Screen resolutions vs. web site legibility

[Chris Morris]

Re: Screen resolutions vs. web site legibility

Neal <neal413@yahoo. com> writes:[color=blue]
> On 28 Aug 2004 14:53:23 +0100, Chris Morris <c.i.morris@dur ham.ac.uk>
> wrote:[color=green]
> > That I don't know. But to object to font-size: 99% would seem
> > ridiculous. (It would be almost equally pointless to set it, true). So
> > there must be a safe region. Perhaps it's only +/- 5% each way,
> > perhaps it's +/- 10%, perhaps it's +/- 20%.[/color]
>
> Perhaps. And perhaps we make our lives much easier by setting it as
> the default and leaving it.[/color]

Actually I'd make my life much easier by setting it to 10pt, since
that tends to be what designers want and it'd keep them quiet (and
they're here and the users aren't) - but I don't want to do that for
the obvious reason.
[color=blue][color=green]
> > Hmm. I would say that the equivalences were:
> > Set colour without background <=> pt/px font size (very risky)
> > Set colour and background <=> % font size other than 100% (less
> > risky [1])
> > Set no colour or background <=> 100% font size (never risky)
> >
> > [1] With the risk greater with some settings than others.[/color]
>
> Don't agree with your matchups. For one, if a user stylesheet changes
> your colors, it does not break the graphic layout as changing font
> size could. So setting colors and backgrounds would be equivalent to
> 100%.[/color]

color: shade-of-red;
background: shade-of-green;

Don't tell me that's not more harmful than font-size: 90%; - the worst
that can do is stop you reading what the text says - that combination
could make the text invisible for some people.

And if a design breaks (as opposed to changes) under UA font-size
changes at least as large as IE's scale provides for, then something's
wrong with the design.

If a user stylesheet sets div {position: static; float: none;} it will
break the design, far more than most font-size changes would. But that
doesn't mean that those should be used less than font-size.

I have a feeling I've misunderstood what you meant by that paragraph.
[color=blue]
> Let me ask you this: if you HAD to use 100% for all body text, really,
> what is lost? And what is gained as a result?[/color]

Nothing, in many senses. What is lost if you don't use colour and
background?

--
Chris

[Brian]

Re: Screen resolutions vs. web site legibility

Chris Morris wrote:
[color=blue]
> color: shade-of-red;
> background: shade-of-green;
>
> Don't tell me that's not more harmful than font-size: 90%;[/color]

Here's how I'm reading this: for the sake of consistency, if we argue
that font-size < 100% reduces readability, we must also argue against
*any* use of color because one possible combination is red on green. But
suggesting colors by itself does not reduce readability (color: #020,
background: #eee; color: #000; background: #ddd; color: 900; background:
#fff; etc.).

If someone asked, "how do I change the text on my page to bright red on
bright green", I'm quite sure that ciwa* regulars would explain that
such a scheme is a bad idea and would reduce readability, much like we
argue that font-size: 85% is bad for readability.

--
Brian (remove ".invalid" to email me)

http://www.tsmchughs.com/

[Neal]

Re: Screen resolutions vs. web site legibility

On Sun, 29 Aug 2004 23:00:44 -0400, Brian
<usenet3@juliet remblay.com.inv alid> wrote:
[color=blue]
> Chris Morris wrote:
>[color=green]
>> color: shade-of-red;
>> background: shade-of-green;
>>
>> Don't tell me that's not more harmful than font-size: 90%;[/color]
>
> But
> suggesting colors by itself does not reduce readability (color: #020,
> background: #eee; color: #000; background: #ddd; color: 900; background:
> #fff; etc.).[/color]

In fact, here's a good tool to test your colors.

http://www.juicystudio.com/services/colourcontrast.asp

My math may be wrong, but I estimate about 1/3 of possible color
combinations are OK according to the adopted standard. You'll also note
that no provision can be made for monitors which display color differently
than what the standard expects.

However, it's a necessity of usability and accessibility to accept that
you cannot directly control the actual size of rendered text without
causing problems for the user. Therefore, we must assume a different way
of thinking than we did with colors. We must assume there is no standard,
because there cannot be, and we have no way of knowing whether one or
another chosen font size can be better than another because we cannot tell
all the information about the user we need to make that determination.
It's as if each monitor in the world displayed colors significantly
differently. We could have no control over colors. All the above standards
would be impossible to set.

We can also assume the browser's default or reset font size is too small
to read, the right size, or a little larger than what's preferred for any
individual user, _no_matter_what _ we_do_. So the best solution is for all
of us to agree on one font size to deliver text, in a way a user can
adjust. In other words, we set a different standard.

100% is by definition the set size in the user's UA. It seems the best one
to choose, as it's the simplest. So we adopt this as the standard, as it
is exactly what the UA is set up to do, no more, no less.

I do see your point. The power running my computer is 110V. If the
electric company wanted to change this standard, it could fry my computer.
Similarly, if everyone was already using 80%, that would be the accepted
standard. But I don't see evidence that most authors have font sizes set
at 80%. Lacking that evidence, it's more reasonable to assume that 100% is
a better choice.

[John C. Ring, Jr.]

Re: Screen resolutions vs. web site legibility

In article <874qmo7br1.fsf @dinopsis.dur.a c.uk>, Chris Morris <c.i.morris@dur ham.ac.uk> wrote:[color=blue]
>Brian <usenet3@juliet remblay.com.inv alid> writes:[color=green]
>> You keep arguing that smaller is better, but other than "everyone else
>> does it", you've presented no reason why.[/color]
>
>That is, as the Neilsen article points out, sometimes a good
>reason. Why does it apply to link colour and underlining, making the
>conventional but sub-optimal underlined blue a very good idea (links
>want to be links, and other references) but doesn't apply to small
>font sizes?[/color]

Because when a user is confused on what is/is not a link, that very clearly
impacts a site's usability is a rather large and adverse way.

On the other hand, granting your point that 100% is actually larger then what
most users normally see, it does not seem to me that the font being slightly
larger then "normal" will have a large, adverse effect on *usability* rather
then aesthetics.

[Steven]

Re: Screen resolutions vs. web site legibility

Kris wrote:<br>[color=blue]
> In article <pan.2004.08.26 .21.51.14.67000 @spam.matt.blis sett.me.uk>,
> Matt <nospam-1@spam.matt.bli ssett.me.uk> wrote:
>[color=green][color=darkred]
>>>Except that font-size: 90% is just as resizable as font-size: 100%. Just
>>>in case I wasn't clear earlier, I am *not* in any way suggesting px/pt is
>>>a good unit for WWW font-sizing.[/color]
>>
>>Why 90%? Why not 95%? Or 103%? If the user has chosen an optimum font
>>size, then that is chosen for 100%. These users have chosen to adjust
>>their font size for a reason: to make it easier to read web sites. By
>>changing the size, you are making it harder to read. Do you really want to
>>make it harder for users to read your site?[/color]
>
> There is more to legibility than font-size. Don't discount line-height,
> contrast between fore- and background, perceived width of text and
> surrounding whitespace.
>
> Lowering font-size is often perfectly justified when complemented with
> attention to above properties.[/color]

Let me show you an example:

http://www.sifre.demon.nl/temp/legibility.html

The example shows how I see this page with and without stylesheet using
my own browser text size preference. I use plain html text as a
reference to set my text size preference.

In the example I see very small text in the styled page combined with
very low contrast (light blue on dark blue) and no extra linespacing or
extra surrounding whitespace. If you scroll down you see the same page
without your stylesheet. This shows the text at my preferred size.

Do you think lowering font-size is "perfectly justified" in this case?

Steven