Readability of Reverse Color Text

19 Nov 2008 - 2:10pm
5 years ago
11 replies
2714 reads
Loren Baxter
2007

Hey all,

Reverse color text (white text on a black background) seems to come up in
our discussions from time to time, and usually someone chimes in that it's
less readable. However, I know quite a few programmers who switch their
editors to a reverse color scheme due to eyestrain, including myself. And
the statement that reading black text on a white screen "is like looking at
ants on a lightbulb" has always rung true with me.

So, here's a little research on the subject:
http://www.joedolson.com/articles/2006/08/on-the-readability-of-inverted-color-schemes/

According to the study, the readability is the same either way so long as
the type is adjusted correctly, and it comes down to personal preference.
Just thought I would share that and see if anyone has thoughts.

Loren

-----
http://acleandesign.com

Comments

19 Nov 2008 - 8:08pm
mark ahlenius
2008

Hi,

I did not read all this research but I am curious if studies have shown
black on white to be more tiring on the eyes for a length of time or
other side effects (due to the back lit nature of the screen and staring
at it for a long time). I wonder if any optometry or ophthalmologist
schools/research centers have looked into this. Actually come to think
about it, all the eye charts I've been tested with are black on white.
I'd be surprised if there studies out there to test this. ?

'mark

Loren Baxter wrote:
> Hey all,
>
> Reverse color text (white text on a black background) seems to come up in
> our discussions from time to time, and usually someone chimes in that
> it's
> less readable. However, I know quite a few programmers who switch their
> editors to a reverse color scheme due to eyestrain, including myself. And
> the statement that reading black text on a white screen "is like
> looking at
> ants on a lightbulb" has always rung true with me.
>
> So, here's a little research on the subject:
> http://www.joedolson.com/articles/2006/08/on-the-readability-of-inverted-color-schemes/
>
>
> According to the study, the readability is the same either way so long as
> the type is adjusted correctly, and it comes down to personal preference.
> Just thought I would share that and see if anyone has thoughts.
>
> Loren
>
> -----
> http://acleandesign.com
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

19 Nov 2008 - 9:01pm
Loren Baxter
2007

I'd be interested to see that research too.

There is a difference between black text on a white screen, and black ink on
white paper. The screen is much brighter and at a much worse resolution -
so the eye tests (or any other printed text) are difficult to compare to a
computer.

Loren

-----
http://acleandesign.com

On Wed, Nov 19, 2008 at 5:08 PM, Mark Ahlenius <m_ahlenius at comcast.net>wrote:

> Hi,
>
> I did not read all this research but I am curious if studies have shown
> black on white to be more tiring on the eyes for a length of time or other
> side effects (due to the back lit nature of the screen and staring at it for
> a long time). I wonder if any optometry or ophthalmologist schools/research
> centers have looked into this. Actually come to think about it, all the eye
> charts I've been tested with are black on white. I'd be surprised if there
> studies out there to test this. ?
>
> 'mark
>
> Loren Baxter wrote:
>
>> Hey all,
>>
>> Reverse color text (white text on a black background) seems to come up in
>> our discussions from time to time, and usually someone chimes in that it's
>> less readable. However, I know quite a few programmers who switch their
>> editors to a reverse color scheme due to eyestrain, including myself. And
>> the statement that reading black text on a white screen "is like looking
>> at
>> ants on a lightbulb" has always rung true with me.
>>
>> So, here's a little research on the subject:
>>
>> http://www.joedolson.com/articles/2006/08/on-the-readability-of-inverted-color-schemes/
>>
>> According to the study, the readability is the same either way so long as
>> the type is adjusted correctly, and it comes down to personal preference.
>> Just thought I would share that and see if anyone has thoughts.
>>
>> Loren
>>
>> -----
>> http://acleandesign.com
>> ________________________________________________________________
>> Welcome to the Interaction Design Association (IxDA)!
>> To post to this list ....... discuss at ixda.org
>> Unsubscribe ................ http://www.ixda.org/unsubscribe
>> List Guidelines ............ http://www.ixda.org/guidelines
>> List Help .................. http://www.ixda.org/help
>>
>> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

19 Nov 2008 - 10:35pm
jet
2008

Disregarding how quickly or efficiently studies have shown that we can
parse text under different color combinations...

I'm somewhat light sensitive and I can easily say that black text on a
white screen is something I can only handle for an hour or two at most
when I'm looking at a computer screen. I use a Mac and Adobe CS*, so
this means I take breaks on a regular basis to let my eyes "cool down".

When I'm writing code, however, green or orange text on a black
background is wonderful. I can sit in front of an editor all day long
in these colors and only take breaks to stretch my arms. It might be
more difficult for me to read the text (I honestly haven't noticed a
difference) but the overall amount of light hitting my eyes is much
lower and I can work more hours.

Back in the late 80s, I remember reading studies showing that
medium-orange on black was one of the best combinations in terms of
readability and eye strain. No idea who did the study or if it was ever
backed up with further research.

--
J. Eric "jet" Townsend, CMU Master of Tangible Interaction Design '09

design: www.allartburns.org; hacking: www.flatline.net; HF: KG6ZVQ
PGP: 0xD0D8C2E8 AC9B 0A23 C61A 1B4A 27C5 F799 A681 3C11 D0D8 C2E8

20 Nov 2008 - 12:00pm
Caroline Jarrett
2007

From: J. Eric "jet" Townsend,
<snip - personal preferences>

> Back in the late 80s, I remember reading studies showing that
> medium-orange on black was one of the best combinations in terms of
> readability and eye strain. No idea who did the study or if it was
> ever backed up with further research.

As you, and a few others, have pointed out in this thread, there's a huge
difference between what 'studies show' and what individuals actually prefer.
The great thing is to try to offer flexibility, so that people can set their
personal experiences up in a way that suits them.

But also, to offer a decent original experience. And for most people, that's
probably going to be a large size, familiar font that offers a good contrast
between foreground and background.

My personal recent experience of light fonts on a dark background is that
many web sites in that combination have fonts that are too small. Is it a
'designer' thing? Who knows.

A couple of further points (forgive me if you've heard this from me before,
it's a constant rant):
- it's hard to trust legibility research. The studies often over-simplify,
use poorly designed materials, fail to understand the interrelationship of
different factors in the examples, used technologies that are now years out
of date, and have lots of other defects. I've been heard to say 'all
legibility research is useless' but I probably exaggerated (a little).
For a longer version of this rant, see my essay available from:
http://www.upassoc.org/upa_publications/jus/2007november/jarrett.html

- it's particularly challenging for users if you interleave white-on-dark
with light-on-dark. I've repeatedly seen users fail to absorb information
that's placed in white-on-dark headers between chunks of light-on-dark body
text, and effect that is especially strong in forms.

Best
Caroline Jarrett

Out now: "Forms that work: Designing web forms for usability"
http://www.amazon.co.uk/Forms-that-Work-Interactive-Technologies/dp/15586071
02

20 Nov 2008 - 12:08pm
Adrian Howard
2005

On 20 Nov 2008, at 17:00, Caroline Jarrett wrote:
[snip]
> As you, and a few others, have pointed out in this thread, there's a
> huge
> difference between what 'studies show' and what individuals actually
> prefer.
> The great thing is to try to offer flexibility, so that people can
> set their
> personal experiences up in a way that suits them.

That may depend on your goals. I recall reading a study (and Google is
failing me here - so I hope somebody can back me up here - it was
about 10/15 years ago) where they monitored completion times for a
task along with user satisfaction. Allowing folk to change the colour
scheme made user satisfaction improve - but made task completion time
slower. This was a desktop app - not a web app.

Happy/happier users vs efficient users.

Cheers,

Adrian

>

20 Nov 2008 - 1:37pm
Caroline Jarrett
2007

From: Adrian Howard

> That may depend on your goals. I recall reading a study (and Google is
> failing me here - so I hope somebody can back me up here - it was
> about 10/15 years ago) where they monitored completion times for a
> task along with user satisfaction. Allowing folk to change the colour
> scheme made user satisfaction improve - but made task completion time
> slower. This was a desktop app - not a web app.
>
> Happy/happier users vs efficient users.

Hmm. I expect that the study did show that, but surely it was testing
one-off approach-and-use tasks rather than longer term usage?

Just thinking it through: if you allow users to include an extra step
("Adjust text so that it is comfortable") then I'd definitely expect task
time to increase just because of the extra step. But I'd also expect users
to prefer the comfortable display.

This time penalty reduces for each subsequent task (providing the display
corrections persist - I can imagine users getting rather grumpy if they had
to tweak the display for every darn task).

In real life, I suspect that users probably won't take the time to adjust
the text etc for a one-off use of a web site or application. They are more
likely to live with it for a short experience - particularly if it's only
slightly unpleasant or mildly uncomfortable. If it's really horrible for
them, or they're planning on longer term use, then it's worth the effort to
do the adjustment.

Thus, I repeat my recommendation that we should be offering a decent default
setup.

Best
Caroline Jarrett

Out now: "Forms that work: Designing web forms for usability"
http://www.amazon.co.uk/Forms-that-Work-Interactive-Technologies/
dp/1558607102
http://www.amazon.com/Forms-that-Work-Interactive-Technologies/dp/product-de
scription/1558607102

20 Nov 2008 - 3:04pm
Jack L. Moffett
2005

On Nov 20, 2008, at 1:37 PM, Caroline Jarrett wrote:

> Just thinking it through: if you allow users to include an extra step
> ("Adjust text so that it is comfortable") then I'd definitely expect
> task
> time to increase just because of the extra step. But I'd also expect
> users
> to prefer the comfortable display.

Of course, you are assuming that the result is a more comfortable
display. Just because people think bright yellow text on a bright blue
background is good doesn't make it so.

Best,
Jack

Jack L. Moffett
Interaction Designer
inmedius
412.459.0310 x219
http://www.inmedius.com

My goal is to build elegant products.
The products that don't make people think
when they should be doing,
make people think
when they should be learning,
compel them by relating to them,
and simply work.

- Josh Viney

20 Nov 2008 - 3:20pm
Caroline Jarrett
2007

I wrote:
> Just thinking it through: if you allow users to include an extra step
("Adjust text so that it is comfortable") then I'd definitely expect task
time to increase just because of the extra step. But I'd also expect users
to prefer the comfortable display.

Jack Moffett replied:

> Of course, you are assuming that the result is a more comfortable display.
Just because people think bright yellow text on a bright blue background is
good doesn't make it so.

Eh? I don't get it. Are you saying that you wouldn't allow users to judge
for themselves whether they consider something to be set up to be
comfortable or not?

Well, I suppose there is a risk. We've probably all come across someone
who's been experimenting with tuning a display and accidentally ended up
with (say) white text on a white background.

But so far, I've never come upon someone who has adjusted a display that to
something that they find uncomfortable and kept it that way. I *have*
observed users who have set their display up in a way that I considered to
be thoroughly unpleasant, but on investigation there's always been a good
reason for it.

Example 1: (I've mentioned this before) A display set up with a clashing
scheme of acid colours. Reason: user was colour-blind and suffered from
migraines. He'd learnt over long experience what mixture worked for him.

Example 2: (Just last week) An LCD display set up with nastily fuzzy text. A
moment's diagnosis showed that the user had opted for a display resolution
of about 75% of the native pixels for the display. Result: larger but (to my
eyes) unpleasantly fuzzy. Reason: user had tried native resolution, and
found it was too small to be comfortable. She'd tried both ways and was
happier with the larger size. (Longer term solution would be to buy her a
bigger monitor, but that's another problem).

This whole sight thing is quite complex. As has been pointed out in this
thread, with some people saying that the (generally recommended) convention
of dark-on-light isn't good for them.

Best
Caroline Jarrett

Out now: "Forms that work: Designing web forms for usability"
http://www.amazon.co.uk/Forms-that-Work-Interactive-Technologies/
dp/1558607102
http://www.amazon.com/Forms-that-Work-Interactive-Technologies/dp/product-de
scription/1558607102

20 Nov 2008 - 4:35pm
Jack L. Moffett
2005

On Nov 20, 2008, at 3:20 PM, Caroline Jarrett wrote:

> Eh? I don't get it. Are you saying that you wouldn't allow users to
> judge
> for themselves whether they consider something to be set up to be
> comfortable or not?

No, I was simply pointing out that task time might increase because
the user unwittingly ended up with a customized display that, despite
their claim to prefer it, may be more difficult to view.

Best,
Jack

Jack L. Moffett
Interaction Designer
inmedius
412.459.0310 x219
http://www.inmedius.com

Good designers relentlessly generate lots of ideas
and open-mindedly consider alternative solutions.
At no time are good designers frightened to entertain
a crazy, competing, or uncomfortable idea.

- Karl Ulrich

19 Nov 2008 - 9:14pm
Tomas Garcia Ferrari
2008

Back in 2002 we have conducted a test on Legibility and Readability on
the screen. One of the axis analyzed was color.

With the limitations of our research, the differences when we were
considering objective results where minimum and get clearer when we
ask subjective opinions.

The results are published here:

Legibility and readability on the World Wide Web

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=35797

20 Nov 2008 - 12:32pm
L.A. King
2008

The one problem I see with white text on a black background vs. black
text on a white background is switching from a page that uses one to
a page that uses the other. When this happens, it is really difficult
for my eyes to adjust to the change.

Since most pages on the net are black text on a white background it
is probably a good idea to make all of your continuous text match
this configuration. Otherwise, it will be difficult for your user's
eyes to make the adjustment.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=35797

Syndicate content Get the feed