About font sizes

[dingbat@codesmiths.com]

Re: About font sizes


Darin McGrew wrote:
[color=blue]
> Does it really matter whether my default font size is 13px and your heading
> (with a font-size of 130%) is 17px, or my default font size is 16px and
> your heading is 22px?
>
> If so, then HTML+CSS is the wrong medium for your content.[/color]

If it does matter (and sometimes, legitimately, then it might matter)
HTML+CSS is still an appropriate medium. The CSS protocol allows the
_choice_ of sizing behaviour, it's the mis-use of CSS dimension units
that has caused the problem. If it doesn't matter, then use ems. If it
does matter then use pixels. Everyone is happy -- or at least they are
until some dezyner insists their presentation is more important than
usable access to their content.


....And of course, Firefox has broken the whole model, from the best of
pragmatic intentions.

[Matt Silberstein]

Re: About font sizes

What would a dingbat know about character fonts anyway?

;-)



--
Matt Silberstein

Do something today about the Darfur Genocide

http://www.beawitness.org

http://www.darfurgenocide.org

http://www.savedarfur.org

"Darfur: A Genocide We can Stop"

[Harlan Messinger]

Re: About font sizes

Jukka K. Korpela wrote:[color=blue]
> David Dorward <dorward@yahoo. com> scripsit:
>[color=green]
>> Systems can round font sizes to something acceptable, so there's not
>> really any point in worrying about it.[/color]
>
> Unfortunately, there is.
>
> Typography rules say, among other things, that a change in font size
> should be clear enough so that it does not look like an error. The
> mileage varies, but let us assume that a 10% increase is desired. So if
> we say font-size: 110%, what will happen? The browser computers the
> absolute font size and then selects the closest actual font size it can
> use - at the simplest, selecting among differently sized fonts for the
> given font family (typeface). For all we can know, that size could be
> identical to the basic font size (when only a few sizes exist). Or it
> might be almost 20% larger than the basic font size.
>
> For example, if the basic font size is 12pt, then 110% yields 13.2pt,
> which I would expect to get rounded to 13pt. However, for Times New
> Roman, I seem get a bigger font (13.5pt?), which is rather big and looks
> bolded.[/color]

Since the quantum of the screen is the pixel, isn't it a question of
whether the number of pixels that you think looks bigger than 13.2pt is
a better fit than one pixel less would be, which might be more like 12.75pt?
[color=blue]
> If I set the font size to a smaller percentage, there's the risk
> of getting no font size increase, for some combinations of font face and
> basic size.
>[/color]

[Darin McGrew]

Re: About font sizes

I wrote:[color=blue][color=green]
>> Does it really matter whether my default font size is 13px and your heading
>> (with a font-size of 130%) is 17px, or my default font size is 16px and
>> your heading is 22px?
>>
>> If so, then HTML+CSS is the wrong medium for your content.[/color][/color]

<dingbat@codesm iths.com> wrote:[color=blue]
> If it does matter (and sometimes, legitimately, then it might matter)
> HTML+CSS is still an appropriate medium. The CSS protocol allows the
> _choice_ of sizing behaviour, it's the mis-use of CSS dimension units
> that has caused the problem. If it doesn't matter, then use ems. If it
> does matter then use pixels.[/color]

But CSS is optional, by design. If the size in pixels (or points, or
inches, or...) really matters, then the distinction will be lost when the
author's CSS is disabled/ignored. Or in media where the concept of pixels
(or points, or inches, or...) is irrelevant.

If the exact font size matters, then HTML+CSS is still the wrong medium.
--
Darin McGrew, mcgrew@stanford alumni.org, http://www.rahul.net/mcgrew/
Web Design Group, darin@htmlhelp. com, http://www.HTMLHelp.com/

"Cheaters never win; they just finish first." - Johhny Hart

[deko]

Re: About font sizes

>>> Does it really matter whether my default font size is 13px and your heading[color=blue][color=green][color=darkred]
>>> (with a font-size of 130%) is 17px, or my default font size is 16px and
>>> your heading is 22px?
>>>
>>> If so, then HTML+CSS is the wrong medium for your content.[/color][/color]
>
> <dingbat@codesm iths.com> wrote:[color=green]
>> If it does matter (and sometimes, legitimately, then it might matter)
>> HTML+CSS is still an appropriate medium. The CSS protocol allows the
>> _choice_ of sizing behaviour, it's the mis-use of CSS dimension units
>> that has caused the problem. If it doesn't matter, then use ems. If it
>> does matter then use pixels.[/color]
>
> But CSS is optional, by design. If the size in pixels (or points, or
> inches, or...) really matters, then the distinction will be lost when the
> author's CSS is disabled/ignored. Or in media where the concept of pixels
> (or points, or inches, or...) is irrelevant.
>
> If the exact font size matters, then HTML+CSS is still the wrong medium.[/color]

I don't think format is relevant to this discussion (which I've been enjoying,
by the way).

I see only two issues here. One is about usability. And this is the far more
important issue - users should be able to adjust the text size in a page
(regardless of format, pixels, percent, etc). With modern browsers/viewers,
they can, and I think we all agree on this.

That's pretty much the end of it for me, but here's the other issue:

The second issue is one of accommodating user's preferences. That is, some
users (a small minority) want to be able to scale the page author's proposed
font sizes automatically, according to their predefined browser settings. So,
when a page is rendered, 100% is mapped to 80%, 30% is mapped to 60%, and so on.
This has nothing to do with usability. It's pure preference. The problem for
users who set their browsers to do this - to change the way a page has been
authored to meet their preferences - is that current browser technology cannot
easily map px-designated font sizes to their preferences (while percentage-based
font sizes are more readily adjustable).

So the px vs. percentage debate is a tempest in a teapot. It makes absolutely
no difference in regard to usability, and very little difference for those very
few who want to alter a page author's design.

[Darin McGrew]

Re: About font sizes

deko <deko@nospam.co m> wrote:[color=blue]
> So the px vs. percentage debate is a tempest in a teapot. It makes
> absolutely no difference in regard to usability, and very little
> difference for those very few who want to alter a page author's design.[/color]

By definition, px font sizes do not adapt to the user's default font size.
Sure, readers can override px font sizes (e.g., with a minimum font size
setting or by ignoring document font sizes).

By definition, percentage font sizes (or em font sizes, but in practice you
need to work around a bug in MSIE) do adapt to the user's default font
size. Sure, web deezyners can still specify font sizes smaller than 100%
(1em), and readers can still override such microfonts. But at a fundamental
level, percentage (em) font sizes adapt to the user's default font size in
a way that px font sizes don't.

Maybe it's time to bring this quote up again:

The font size chosen by the user as a comfortable default (1 em)
[or 100%] provides more truly useful information about the
rendering environment than all the resolution-sniffing,
window-querying, "open-this-wide" logic you can throw at the
problem. -- Todd Fahrner
--
Darin McGrew, mcgrew@stanford alumni.org, http://www.rahul.net/mcgrew/
Web Design Group, darin@htmlhelp. com, http://www.HTMLHelp.com/

"Experience is what allows you to recognize a mistake when you make it again."

[Jukka K. Korpela]

Re: About font sizes

Harlan Messinger <hmessinger.rem ovethis@comcast .net> scripsit:
[color=blue]
> Since the quantum of the screen is the pixel, isn't it a question of
> whether the number of pixels that you think looks bigger than 13.2pt
> is a better fit than one pixel less would be, which might be more like
> 12.75pt?[/color]

Not really. It is possible that the available font sizes, which are selected
by the browser by rounding the computed size to the closest of them, will be
further "rounded" to pixels, i.e. the glyphs will be pixelized. This is,
however, another aspect that does not change the primary issue that the
repertoire of actual font sizes is, or could be, discrete and even
relatively small. Moreover, on paper, different considerations apply. Even
on screen, smoothing may imply that in practice, the pixel does not act as a
simple quantum.

--
Jukka K. Korpela ("Yucca")

http://www.cs.tut.fi/~jkorpela/

[dingbat@codesmiths.com]

Re: About font sizes


Matt Silberstein wrote:[color=blue]
> What would a dingbat know about character fonts anyway?[/color]

Oddly that's _not_ where the username came from. It began as a
mainframe user home directory called DING$BAT, about 20 years ago.

[Alan J. Flavell]

Re: About font sizes

On Fri, 16 Jun 2006, Jukka K. Korpela wrote:
[color=blue]
> however, another aspect that does not change the primary issue that
> the repertoire of actual font sizes is, or could be, discrete and
> even relatively small.[/color]

Oh, indeed: on Lynx the repertoire of font sizes is one (and on a
speaking browser, one could say that it is 'none'); but that doesn't
discourage me from proposing sizes that are intended for the more
mainstream browsing situation.

CSS is optional, by design. If, in a small number of cases, a pair of
proposed em or percent sizes results in the same display size, it's
unfortunate, but it's hardly a tragedy, and I wouldn't want to go
against other good principles merely to avoid this occasional
possibility.
[color=blue]
> Moreover, on paper, different considerations apply. Even on screen,
> smoothing may imply that in practice, the pixel does not act as a
> simple quantum.[/color]

Right. But even when designing a stylesheet for print-only, there's
still the paper size question: you don't know whether they use A4 or
US Letter (or maybe something else again). If it's about getting a
paper form printed, that for some reason *has* to be an exact size,
then PDF is still a better medium for doing that, IMHO.

I value HTML+CSS for its flexibility and versatility, on the other
hand.

regards

[deko]

Re: About font sizes

> But at a fundamental level, percentage (em) font sizes adapt to[color=blue]
> the user's default font size in a way that px font sizes don't.[/color]

Absolutely. But for over 90% of users (those who don't change their browser
settings), the default font size is 16px (the browser's default), which is
unattractively large. Most page authors will adjust this down a bit, say to
14px, in an effort to make their pages more attractive. An author can specify
either 87.5% or 14px to make that adjustment. Either way he can expect his
design to be rendered properly for those 90% of users.

For those few users who *do* change their browser's default font size, we have a
problem.

If they set the default size to 14px, then the author's 87.5% becomes 12.25px,
which is a bogus size. There is no 12.25px font size. We can hope it will be
rounded, but we really don't know what we're going to get. This results in a
"crap shoot" when it comes to rendering the author's design. An author who uses
percentage-based font sizes, therefore, is effectively making his design
meaningless for users who adjust their browser's default font size. For those
who don't change their default font size, it makes no difference how the author
designated his font sizes. Nevertheless, everyone still has the ability to
*manually* adjust the font size (the browser's override feature).

The downside of px-based font size designations is that the author's design will
not *automatically* adjust to user-defined preferences. As I mentioned earlier,
this is a preference issue, not a usability issue.

The root of the problem is that browsers' default settings are specified in px -
and percentages cannot accurately be mapped to px. If browsers specified more
intuitive sizes - like small, medium, large, etc. - there would not be a problem
(other than coming to an agreement on what "medium" is). Authors would then
have assurance that their designs would scale proportionately to accommodate
user preference.

[Darin McGrew]

Re: About font sizes

I wrote:[color=blue][color=green]
>> But at a fundamental level, percentage (em) font sizes adapt to
>> the user's default font size in a way that px font sizes don't.[/color][/color]

deko <deko@nospam.co m> wrote:[color=blue]
> Absolutely. But for over 90% of users (those who don't change their browser
> settings), the default font size is 16px (the browser's default), which is
> unattractively large. Most page authors will adjust this down a bit, say to
> 14px, in an effort to make their pages more attractive.[/color]

IMHO, it's a misguided effort. See also

http://www.xs4all.nl/~sbpoley/webmatters/fontsize.html

[color=blue]
> An author can specify either 87.5% or 14px to make that adjustment.[/color]

There's a quote in my .sig file that seems appropriate:
"There is no right way to do the wrong thing."
[color=blue]
> If they set the default size to 14px, then the author's 87.5% becomes
> 12.25px, which is a bogus size. There is no 12.25px font size. We can
> hope it will be rounded, but we really don't know what we're going
> to get. This results in a "crap shoot" when it comes to rendering
> the author's design. An author who uses percentage-based font sizes,
> therefore, is effectively making his design meaningless for users who
> adjust their browser's default font size.[/color]

IMHO, a font size smaller than 100% is appropriate only for legalese and
similar fine print that the average reader can safely ignore. Once you make
the mistake of setting the font size for body text at 87.5%, all the issues
you're so concerned about become relatively minor.
[color=blue]
> Nevertheless, everyone still has the ability to *manually* adjust the
> font size (the browser's override feature).[/color]

Wouldn't it be better to adapt to the browser configuration, so the user
isn't expected to override anything?
[color=blue]
> The downside of px-based font size designations is that the author's
> design will not *automatically* adjust to user-defined preferences.
> As I mentioned earlier, this is a preference issue, not a usability
> issue.[/color]

Or maybe it's an accessibility issue. I find microfonts to be a significant
usability problem, and I've got better than 20/20 vision. That's why I
enforce a minimum font size. But you're free to write that off as merely my
"preference " if you like.
[color=blue]
> If browsers specified more intuitive sizes - like small, medium, large,
> etc. - there would not be a problem (other than coming to an agreement
> on what "medium" is). Authors would then have assurance that their
> designs would scale proportionately to accommodate user preference.[/color]

A very common browser-like OS component (you may have heard of it) allows
the user to specify font sizes only as Smallest, Smaller, Medium, Larger,
and Largest. CSS provides the font sizes xx-small, x-small, small, medium,
large, x-large, and xx-large, as well as larger and smaller.

How has that assured authors of anything? Web sites with microfonts are
still common enough that web browsers provide minimum font size settings.
--
Darin McGrew, mcgrew@stanford alumni.org, http://www.rahul.net/mcgrew/
Web Design Group, darin@htmlhelp. com, http://www.HTMLHelp.com/

"Shin: a device for finding furniture in the dark." - Steven Wright

[Alan J. Flavell]

Re: About font sizes

On Fri, 16 Jun 2006, deko wrote:
[color=blue]
> Absolutely. But for over 90% of users[/color]

You've actually been out there and looked over their shoulders and
counted them?
[color=blue]
> (those who don't change their browser settings), the default font
> size is 16px (the browser's default), which is unattractively large.[/color]

*You* might think so. But if *they* thought so, don't you suppose
they would start asking why, and what they could do about it?
[color=blue]
> Most page authors will adjust this down a bit,[/color]

So now we're into a deadly spiral. Authors crank the fonts down to
smaller than the user wants, so users crank it up to more than the
author wants. The authors then will respond by cranking it down even
smaller, and the users will want to compensate even further.

So why not get out of this spiral before it's even started. Design
your pages so that they can *also* cope with readers who display text
larger than you would prefer it to be.
[color=blue]
> Either way he can expect his design to be rendered properly for
> those 90% of users.[/color]

"For some value of properly".

There *is* an interworking specification, after all.
[color=blue]
> For those few users who *do* change their browser's default font
> size, we have a problem.[/color]

You have a problem for *all* users, excepting only those who *wanted*
your choice of font size. You have no idea who those are, nor even
how many.
[color=blue]
> If they set the default size to 14px, then the author's 87.5% becomes 12.25px,
> which is a bogus size. There is no 12.25px font size. We can hope it will be
> rounded, but we really don't know what we're going to get.[/color]

Again you're arguing around in circles again.

With 1.0em or 100%, we *know* that users will get the text size that
they selected - which in many cases will be the size chosen by their
chosen OS vendor. Are you *SO* certain that you know so much better
what they could possibly want, than their vendor knows?
[color=blue]
> This results in a "crap shoot" when it comes to rendering the
> author's design.[/color]

It does, if the author is incapable of designing flexibly.
[color=blue]
> An author who uses percentage-based font sizes, therefore, is
> effectively making his design meaningless for users who adjust their
> browser's default font size.[/color]

That's drivel.

I've had enough of this. You're just trolling on and on, round and
round the same imponderables, and getting nowhere. The reason is that
you're focussing on fixing factors which the WWW simply refuses to
allow you to fix, and apparently ignoring the solution which has been
staring you in the face all the time.

And then you try to comfort yourself with the thought that you'd only
be excluding your invented figure of 10% of users. That's pathetic.

Bye.

[Harlan Messinger]

Re: About font sizes

deko wrote:[color=blue][color=green]
>> But at a fundamental level, percentage (em) font sizes adapt to
>> the user's default font size in a way that px font sizes don't.[/color]
>
> Absolutely. But for over 90% of users (those who don't change their
> browser settings), the default font size is 16px (the browser's
> default), which is unattractively large. Most page authors will adjust
> this down a bit, say to 14px, in an effort to make their pages more
> attractive. An author can specify either 87.5% or 14px to make that
> adjustment. Either way he can expect his design to be rendered properly
> for those 90% of users.
>
> For those few users who *do* change their browser's default font size,
> we have a problem.[/color]

On the one hand, all these people, as they browse across the Internet
day after day after day, are comfortable enough with the default font
size not to bother taking five seconds to adjust it on their browser. On
the other hand, when they come to your web site, the same size is so
unattractively large that they rely on your to save them from the
setting that they haven't bothered to change on their own. Seems like a
fallacy to me.
[color=blue]
>
> If they set the default size to 14px, then the author's 87.5% becomes
> 12.25px, which is a bogus size. There is no 12.25px font size.[/color]
[color=blue]
> We can
> hope it will be rounded, but we really don't know what we're going to
> get. This results in a "crap shoot" when it comes to rendering the
> author's design.[/color]

Exactly! It *is* a crap shoot! It isn't desktop publishing, where you
have perfect control over the location and size of every glyph, line,
and image. Rather than remaining in denial over that fact, the best way
to design for the web is to treat it as what it is rather than as what
it isn't.

[deko]

Re: About font sizes

Now that Mr. Flavell has bowed out of the debate, I will declare check and mate
on this argument and conclusively declare px the preferred font size designator
for web development. The ax has already been laid at the root. Now, to the
chagrin of some, the tree will fall.

The pc (em or percentage-size) crowd scoffs at px (pixel-size) as if it were a
fascist plot that imposes the will of the page author upon helpless users. The
reality is that a benevolent designer will not mistreat his users, but rather
will provide a good design. For who do not like the good of the author, or
require special care, browser technology is at their disposal to adjust the
author's design to suit their own preferences.

Now, for those who insist that their preferences are more important than the
designer's (this is a fallacy; there would be no art if this were true),
technology is available in the form of a user stylesheet, which will ignore the
good of the designer and let each man do what is right in his own eyes.

Whatever the medium, format, or content, the designer has a responsibility to
provide a good design, not simply throw text at his users. px ensures the good
of the designer is not corrupted in translation; pc pretends there is no good.
px is the currency of good; pc is the currency of anarchy.

In the end, the pc vs. px debate is about preferences, or perhaps philosophy.
If we are not arguing about the color of the rug, then we are arguing about the
good. And designers, if the word means anything, are architects of good.

[Beauregard T. Shagnasty]

Re: About font sizes

deko wrote:
[color=blue]
> Now that Mr. Flavell has bowed out of the debate, I will declare check
> and mate on this argument and conclusively declare px the preferred
> font size designator for web development. The ax has already been
> laid at the root. Now, to the chagrin of some, the tree will fall.[/color]

No, you still lost. Days ago.

--
-bts
-Warning: I brake for lawn deer