someone was awarded a Masters or a PhD based on the research, and it appears that some clever business types were able to get startup money from a venture capitalist.
I found it hard to read due to bad breaks and short line lengths. But we are all a fairly literate group here (only 1/50th of the spelling/grammar errors I see on other boards). Perhaps it becomes more effective for poor readers.
If so, we will soon see it adopted by the public school systems, where they have no compunction against slowing down the top 80% if it might be effective to the bottom 5%.
28 Apr 2009 — 11:44am
Reading this
makes my brain
sound like
Christopher Walken.
28 Apr 2009 — 12:01pm
I was thinking
more
Shatner!
28 Apr 2009 — 12:52pm
…or
bower
bird.
:-[
28 Apr 2009 — 2:27pm
for me, as a person who reads english as a second language,
even long sentences such as here in typophile can be hard to read.
sometimes the meaning is lost in a sea of words, and i have to re-read the sentences.
so this gadget should be helpful, but the short sentence breaks the syntax of the text and diminishes the original art of typesetting. suddenly the text has no meaning and i have to re-read it and try to re-create what was originaly intended.
the result sounds like a dying robot..
tired
have no
power
battery
off.
28 Apr 2009 — 3:04pm
"like a dying robot.."
Now I know what it reminds me of:
Daisy, Dai…sy…
28 Apr 2009 — 4:59pm
Nina:
I believe that
this would be
really useful for
reading in Geman
nicht.
28 Apr 2009 — 5:13pm
Oh, absolutely,
because our
sentence structure
is such that
it would
for example
allow
for very long
and contorted
and hard to understand
and kind of backwards
(and complex
– if you know what I mean
and are actually reading
these poetic lines –)
elephants
(-:
Seriously, that's a good point that (I think) demonstrates how futile the attempt is to visually group text into those little bites that make sense. Because language doesn't necessarily proceed in a linear fashion. Often the opposite is true; as Yaronimus said, this breaks the syntax.
Now German does often have horrid sentence structure in that regard, like putting the verb in the end of subordinate clauses. That can make for strange surprises a bit akin to what I tried to do above; and I don't think this model would help comprehension of such sentences, more like the opposite. :-\
28 Apr 2009 — 5:41pm
It might just be
that some people
rely on this mind-numbing
effect of the short lines
to hide the fact that
they have in fact
nothing new to say.
28 Apr 2009 — 8:23pm
> Reading this
makes my brain
sound like
Christopher Walken.
Haha, nice one. :D
29 Apr 2009 — 8:42am
Ooof! What has he ever done to you?
He's great in Pulp F. and in Fatboy's video "Weapon of Choice"!
2 May 2009 — 7:50am
> Ooof! What has he ever done to you?
He’s great in Pulp F. and in Fatboy’s video “Weapon of Choice”!
Yes, he's awesome.
2 May 2009 — 9:47am
Kevin,
Citation for "…Eye tracking research shows that we fixate on the last word of a sentence or phrase longer than expected. We seem to try to process one phrase at a time as we read…"?
Stiff did an interesting study on semantic line breaking. Eye tracking on top of this could show some correlational data, but in my opinion "easier to read" should be measured by things such as comprehension, time, recall, visual search &c… Not saccades and fixations. But I have to admit, the eye tracking equipment is pretty fun to use.
Stiff, P. (1996). The end of the line: a survey of unjustified typography. Information Design Journal, 8(2), 125–152.
2 May 2009 — 10:35am
Thanks for the reference Christopher.