David Weinberger's Intranet Buzz:
"ClearType and the visual display of knowledge"
By David Weinberger
Editor, Journal of the Hyperlinked Organization
Computer displays suck. Anything else is a euphemism.
(Oh, ok, I'll use the technical term for computer display quality: they exhibit
suckitude.) Even your swell new 21" flat-screen beauty is totally inadequate for
its main job: displaying text. Small type is unreadable and large type has all
the clarity of a newspaper read through a screen door. If computer displays were
capable of displaying readable text not only would your time on line be more
pleasant, but you might actually consider reading long documents on your
computer. And this in turn would make electronic book publishing feasible,
unleashing the same market forces as MP3 has set loose in the music publishing
world.
There is a standard that begins to make good on the
promise. It's called ClearType(tm) and, gulp, the trademark belongs to
Microsoft. [Late breaking news: Adobe has announced CoolType, which Adobe has
differentiated as being "Exactly the same thing as ClearType, but we own it
instead of Them." The big advantage: CoolType will work across platforms.]
The idea behind ClearType has a pedigree. In fact, you
could claim that Steve Wozniak invented it in 1976 to increase the horizontal
resolution of the Apple II's high resolution graphics system. But, hey, what's a
patent among friends? Besides, Woz (or, as I prefer to call him not that I
ever have "Niack") was designing for non-digital displays whereas ClearType
only works on LCDs.
You see, your LCD screen consists of color triads. When
you see a white pixel, you're actually seeing a red, green and blue sub-pixel.
(Yes, I know that mixing red, green and blue should give you a brown about the
color of where you spilled your Starbucks on your aunt's walnut credenza, but
you're thinking about mixing cooking ingredients whereas LCDs mix light, and
since black is the absence of color, then white must be the fullness of color as
you can prove by shining a very bright light in your eyes and noticing that you
close your eyes and thus see black. This was explained to me by Roy G. Biv.) So,
it takes three sub-pixels to make one colored pixel.
ClearType displays type by managing the individual
sub-pixels, tripling the effective resolution of your screen. Pretty damn
clever, eh?
Having more and smaller dots to play with lets the
computer display better looking type. Font rendering software takes the
information that's included when you install a new font ("A TimesRoman 'I' is a
straight vertical line with two horizontal lines at the end," sort of) and turns
it into instructions for turning on a set of colored pixels ("I want 12 pixels
in a vertical row...," etc.). Normally, font rendering software has to make some
pretty ugly decisions when figuring out how to put a particular letter on the
screen. For the sloping parts of the letter, it only has big (pixel-size) blocks
to use, so you get jaggies. For parts of the letter that should be especially
thick or thin (say, the different legs of the letter "m"), it only has one size
block to play with; sometimes it will want to make a leg 1.5 blocks thick but
will have to round it off to either 1 or 2 blocks.
That's why our eyes hurt when reading text on a screen:
Rounding errors.
Font rendering software mitigates some of the effects of
the coarseness of pixels through anti-aliasing which rounds off the edges of
jaggies by turning on gray pixels around the stair steps. ClearType achieves a
better result by using colors instead of gray and, more important, by having
smaller blocks at its disposal.
The results are impressive, and better seen than
described. Steve Gibson (Gibson Research Corporation) has put up a great site
that explains all this stuff in readable, friendly prose, with pictures. And
he's included a 35K downloadable program that demonstrates ClearType vs.
OrdinaryType on your LCD screen. [See the links at the end of this article.]
But there are some limits. First, ClearType only works
for black and white text there are no pixels left over for rendering color.
Second, it only works on LCDs, although on desktop monitors it can give roughly
the same effect as anti-aliasing. Third, because the sub-pixels are arranged horizontally, it only increases horizontal resolution. Fourth, it's from Microsoft.
Fifth, it's still not good enough to make you think you're reading ink on paper.
The Beast of Redmond announced on January 6 that
ClearType will be part of their new Pocket PCs (Motto: "Keep your Palms out of
your pockets"), meaning that the next "convergence" won't be of PDAs and phones
but of PDAs and books. Or so Microsoft would like.
ClearType is an intermediary step that will make your
text-based knowledge literally clearer. We still wait for the day when the back
of the display problem is broken and we have cheap, webbed, portable reading
devices that are as crisp as ink on paper, but with color and multimedia. That's
the point at which working knowledge intimately tied to its expression will
be redefined.
Steve Gibson's ClearType resources page: http://grc.com/cleartype.htm
Download Steve G.'s demo software: http://grc.com/freeandclear.htm
Microsoft on ClearType:
http://www.microsoft.com/typography/
cleartype/default.htm
Microsoft press release:
http://www.microsoft.com/typography/
links/News.asp?NID=1090
The Author
David Weinberger writes JOHO and is one of the Ringleaders of cluetrain.com,
a manifesto of web ethics. He also provides strategic marketing
consulting to high-tech companies, writes for several magazines
(including Wired)
and is a commentator on NPR's "All Things Considered."
He was, as VP of Strategic Marketing, one of the shapers of Open
Text's intranet strategy. David sits on several conference boards
and is a member of AIIM's Emerging Technology Advisory Group. Reach
him at self@evident.com
.