It's a clever idea, but anecdotally, from my experience, I'm finding it slows down my reading -- I'm having a hard time processing the blurbs because I don't read "linearly" -- I scan content to find the relevant parts, and the color changes are making it difficult to scan (because my eye can no longer use color to determine what is scannable and what isn't), and multiple columns is actually making it even more difficult (it looks like the blue in column 1 leads into the blue in column 2, instead of the blue at the next line of column 1). By trying to force me to read line-by-line, instead of scanning efficiently, it's making me read slower.
But that's just for short-form stuff. It could turn out to be faster for some layouts, and slower for others. But honestly, I've never felt I had difficulty locating the start of the next line... is this a problem that needs solving? But nevertheless, it's certainly a good example of clever out-of-the-box thinking.
My wife reads 10x faster than I do but completely eschews the kind of reading I like (viz, literature). But of course, now that, y'know, she has kids too, her overall reading speed is strangely reduced as well.
Anyway, at a first glance I found the line coloring helpful.
Very many concepts, words, metaphors, etc come from the King James translation.
Ignoring any religious stuff - it's a good read.
EDIT: corrected my st / king error!
A point of view on King Jame Bible: https://bible.org/article/why-i-do-not-think-king-james-bibl...
"It was my resolve to live watchfully, and never use my tongue amiss; still, while I was in the presence of sinners, I kept my mouth gagged, dumb and patient, impotent for good. But indignation came back, and my heart burned within me, the fire kindled by my thoughts, so that at last I kept silence no longer.
"Lord, warn me of my end, and how few my days are; teach me to know my own insufficiency. See how thou hast measured my years with a brief span, how my life is nothing in thy reckoning! Nay, what is any man living but a breath that passes? Truly man walks the world like a shadow; with what vain anxiety he hoards up riches, when he cannot tell who will have the counting of them! What hopes then is mine, Lord? In thee alone I trust. Clear me of that manifold guilt which makes me the laughing-stock of fools, tongue-tied and uncomplaining, because I know that my troubles come from thee; spare me this punishment; I faint under thy powerful hand. When thou dost chasten man to punish his sins, gone is all he loved, as if the moth had fretted it away; a breath that passes, and no more. Listen, Lord to my prayer, let my cry reach thy hearing, and my tears win answer. What am I in thy sight but a passer-by, a wanderer, as all my fathers were? Thy frown relax, give me some breath of comfort, before I go away and am known no more."
-Psalm 38 (39) from Knox's Translation of the Vulgate.
Knox's translation is lucid when compared with the "correct" and dead modern translations of the bible. Try it if King James does not speak to you.
But if you choose to read it as literature, I certainly won't stop you!
Besides we have MUCH better access to older sources now, and what do we find? Contradictions, contradictions everywhere.
Set the chunk size to three for best effect. Just gradually increase it each time, and also try using it occasionally with the speed far too fast. You'll get used to subvocalizing less, and reading multiple words at once.
Of course, for complex texts like the ones you read, it's natural to read much slower. I read 600 WPM for most internet articles, but I'm sure I wouldn't read plato that fast, nor as continuously.
That technique is called Rapid Serial Visual Presentation: http://en.wikipedia.org/wiki/Rapid_Serial_Visual_Presentatio...
(The Wikipedia article should mention saccades (http://en.wikipedia.org/wiki/Saccade) but strangely it doesn't.)
I wrote an RSVP reader in the mid-to-late 90s in QuickBasic. I'll have to see if I can dig up that code. :)
I am still annoyed that things like the Kindle don't have that mode.
Has anyone made a syllabic RSVP reader before? Might be interesting to see how much more natural it would be (my hunch is "a lot").
In terms of increasing reading speed, you may want to try the utility "dictator" for those texts which are electronic. For physical books I would run a finger under text to maintain momentum as Adler suggested in "How To Read a Book". Make notes in the text, being sure to use a custom set of symbols to speed notation. Write concise thoughts in the margin, don't go overboard.
For a hacker, it is normal to not subvocalize.
I think it is also one of the reasons that we hate meetings so much. An email can be written at about 40 wpm, and read much, much, much faster. So past 2 or 3 people, the dominant strategy should be text, unless sidechannel meaning is needed (where emotions are involved or whatever).
That's been debunked a while ago. There's no such thing.
What it doesn't help with is transposing from adjacent lines when reading the middle of a column, which happens to me all the time. The middle of the column is always black, so there's not much of a cue to keep you on the same line.
In making my workaround to get a similar effect in calibre, I made a variation I much prefer which cycles red->green->blue, so that column centers are always easy to separate.
You can change the font size and line height by a little bit without needing to adjust anything (since the color gradient is continuous vertically, it is a bit tolerand), but if you make a dramatic change, you'll need to adjust the background height from 72px to the height of three text lines.
Would love to see it confirmed or falsified.
(As a note, I am probably under this hypothetical threshold; I'm not a slow reader, but I find that efforts to speed up my reading speed generally push it down instead.)
Anyway, my experience from this is that it was much easier to differentiate, parse, and scan with the ugly colours (43% faster apparently). However, my natural slowness in reading is in both understanding and committing to memory. I get the feeling that this did not help much with regard to this..
For me, I found the changing color to be too distracting. My eye was constantly being drawn away from the text I was reading to another color on a line below or above.
It may be worth noting that I only have one eye, and it's not a very good one - 20/100 visual acuity with about a 15deg. visual field.
If you don't mind me postulating, I wonder whether this difference is something intrinsic to the focussing effect of binocular vision, or if your field of view has maybe trained your brain to be more urgently perceptive to peripheral vision?
then again, maybe its just colour perception.. or different strokes (and thanks, I just learnt about binocular summation.. fascinating)
Either way, quick experiment shows it to be horrid for me.
Here's what I mean: http://www.stationerytrade.com/upfile/images/big/20122221112...
What about using eye tracking? A device could just detect you're at the end of a line, and put a small marker at the start of the next line.
But I agree that it sucks for quick scans - the different colors just add another layer of complexity for our brain to process...
I felt (without experimenting in a measurable way) that particularly the blue-red transitions _really_ slowed down my "normal" reading…
edit: although in this case, they have a patent pending and apps in development, and are offering licensing agreements.
Regarding your scanning approach. An idea I've had is to format each word based on how common it is.
Still, I think it is a cool idea and will give the bookmarklet a try for a while.
"It looks like BeeLine didn't improve your reading speed this time through."
Then I read it. Fast. Consuming nearly at a paragraph at a glance when I usually can digest only a fragment of a sentence up to a couple of sentence.
It's not attractive, but it is clever and innovative - well done!
I 'constantly' have to re-read entire paragraphs because I realised I've looked at them without really taking anything in. It was really strange to feel like i was processing the text as I was reading it.
I'd especially be interested in exploring ways to incorporate this into better designed color schemes so that it doesn't look so much like a unicorn vomited on the page while preserving the benefits and usability.
I'm also less inclined to dismiss improvements like these after misinterpreting the occasional email from colleagues lately. I don't know if it is assuming I know the full contents from the 3 line summary on mobile devices, processing too much email, or simply not paying enough attention but I've had to slow down and make sure I get things right.
Or, to mix in another idea (from philip1209) downthread: have the hover-effect add color-matched dots (or other glyphs) at the corresponding line-ends, currently under the pointer.
Moving the mouse/cursor down the page line at a time to read? _Seriously?_
What I'm worried about is comprehension.
So why isn't the study linked?
Regardless of whether or not the claims are true, who in the hell decided for red and blue for the demo's default? The blues/grays themes look okay. IMO, saturated red and blue and probably the two worst colors to use together in a design.
I'm curious what others experienced?
According to them I read 11% faster. Although I guessed on 2 out 3 questions, compared to knowing all 3 from b&w text.
Thus, I call BS on their testing experience. These two texts are way too different. And picking an easier one for beeline text does a disservice to this hopefully legit fast reading method.
That said, I often find myself fiddling with the cursor to mark my place when I'm reading something long, so the idea that extra indicators of place might help doesn't surprise me.
I feel better line spacing might also help - their test seemed to have the lines pretty tight together, although I'm not surprised a test would be designed to best showcase their software.
Then I took it again with a different subject but I timed it.
It took me ~81 sec to read the colored, ~92 to read the black. It said I read 23% faster with BeeLine. What gives?
i found the colors to be rather distracting and not helpful. the results of two tests taken seem to agree with that.
EDIT: Some feedback after reading a Cracked article with it.
First of all, since the inception of the Readability bookmarklet I've always read online articles with some kind of tool (I started with Readability, then passed to the Safari version and now I've been using Clearly for quite some time and I'm pretty happy about it) and now I'm so used to it that if a particular article doesn't render properly, I just straight out don't read it. The first thing I noticed is that the coloration is applied even to single-line titles, I would do away with it and (maybe) apply it only on multi-line paragraphs' titles. The other thing that irked me is that small images are put on the left side instead of being centered, even worse is the fact that text appears on the right side of the images; I would follow Clearly steps in this regard and always put the text under the centered images. Lastly, I would reduce the text area to 600px of width or better yet, dynamically size it so as to accommodate around 60 characters. As far as I can tell, you totally nailed the font size.
Please give more options for colors and gradients. I'm sure everyone has different tastes and habits.
FWIW, I liked it.
So it is good to remember that it might or might not work depending on the way your individual brain works, independently of what the person next to you gets from it.
"BeeLine Reader is an exciting new technology? That helps people read faster? On computers?"
Maybe I'd get used to it. But if not, a 100x speed increase wouldn't be worth having that in my head. Like, all day?
Line height, line width, and font size, are the 3 elements of a body of large text that must be balanced. There is some basic arithmetic to this, and it appears everyone seems to ignore it. (HN's design is a perfect example of the 'meh' attitude toward readability.)
I think this article does a good job of visually explaining this balance and how to accomplish it: http://www.pearsonified.com/2011/12/golden-ratio-typography....
The catch is, you will almost certainly read the second passage slower than the first, since you're now looking to retain information for the questions!
The colored passages _feel_ faster, but I'm not sure that counts for much.
Perhaps the aggregate A/B numbers make a more compelling case for using BeeReader?
I wonder if the color combo choice has any affect on the speed/comprehension of the text.
This thing combines the old Readability bookmarklet with the gradient. I saw the improvement right away, following the line is much easier now!
tl;dr - this is awesome!
Why not show me the stats? I'd like to know, even if it doesn't confirm what you want it to.
Surely the author is not claiming that putting color on text gives them som sort of patentable intellectual property? If this takes off and people will start incorporating this on their blogs, this company will become one of the biggest patent trolls.
Better idea: Chill out. Then take 1 minute and read some stuff with it on. If you think it feels better try it for longer if not move on with your life.
No has claimed to cure cancer here, just that formatting text differently might give marginal increases in reading speed.
I think this is interesting, but I'd want to see studies on what the incremental cognitive load of "color matching" (not something I'm particularly great at) does to reading comprehension, and what the impact of this is on how much I can read in a sitting. It's no good if you read 50% faster but tire out 3X as fast.
Err. Line transition errors are common on mobile devices, not because lines are short (the shorter the line, the less common line transition errors are), but because people are usually moving, walking, etc while holding a mobile device.
I don't like magicscroll because of the way the lines scroll down; I find it disconcerting. In the case of Beeline, I can't stand the color scheme.
The goals of both software are admirable and I'd love to see more work in this space, but I don't think either of them have it exactly right. If the designer is on here, consider using an interior design color picker website to find a color scheme that works better than the current one.
In short, this is a problem and it would be valuable to somebody like Amazon if it were polished, IMHO.
http://colorschemedesigner.com/ as an example.
First, we wanted to make it really obvious what's going on, and if we'd used a subtle color gradient, it wouldn't have been as obvious. We realize that many/most people won't ultimately use the bright color scheme for one reason or another.
The second point is that people perceive color differently, so what is bright to one person may not seem so bright to someone else. There tend to be age-related correlations/causes here--it's why old people tend to wear lots of bright blues. To them, the blues don't look as bright. What we've found is that younger folks tend like the more subtle colors (eyes are more sensitive) and older folks tend to like the brighter colors (because they can't perceive the color difference in the subtle schemes). We anticipate rolling out our product/feature with a color picker palette so that users can choose the right color scheme for their visual physiology.
Thanks again for all the comments and suggestions, and please feel free to email us through the website if you have an app/site that you'd like to use BeeLine with. We have some code that will make integration pretty easy.
It shows only the first paragraph.
(I know, I'm a sad person, but that's the sort of text-heavy doc I have to read all day).
I would really appreciate some way for it to automatically do it and not take me to a new page, maybe something I could install into my browser?
Insidently, the speak function of iOS is amazing for people with dyslexia. I use it all the time to listen to text at speeds of 300+wpm. I know many of my friends can read at 600+wpm my them selves, but not for a few hours on end. In any case, if you are dyslexic and use iOS, check out the read function under accessibilities.
The other problem I see (mostly on Chrome) is that headings are anti-aliased, but the body text is not. The difference is subtle, but still noticeable.
A few things: it would be nice to be able to configure the plugin to limit the color variations. I'd like to try with only 2 colors and with less drastic contrasts. I suspect that only a slight transition between two close colors would already be helpful enough for me.
Now, I'm afraid to get used to the crutch and find it harder to read books after. After using vim to edit almost anything, I have developed the bad habit of pressing ctrl-[ to go in normal mode any time I'm in some text editor, be it in the browser, email client, word processor, whatever.
The examples in the bottom of the page really helped me realize how much this helps. Seriously, I read those paragraphs with the "Bright" theme and then I read them with BeeLineReader disabled ("Off") and I could notice my brain working harder.
I realize it looks ugly as other commenters have posted before, there's probably another method that doesn't make the text look so "ugly".
A couple of problems: 1) beelining doesn't work well with links in text 2) Doesn't work on Hacker News at all.
Once again, thanks for being honest in your test and not convincing me to use something that might not actually help me.
I'd love to integrate this with my iOS speed-reading app (http://velocireaderapp.com). Any interest in collaborating on a spin-off iOS app, to read, say, DRM-free ePubs? My contact info is in my profile.
1. It'd be very nice if you had a version that tweaks text colors and doesn't touch anything else, i.e. just like on the demo page. When I tested your bookmarklet and it tore up the page I thought something was broken. I only found out that it uses readability because I started digging when it 'broke', you never mention 'readability' in your copy.
2. People will want it enabled by default. You can't do that if you use readability.
It should keep the layout as is, while only adding color, but I have only tested it on wikipedia. The color period is wrong, but it will show you some approximation of how sites will look with a non-interrupting beeline bookmarklet.
Looking at the text on their site, I suspect (aside from issues like dyslexia) that the real problem is that a lot of people are reading text that's too small and probably has overly-wide lines.
I'm wondering if just adding reference points along the margin or between lines could accomplish the same thing without having to change the text color. Something similar to the tick marks along a graph axis.
I created a (very hacky) style sheet to do this in calibre: https://gist.github.com/osuushi/6456804 . It gets a bit out of alignment when a paragraph wraps to a new column or page, but over all it gets the job done.
Edit: I fixed it to do one color transition per line, like the original.
Overall, a very cool idea, I was surprised to find that I read faster. It said only 3% faster, but I searched for an event mentioned in the first set of text which slowed me down.
Apparently, when I'm reading on screen, somewhere towards the middle of the line, I switch my focus from my left eye to my right. This makes it obvious, because with the color at the end of the line, I switch too soon, and miss the third of the line in the middle.
Possibly this is a consequence of wearing glasses.
A different approach but also bookmarklet
Not to say it's not interesting / not a valuable submission. I love the idea, and it seems like it might help me read faster, which is always cool. Just wondering if my memory is correct.
Just wondering what the result would be. Out of curiosity.
PS - I am a voracious reader.
It might be great for some fiction?
The test said that it did not improve my reading but didn't say why.
The reason is because it matches the default colour for links. I wouldn't be surprised if many people tend to read linked text a bit differently.
Have you thought about using colours like orange, green, purple?
Then I got suspicious. I thought that I was subconsciously affecting my own behavior. (Anticipating a test, for instance. Expecting Beeline to speed up my reading, for instance.)
So I did Nature, used Bright, and got no improvement.
...I need a better blind.
I wonder if it would hamper your normal reading abilities if you start reading like this most of the time from young age.
Awesome marklet! Is the source available? I'd be happy to fix this issue up myself if it's in VCS somewhere.
get wiped out, making the text difficult to understand.
Also, be careful to wait for it to work. I didn't think it worked and clicked it a second time. I ended up with funfetti colors, not smooth gradients.
I like the effect, though. I hope they can optimize it technically.
Seriously though, it would be interesting to see this as a feature in new e-readers. I have a feeling that if the e-ink could support it, the effect could be better than books.
Reading super fast over the National Inquirer is not likely to be an overall win.
I just came back just now, just in case it was a co-incidence, and yep, it was not a co-incidence.
FWIW, when I looked the demo, I thought to myself, "Ok, I'm going to try to read this quickly." What was so impressive was how easy that was to do. Total comprehension, less eye strain. Sold.
Why do we need fancy gradients?
Addition: for eink readers, would underline or italics work in place of color gradient?
 http://en.wikipedia.org/wiki/Line_printer#Paper_.28forms.29_... "The paper was usually perforated to tear into cut sheets if desired and was commonly printed with alternating white and light-green areas, allowing the reader to easily follow a line of text across the page. This was the iconic 'green bar' form that dominated the early computer age. " (emphasis mine)
BeeLine doesn't detect line breaks.
Instead it puts a span around every character and sets its color manually.
Anyway the individual spans cause issues for MagicScroll but it should be relatively easy to retrofit MagicScroll to add the beeline gradient to each line.
Not usable for sites, but very much so for articles