Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


The Math of Text Readability 282

Posted by Zonk
from the looks-good-to-me dept.
An anonymous reader writes "Wired magazine has an article that explains The Law of Optical Volumes, a formula for spacing the letters on a printed page that results in maximum readability. Wired's new logo (did anyone notice?) obeys the law. Unfortunately, Web fonts don't allow custom kerning pairs, so you can't work the same magic online as in print. Could this be why some people still prefer newspapers and magazines to the Web?"
This discussion has been archived. No new comments can be posted.

The Math of Text Readability

Comments Filter:
  • by Anonymous Coward on Thursday April 19, 2007 @06:30PM (#18806215)
    "Someday, when we're all running ultra-high-res displays, and someone releases a shitload of completey free (as in beer and freedom), high-quality fonts (I think this is the biggest issue, personally), then we'll all see the same nice fonts on our computers."

    Welcome to slashdot were doing anything that goes on the internet isn't really "work".

    "Traditional fonts usually included several hundred kerning pairs. Hoefler & Frere-Jones' fonts are super-fussy - they can include 10,000 pairs to get every combo of letters exactly right. "

    That's work, and that's only a small part of what's known as typography.
  • Web Site Readability (Score:2, Interesting)

    by hattig (47930) on Thursday April 19, 2007 @06:39PM (#18806289) Journal
    Unfortunately the WIRED headline "underwire" doesn't obey those rules.

    I'm generally unhappy with kerning on websites, unless they use certain fonts (sorry, I've never cared enough to look them up, although oddly enough they were serif fonts whereas I like sans-serif on websites).

    The biggest issue for readability was:

    - not too small
    - decent line spacing
    - NOT black on white. Dark grey on white, or black on pale grey
    - Nice margins to other content

    (aside, remember when people used to call them founts back in the 80s?)

    I've actually found the Wii Opera browser quite readable even on a 576i PAL TV (once zoomed in on the content anyway), and I attribute that to decent fonts and colours.
  • by michaelmalak (91262) <> on Thursday April 19, 2007 @06:42PM (#18806319) Homepage

    Also the simple fact that many people like the freedom of being able to actually hold and move around the thing they are looking at.
    The term you're looking for is Picard's Syndrome [].
  • Disappointing (Score:5, Interesting)

    by Anonymous Brave Guy (457657) on Thursday April 19, 2007 @06:47PM (#18806361)

    I thought someone might finally have come up with some serious research showing how to objectively improve readability, but it's just a summary of kerning.

    Why is this area so bare of real scientific results? There have been a few studies into on-screen readability, typically measuring things like reading speed, accuracy of recollection afterwards, and subjective approval of the document by the reader. However, there are so many variables that people don't seem to control that it's hard to see any general patterns. For example, changing the font from 10pt to 12pt on screen may well not just scale the size by 120%, but also make the dominant strokes two pixels wide rather than one. There is little consistency among conclusions about optimal font size for reading across fonts or whether serif or sans-serif fonts are more readable, perhaps because there are so many variables.

    Oh well, I guess we'll just have to wait a bit longer for comprehensive research.

  • latex (Score:5, Interesting)

    by dheera (1003686) on Thursday April 19, 2007 @06:59PM (#18806475) Homepage
    This is exactly why MS Word sucks and LaTeX is awesome, at least in terms of readability. Try reading a LaTeX'ed documunt on the screen, it is extremely pleasant.
  • by Anonymous Coward on Thursday April 19, 2007 @07:51PM (#18806939)
    When come to font and screen, very few people gets it. Even in the article, "Web fonts in 2007 still don't have kerning pairs", what does it means? Anyone care to exlain? As far as I know, quite a number of fonts in my machine do have kerning pairs, and web browsers rendered them nicely. There are exceptions, of course, but that does not mean Web fonts don't have kerning pairs. I know this, I have done more complex work than just kerning (like mark positiong,ligatures, contextual alternatives etc).

    One thing that the article fails to explain is, when comes to screen reading, kerning is just one facter. The other VERY important factor is how well the glyph is rendered. In truetype fonts, this is achieved through proper hinting, and I really mean PROPER. As far as I can tell, there are very limited fonts that are properly hinted, even the commercial ones. To make matter worse, part of hinting are patented, make it more difficult to be available on platform like Linux (without using the patent, that is).
    Freetype has done a termendous job on getting their "autohinter" working, but it will never be the same as font hinting. This is because when a font is hinted, the designer decides how it will exactly look like at specified resolution, while with autohinter, the program have to "figure out" how it suppose to look like. You know which one will produce a better output. Of course, all this does not apply to printed material where the printers are high resolution.
    So far, even though truetype hinting is aging, it seems like there is no better substitude in the pipeline.
  • by Anonymous Coward on Thursday April 19, 2007 @10:11PM (#18808039)
    It's not purely a matter of numbers. It's a matter of skill and experience (read scarcity of talent). And yes, if one truely wants quality fonts, it takes lots of work (especially a "shitload"). Read up on the history of some of the more popular fonts. While some company or organization may donate some fonts, that in no way mitigates my argument. And yes it does bother me (hence my reply) that slashdot has little to no understanding of how much work goes into anything digital. You're not going to hear a post were someone says "a shitload of houses are suddenly going to appear" because most people understand physical goods. They don't understand digital, and it shows. e.g. piracy.
  • by Sique (173459) on Friday April 20, 2007 @05:33AM (#18809877) Homepage
    The main difference between normal text and programs is that in texts the smallest semantic unit is the morphem (which is mostly a syllable), while in programs the smallest semantic unit is the single character (or symbol). It thus makes sense for programs to use monospaced fonts, because then every semantic unit has the same size.

    But we read text by reading morphems, and the reader even can easily be confused by hyphenation through morphems (re-adi-ng is difficult to read compared to read-ing), and morphems don't map to symbols in the letter based alphabets anyway, so each morphem has its individual size to begin with. Monospaced fonts don't help spotting morphems in letter based alphabets.

    It's different in Chinese or Korean though, where each morphem has its own character, and those languages use - monospaced fonts!
  • by kalidasa (577403) on Friday April 20, 2007 @07:18AM (#18810303) Journal
    Windows can do it in some applications. Actually, the one thing you can't fault Microsoft on is their work on fonts, typesetting, and encoding. Security? They're idiots. Usability? They're at best semicompetent. But in the NT/2K/XP/Vista line, they've done as much with OpenType, Uncode compability, and readability as any other OS vendor/group.
  • by Anonymous Coward on Friday April 20, 2007 @09:54AM (#18811697)
    Looks fine on my system. But then I'm on an OS with competent font rendering. If you want to see what well-rendered "web fonts" (a thorougly bogus term) look like, see these three screengrabs I've just taken: [] [] []

    It's particularly noticeable for the word 'Lyrid' in 'Lyrid Meteor Shower', and the VAVAVA line in the second post. Oh look, it's even kerning it correctly in this text box as I type.

    I run at 1360x1024 on a 17" display, by the way. The first grab appears to me at about 3" by 5", the second at about 2.5" by 6". Give it another 5 years or so, other OSes should have caught up by then . . .
  • by pz (113803) on Friday April 20, 2007 @10:38AM (#18812307) Journal
    No, you're wrong. On the screen font you happen to be using, under the OS you happen to be using, on the browser you happen to be using, using the rendering engine you happen to be using, it might or might not support kerning. So, you might or might not see the effect.

    To a great extent, the resolutions available for computer screens sufficient to even think about kerning have only recently become available to the mainstream. Kerning is a subtle, but important, effet that most screen fonts are designed to not require (because of the limited rendering resolution).

    You can, however, easily tell the difference between a properly hinted font where the hints have been correctly used and one where things are just wrong by printing out some text on a decent printer (ie, nearly anything manufactured in the last 2 years). If you typeset, for example "VA" on a kerned font, and very very carefully compare where the "V" ends and the "A" begins, you'll see they overlap just a hair. As in 1/72nd of an inch. The difference between a properly typeset font and one that's lacking kerning is the difference between a beautifully drawn pen-and-ink illustration and something hacked together with Powerpoint.

No problem is insoluble in all conceivable circumstances.