I've run CSSCheck on my style sheets and I always get a warning
similar to this:

"font: bold 9pt/100% sans-serif

Warning: Absolute length units should not generally be used on the Web ..."

Yet if I use 'x-small' instead of 9pt, I get bigger type on IE6 and smaller
type on
Mozilla. My choices seem to be:

1. keep the absolute 9pt setting in order to maintain a much more consistent
display between browsers (and break the spirit of the stylesheets)

2. use 'x-small' and live with the different display results (yet be
compatible with
user defined stylesheets, and other output media).

Is that it?

"J" <sorrynospam@nobody.com> wrote in message
news:400d5ab6$0$43853$39cecf19@news.twtelecom.net. ..
...
| 2. use 'x-small' and live with the different display results
(yet be
| compatible with
| user defined stylesheets, and other output media).

It is not you, but your users who have
to put up with it. They will set fonts
to a size of their own choosing if you
do not stuff them up by specifying
fonts in pixels.

For the sake of useability, you should
never specify fonts in pixels..

--
Andrew Thompson
* http://www.PhySci.org/ PhySci software suite
* http://www.1point1C.org/ 1.1C - Superluminal!
* http://www.AThompson.info/andrew/ personal site

"J" <sorrynospam@nobody.com> wrote in message
news:400d5ab6$0$43853$39cecf19@news.twtelecom.net. ..[color=blue]
> I've run CSSCheck on my style sheets and I always get a warning
> similar to this:
>
> "font: bold 9pt/100% sans-serif
>
> Warning: Absolute length units should not generally be used on the Web[/color]
...."[color=blue]
>
> Yet if I use 'x-small' instead of 9pt, I get bigger type on IE6 and[/color]
smaller[color=blue]
> type on
> Mozilla. My choices seem to be:
>
> 1. keep the absolute 9pt setting in order to maintain a much more[/color]
consistent[color=blue]
> display between browsers (and break the spirit of the stylesheets)[/color]

Theoretically, it's not your job to assure absolute consistency among the
browsers. It's up to the browsers to decide what base size to use for text
they display. If a user doesn't like his browser's default, then he can use
the browser's mechanism for increasing or decreasing the text size. If he
prefers the way pages look in a different browser, then he'll use a
different browser. Life is just too short, unless the demand for consistency
comes from your customer and you're getting paid by the hour.

Having said that, I think it's reasonable to specify a font size from within
a range. Larger text is more open, more friendly; smaller text is more
formal, more serious, and makes better use of the available space. Moreover,
if I set

body { font-size: 100%; sans-serif; }

then the text seems ridiculously large to me in ALL the graphical browsers.
So if my general body text is to be sans-serif, then I'm inclined to use
font-size of 80%-90%. For serif, 90%-100%. Hmm, using "small" instead of
percentages, I seem to get decent results with both serif and
sans-serif--and I find x-small much too small in all browsers.
[color=blue]
>
> 2. use 'x-small' and live with the different display results (yet be
> compatible with
> user defined stylesheets, and other output media).[/color]

In article <TqdPb.20593$Wa.20283@news-server.bigpond.net.au>, andrew64
@bigNOSPAMpond.com enlightened us with...[color=blue]
> "J" <sorrynospam@nobody.com> wrote in message
> news:400d5ab6$0$43853$39cecf19@news.twtelecom.net. ..
> ..
> | 2. use 'x-small' and live with the different display results
> (yet be
> | compatible with
> | user defined stylesheets, and other output media).
>
> It is not you, but your users who have
> to put up with it. They will set fonts
> to a size of their own choosing if you
> do not stuff them up by specifying
> fonts in pixels.
>
> For the sake of useability, you should
> never specify fonts in pixels..
>[/color]

You know, no one told me that (nor did I see it in the beginner's sites
I was using at the time) when I started out with web design (personal
sites).
I found out quite some time later that when I specified font in pixels,
trying to up the size in the browser did nothing.
I immediately changed all my sizes to small, medium, etc.

So, add my opinion to this one - don't use pixels. It's mean to normal
users and downright awful to visually impaired people.

It is not your job to worry about how the browser renders those sizes -
because if you specify them in relative terms like percents or small,
medium, etc, the user can easily change the size if they find it too
small. They can't do that if you use pixels.

--
--
~kaeli~
A man's home is his castle..., in a manor of speaking.
http://www.ipwebdesign.net/wildAtHeart
http://www.ipwebdesign.net/kaelisSpace

J wrote:
[color=blue]
> I've run CSSCheck on my style sheets and I always get a warning
> similar to this:
>
> "font: bold 9pt/100% sans-serif
>[/color]

Unit "pt" is for print, not Web (as opposed to "px"). On the web you
better keep the font-sizes to the default size. Why would you want to
decrease that? It's one if not _the_ biggest accessibility issue on
today's Web and you want to repeat it? If it's something like a
footnote, try "90%" and the like.

J <sorrynospam@nobody.com> wrote:[color=blue]
> I've run CSSCheck on my style sheets and I always get a warning
> similar to this:
>
> "font: bold 9pt/100% sans-serif
>
> Warning: Absolute length units should not generally be used on the Web ..."[/color]

That is correct. See also http://css.nu/faq/ciwas-aFAQ.html#QA02
[color=blue]
> Yet if I use 'x-small' instead of 9pt, I get bigger type on IE6 and smaller
> type on
> Mozilla. My choices seem to be:
>
> 1. keep the absolute 9pt setting in order to maintain a much more consistent
> display between browsers (and break the spirit of the stylesheets)[/color]

Why is consistency an issue? Who (besides you) will view your site with
more than one browser? And of the few who do view your site with more than
one browser, how many will care if the presentation is different?

And FWIW, 9pt is smaller than the minimum font size I've configured my
browser to use.
[color=blue]
> 2. use 'x-small' and live with the different display results (yet be
> compatible with
> user defined stylesheets, and other output media).[/color]

That's going to be a size or two smaller than the browser's normal font
size, which is appropriate only for legalese and similar fine print that
the reader can safely ignore.
[color=blue]
> Is that it?[/color]

You could use percentages to specify fonts relative to the browser's normal
font size. I've read that this generally produces better results than using
the keywords (xx-small through xx-large).

Or you could leave the font size alone, which is a very good idea for
normal body text anyway.
--
Darin McGrew, mcgrew@stanfordalumni.org, http://www.rahul.net/mcgrew/
Web Design Group, darin@htmlhelp.com, http://www.HTMLHelp.com/

"Good teachers are costly. Bad teachers cost more." - Bob Talbert

kaeli wrote:[color=blue]
>
> It is not your job to worry about how the browser renders those
> sizes - because if you specify them in relative terms like percents
> or small, medium, etc, the user can easily change the size if they
> find it too small.[/color]

And if you don't specify any size for body text (body, p, li, etc),
then the user likely won't have to change the size of the text. And if
they do need to change it, their changes will work on every properly
authored site without breaking anything on badly designed ones.
[color=blue]
> They can't do that if you use pixels.[/color]

Sure. So use percent when you specify e.g. headings. But leave the
body text alone, and it's irrelvant.

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Harlan Messinger wrote:[color=blue]
>
> Theoretically, it's not your job to assure absolute consistency
> among the browsers.[/color]

And practically, it's not possible to do that.
[color=blue]
> It's up to the browsers to decide what base size to use for text
> they display. If a user doesn't like his browser's default, then he
> can use the browser's mechanism for increasing or decreasing the
> text size.[/color]

Right.
[color=blue]
> Having said that, I think it's reasonable to specify a font size
> from within a range.[/color]

To suggest sizes for headings, and the like, sure.
[color=blue]
> if I set
>
> body { font-size: 100%; sans-serif; }
>
> then the text seems ridiculously large to me in ALL the graphical
> browsers.[/color]

Funny.

body { font-size: 100%;}

produces text that is *precisely* the right size in my graphical
browser. Oh, right. I /configured/ my browser.
[color=blue]
> So if my general body text is to be sans-serif, then I'm inclined
> to use font-size of 80%-90%. For serif, 90%-100%. Hmm, using
> "small" instead of percentages, I seem to get decent results with
> both serif and sans-serif[/color]

Perhaps you should configure *your* browser instead of solving your
browsing problem from the authoring end.

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Harlan Messinger wrote:[color=blue]
> "J" <sorrynospam@nobody.com> wrote in message
> news:400d5ab6$0$43853$39cecf19@news.twtelecom.net. ..
>[color=green]
>> I've run CSSCheck on my style sheets and I always get a warning
>> similar to this:
>>
>> "font: bold 9pt/100% sans-serif
>>
>> Warning: Absolute length units should not generally be used on the
>> Web[/color]
>[color=green]
>> Yet if I use 'x-small' instead of 9pt, I get bigger type on IE6 and
>> smaller type on Mozilla.[/color][/color]

The CSS spec doesn't define how 'smaller' should be rendered. It's only
defined to be smaller or equal in size to 'small'. Nobody should ever
use 'small', 'smaller' or 'x-small' for body text. See below.
[color=blue]
> Theoretically, it's not your job to assure absolute consistency among
> the browsers. It's up to the browsers to decide what base size to use
> for text they display. If a user doesn't like his browser's default,[/color]

Actually, it's up to the user to decide what the base size for text
should look like. Unfortunately, many users and even some developers
using Microsoft's products are unaware of this fact simply because MSIE
lacks the user interface to set base size. The user is supposed to hint
his 'User Agent' (a.k.a. browser) how he wants to view pages and then
User Agent proceeds to render the page the view user wants. The User
Agent may be a graphical browser like MSIE, Mozilla or Opera, or it may
be a text-only browser like Lynx or Links. Or it could be output driver
of a speech synthesis software or braille display.

It's not your work - as a page author - to specify some pixel or point
size for the text. Use percentages instead: if something should be
emphasized when compared to body text, make it a bit larger (perhaps
110%), if it's less important than body text, make it a bit smaller
(like 90%). Be warned that MSIE (and Opera?) has a bug that causes weird
rendering unless you say also "body { font-size: 100%; }", which is
interesting because that should be the default...

The 100% is the correct setting for the body text. This is for two reasons:

1) If the user is aware that he can set the default body text size (like
me) he has already adjusted it and 100% is THE BEST possible font size
(he selected it, after all, how could you know any better than him??)

2) If the user isn't aware about the possibility, he doesn't care enough
to look for it. Ergo, he's already HAPPY with the font size of 100%. Why
should you then make it smaller??
[color=blue]
> body { font-size: 100%; sans-serif; }
>
> then the text seems ridiculously large to me in ALL the graphical
> browsers. So if my general body text is to be sans-serif, then I'm[/color]

Yes, the DEFAULT SETTINGS of all graphical browsers DEFAULT to pretty
large body text. Why? Perhaps, because if you have to choice between
text being RIDICULOUSLY LARGE or UNREADABLE SMALL, then the former is
lesser of two evil. You cannot select a default settings that is perfect
for everybody, can you? You're supposed to change the settings of YOUR
user agent as you want. I repeat, if you don't change the setting or you
don't even bother for looking it, you don't mind enough and any web page
using the default size will be OK, really.

FYI, my MINIMUM font size is 11px. I'm using high resolution display and
using more pixels per character results to better rendering. Do not try
to destroy my beautifully shaped letters.

[ Note that 100% in the root element is supposed to be equal to
'medium'. However, you shouldn't use 'medium' or 'small' keywords for
any normal text because both of those keywords have different meaning in
different versions of MSIE -- that is, some versions interprete 'small'
as 100% and 'medium' as something like 110%. And some other versions of
the same browser interprete 'medium' as 100% and small as something like
90%. Considering that MSIE is the most used browser one cannot really
use keywords 'small' or 'medium' for anything. ]

--
Mikko

On Tue, 20 Jan 2004 12:32:34 -0500, "Harlan Messinger"
<h.messinger@comcast.net> wrote:
[color=blue]
>Moreover, if I set
>
> body { font-size: 100%; sans-serif; }
>
>then the text seems ridiculously large to me in ALL the graphical browsers.[/color]

Well, if you haven't told the browser what size you like, it might.
Defaults tend to be on the large size for the benefit of people with
less good eyesight.

http://www.xs4all.nl/~sbpoley/main/adjust.html

--
Stephen Poley

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

"Brian" <usenet2@julietremblay.com.invalid-remove-this-part> wrote in
message news:EqgPb.107810$8H.205662@attbi_s03...[color=blue]
> Harlan Messinger wrote:[color=green]
> >
> > Theoretically, it's not your job to assure absolute consistency
> > among the browsers.[/color]
>
> And practically, it's not possible to do that.
>[color=green]
> > It's up to the browsers to decide what base size to use for text
> > they display. If a user doesn't like his browser's default, then he
> > can use the browser's mechanism for increasing or decreasing the
> > text size.[/color]
>
> Right.
>[color=green]
> > Having said that, I think it's reasonable to specify a font size
> > from within a range.[/color]
>
> To suggest sizes for headings, and the like, sure.
>[color=green]
> > if I set
> >
> > body { font-size: 100%; sans-serif; }
> >
> > then the text seems ridiculously large to me in ALL the graphical
> > browsers.[/color]
>
> Funny.
>
> body { font-size: 100%;}
>
> produces text that is *precisely* the right size in my graphical
> browser. Oh, right. I /configured/ my browser.[/color]

Well, good for you. ;-) Consider that, in the US at least, it isn't
surprising to find on a can of franks and beans:

(1) Open can. (2) Empty contents into saucepan. (3) Heat until warm.

Besides, at least with IE, as far as I can tell it won't remember what text
size you set previously, so you have to do it each time. Ick. So, anyway, I
figure plenty of people have their browsers permanently stuck on medium, and
I just think the default actual size of text is simply unwieldy on that
setting. I also have a feeling that people with visual difficulties are more
likely to have actually bothered to figure out how to fix their browsers to
suit them, than are people who might think the text is too large, but for
whom it's only an esthetic concern.
[color=blue]
>[color=green]
> > So if my general body text is to be sans-serif, then I'm inclined
> > to use font-size of 80%-90%. For serif, 90%-100%. Hmm, using
> > "small" instead of percentages, I seem to get decent results with
> > both serif and sans-serif[/color]
>
> Perhaps you should configure *your* browser instead of solving your
> browsing problem from the authoring end.[/color]

Or I can choose not to be dogmatic, and gear things just a little bit for
some of the most common viewing circumstances, as long as the ability of
others to adjust things remains.

"Mikko Rantalainen" <mira@st.jyu.fi> wrote in message
news:buk4n8$bm5$1@mordred.cc.jyu.fi...[color=blue]
> 2) If the user isn't aware about the possibility, he doesn't care enough
> to look for it. Ergo, he's already HAPPY with the font size of 100%.[/color]

Non sequitur.
[color=blue]
> Why
> should you then make it smaller??[/color]

Because my intuition tells me that, all things considered, it's a net
benefit. I could be wrong but it's my sense that I'm right. And that, after
all, is the kind of judgment that professionals in every field and in every
era have had to bring to the table.
[color=blue]
>[color=green]
> > body { font-size: 100%; sans-serif; }
> >
> > then the text seems ridiculously large to me in ALL the graphical
> > browsers. So if my general body text is to be sans-serif, then I'm[/color]
>
> Yes, the DEFAULT SETTINGS of all graphical browsers DEFAULT to pretty
> large body text. Why? Perhaps, because if you have to choice between
> text being RIDICULOUSLY LARGE or UNREADABLE SMALL, then the former is
> lesser of two evil. You cannot select a default settings that is perfect
> for everybody, can you? You're supposed to change the settings of YOUR
> user agent as you want. I repeat, if you don't change the setting or you
> don't even bother for looking it, you don't mind enough[/color]

You think I shouldn't second-guess people's preferences and motivations, but
you seem happy enough to do it, as you again show here.
[color=blue]
> and any web page
> using the default size will be OK, really.
>
> FYI, my MINIMUM font size is 11px. I'm using high resolution display and
> using more pixels per character results to better rendering. Do not try
> to destroy my beautifully shaped letters.
>
> [ Note that 100% in the root element is supposed to be equal to
> 'medium'. However, you shouldn't use 'medium' or 'small' keywords for
> any normal text because both of those keywords have different meaning in
> different versions of MSIE -- that is, some versions interprete 'small'
> as 100% and 'medium' as something like 110%. And some other versions of
> the same browser interprete 'medium' as 100% and small as something like
> 90%. Considering that MSIE is the most used browser one cannot really
> use keywords 'small' or 'medium' for anything. ][/color]

First you tell me that I should defer to the wisdom of the browser
manufacturers concerning the proper default text size because they
determined that too large is better than too small, and they know best. But
then the way they handle "small" and "medium" is dumb and we shouldn't trust
their judgment at all. I'm finding it very confusing being told I'm wrong
about one thing because the browser manufacturers are smart, but I'm wrong
about another thing because the browser manufacturers are dumb. Consistency
of argument is a virtue!

in post: <news:buk9ou$ihn39$1@ID-114100.news.uni-berlin.de>
"Harlan Messinger" <h.messinger@comcast.net> said:
[color=blue]
> Besides, at least with IE, as far as I can tell it won't remember what text
> size you set previously,[/color]

it does


--
brucie - i usenet nude

*Philipp Lenssen*:[color=blue]
>
> Unit "pt" is for print, not Web (as opposed to "px").[/color]

The Web can be printed as well, just not all at once.

I assume you meant "screen" instead of "Web".

--
"The scientific name for an animal that doesn't
either run from or fight its enemies is lunch."
Michael Friedman

Harlan Messinger / 2004-01-21 00:33:[color=blue]
> "Mikko Rantalainen" <mira@st.jyu.fi> wrote in message
> news:buk4n8$bm5$1@mordred.cc.jyu.fi...
>[color=green]
>>2) If the user isn't aware about the possibility, he doesn't care enough
>>to look for it. Ergo, he's already HAPPY with the font size of 100%.[/color]
>
> Non sequitur.[/color]

You did read the rest of my post, right? I explained, that if the
user really were unhappy with the font size of 100%, he would do
something. After that user has fixed the problem, font size of 100%
would be the best possible font size for that user. You cannot know
if an user has already adjusted his browser as he likes. If he has
already adjusted his browser and 100% is the perfect font size for
him, wouldn't it be a little dumb to readjust the font size to some
random smaller size? And if he hasn't adjusted the font size, does
that mean that he /really/ wants to read smaller text, but doesn't
have enough knowledge to fix the issue, or is it that the font size
he's seeing is already the best one for him?

The bottom line is that for advanced users that CAN configure their
browsers correctly, 100% is THE BEST possible font size for the body
text. And for the rest of the users, you cannot know which font size
would be the best.

If you know that 100% is THE BEST possible font size for some people
and the other people really wanted to use some smaller or bigger
font but don't know how to fix the problem, HOW do you know the
correct setting?

[color=blue][color=green]
>>Why should you then make it smaller??[/color]
>
> Because my intuition tells me that, all things considered, it's a net
> benefit. I could be wrong but it's my sense that I'm right. And that, after
> all, is the kind of judgment that professionals in every field and in every
> era have had to bring to the table.[/color]

Ooh! Your intuition tells you that? Are you suggesting that you know
better than professional developers at Microsoft / Netscape / Opera
/ Apple / SomeBigCompany which font size an average user wants to
see? How many people have you interviewed? The fact that you like
smaller font than the browser's default isn't a good enough reason
to bug everybody, IMO.
[color=blue][color=green][color=darkred]
>>>body { font-size: 100%; sans-serif; }
>>>
>>>then the text seems ridiculously large to me in ALL the graphical
>>>browsers. So if my general body text is to be sans-serif, then I'm[/color]
>>
>>Yes, the DEFAULT SETTINGS of all graphical browsers DEFAULT to pretty
>>large body text. Why? Perhaps, because if you have to choice between
>>text being RIDICULOUSLY LARGE or UNREADABLE SMALL, then the former is
>>lesser of two evil. You cannot select a default settings that is perfect
>>for everybody, can you? You're supposed to change the settings of YOUR
>>user agent as you want. I repeat, if you don't change the setting or you
>>don't even bother for looking it, you don't mind enough[/color]
>
> You think I shouldn't second-guess people's preferences and motivations, but
> you seem happy enough to do it, as you again show here.[/color]

I cannot understand where the word "here" refers to.

I'll list some facts:
1) You cannot make everybody happy. There isn't true
one-size-fits-all font size when it comes to personal preferences of
reading text on different devices. No matter what the default font
size is, somebody will be unhappy with it.
2) If you cannot provide PERFECT environment, it's better to provide
USABLE environment than UNUSABLE.
3) Reading text that's too large is uncomfortable (worst case), but
it might be IMPOSSIBLE to read text that's too small.
4) It's better to make sure that everybody can USE the environment
than to make it PERFECT for some and UNUSABLE for some.

I think we can agree that points 1)-3) are facts. I'm not sure if
you agree that 4) is a fact. However, if we take that 1)-4) are
facts, then it seems pretty simple to deduce that optimal
one-size-fits-all-default is to make default text big enough so that
almost everybody can read it (the legally-blind people would require
such a huge font size that the text would be PRACTICALLY UNUSABLE
for normal people).

[color=blue][color=green]
>>[ Note that 100% in the root element is supposed to be equal to
>>'medium'. However, you shouldn't use 'medium' or 'small' keywords for
>>any normal text because both of those keywords have different meaning in
>>different versions of MSIE -- that is, some versions interprete 'small'
>>as 100% and 'medium' as something like 110%. And some other versions of
>>the same browser interprete 'medium' as 100% and small as something like
>>90%. Considering that MSIE is the most used browser one cannot really
>>use keywords 'small' or 'medium' for anything. ][/color]
>
> First you tell me that I should defer to the wisdom of the browser
> manufacturers concerning the proper default text size because they[/color]

Perhaps I should have pointed out more clearly that you shouldn't
use 'small' or 'medium' if you care about MSIE users because some
versions of MSIE have BUGS (namely, 'small' and 'medium' have
different meanings than CSS spec says). Unless MSIE were such
popular the issue wouldn't be worth mentioning. But because it seems
that many people in this thread are suggesting that you should use
keywords to define font size, I decided to warn against it.
Percentages should work fine for defining the font size of other
than body text IF you define body { font-size: 100%; }, which is
only a workaround for an another MSIE bug.

I still think that the default font size of many browsers is the
best possible DEFAULT. If YOU think that the default font size is
too large, perhaps YOU should configure your browser. If you modify
the page instead, you get PERFECT text for you, but every visitor to
the page gets TOO SMALL text.


The reason I explain this issue so much is that I really don't want
the future where
1) Page author THINKS that the default font size is too large
2) Page author decides to set body text size to 80%
3) Majority of the page authors do steps 1) and 2) but they disagree
on the percentage; some set it to 95%, some set it to 65%.

Now, if I KNOW that perfect font size FOR ME would be EXACTLY 86% of
the default font size, what size I should set the base font size to
get correct end result after the fact that the page author
readjusted my body text size to "better" value? This is a real
problem already today, but I try to prevent it getting any worse.

--
Mikko

"Mikko Rantalainen" <mira@st.jyu.fi> wrote in message
news:bumreo$3fr$1@mordred.cc.jyu.fi...[color=blue]
> Harlan Messinger / 2004-01-21 00:33:[color=green]
> > "Mikko Rantalainen" <mira@st.jyu.fi> wrote in message
> > news:buk4n8$bm5$1@mordred.cc.jyu.fi...
> >[color=darkred]
> >>2) If the user isn't aware about the possibility, he doesn't care enough
> >>to look for it. Ergo, he's already HAPPY with the font size of 100%.[/color]
> >
> > Non sequitur.[/color]
>
> You did read the rest of my post, right? I explained, that if the
> user really were unhappy with the font size of 100%, he would do
> something.[/color]

Non sequitur. In other words, I don't believe as you do that one can or
should act on the assumption that stimulus and reaction go together in the
inevitable manner that you suggest.
[color=blue]
> After that user has fixed the problem, font size of 100%
> would be the best possible font size for that user. You cannot know
> if an user has already adjusted his browser as he likes. If he has
> already adjusted his browser and 100% is the perfect font size for
> him, wouldn't it be a little dumb to readjust the font size to some
> random smaller size?[/color]

Isn't it dumb to assume that someone has exactly one perfect font size and
is incapable of dealing with any other? Since, as you keep repeating, I
can't know who the reader is, isn't it dumb to assume the reader *is*
someone who has made the adjustments that you claim?
[color=blue]
>And if he hasn't adjusted the font size, does
> that mean that he /really/ wants to read smaller text, but doesn't
> have enough knowledge to fix the issue, or is it that the font size
> he's seeing is already the best one for him?[/color]

Any of these. It varies person by person. It is presumptuous of you to
assume he is happy with the status quo, or that he has even considered
whether he's happy with it.
[color=blue]
>
> The bottom line is that for advanced users that CAN configure their
> browsers correctly, 100% is THE BEST possible font size for the body
> text. And for the rest of the users, you cannot know which font size
> would be the best.
>
> If you know that 100% is THE BEST possible font size for some people
> and the other people really wanted to use some smaller or bigger
> font but don't know how to fix the problem, HOW do you know the
> correct setting?
>
>[color=green][color=darkred]
> >>Why should you then make it smaller??[/color]
> >
> > Because my intuition tells me that, all things considered, it's a net
> > benefit. I could be wrong but it's my sense that I'm right. And that,[/color][/color]
after[color=blue][color=green]
> > all, is the kind of judgment that professionals in every field and in[/color][/color]
every[color=blue][color=green]
> > era have had to bring to the table.[/color]
>
> Ooh! Your intuition tells you that? Are you suggesting that you know
> better than professional developers at Microsoft / Netscape / Opera
> / Apple / SomeBigCompany which font size an average user wants to
> see?[/color]

Wow. And I suppose everything else Microsoft, or Netscape, has done in the
best is smarter than anything I would come up with us well. Because no one
*ever* criticizes Microsoft or Netscape. Everyone in the world thinks that
everything about their browsers is just as good as good can possibly be. Oh,
and everybody in the world loves Clippy, the Help Agent in Microsoft Office
products. (The fact that their on-line help used to be context sensitive and
now no longer is is beside the point, I suppose.) And they were right on
about the need for a UI like Microsoft Bob, weren't they?

Don't *ever* try to get me to cave in on an opinion by holding the
"brilliance" of Microsoft, or of any other company, over my head.

[color=blue]
> How many people have you interviewed? The fact that you like
> smaller font than the browser's default isn't a good enough reason
> to bug everybody, IMO.[/color]

I've already explained the theory behind choice of type size. I didn't make
it up.
[color=blue]
>[color=green][color=darkred]
> >>>body { font-size: 100%; sans-serif; }
> >>>
> >>>then the text seems ridiculously large to me in ALL the graphical
> >>>browsers. So if my general body text is to be sans-serif, then I'm
> >>
> >>Yes, the DEFAULT SETTINGS of all graphical browsers DEFAULT to pretty
> >>large body text. Why? Perhaps, because if you have to choice between
> >>text being RIDICULOUSLY LARGE or UNREADABLE SMALL, then the former is
> >>lesser of two evil. You cannot select a default settings that is perfect
> >>for everybody, can you? You're supposed to change the settings of YOUR
> >>user agent as you want. I repeat, if you don't change the setting or you
> >>don't even bother for looking it, you don't mind enough[/color]
> >
> > You think I shouldn't second-guess people's preferences and motivations,[/color][/color]
but[color=blue][color=green]
> > you seem happy enough to do it, as you again show here.[/color]
>
> I cannot understand where the word "here" refers to.[/color]

Here. This place. The place where you used your psychic powers to determine
that if a user hasn't looked into whether his settings can be changed, it
can only mean that he doesn't "mind enough", and that you can proceed from
that conclusion.

[snip]

> > >>>body { font-size: 100%; sans-serif; }


Why not just omit the 100% altogether?

Leave my fonts ALONE!

On Wed, 21 Jan 2004, Harlan Messinger wibbled:
[color=blue]
> Any of these. It varies person by person. It is presumptuous of you to
> assume he is happy with the status quo,[/color]

Equally, it is presumptuous of you to assume that his vendor has
provided him with something that he cannot use.
[color=blue]
> or that he has even considered whether he's happy with it.[/color]

Yes, I've seen this line of argument before.

1. "the vendor has provided the user with something horrible"
(unproven)

2. "the reader is too dumb to do anything about it" (unproven)

3. "everybody else sets an absolute font size, so the reader has
never seen their vendor's default font size. They would be appalled
if I were to use it" (unproven)

4. "if everyone else is delivering crap, then my only option is
to deliver the same crap". (Oh yes, no argument here!)

Well, it's a point of view. Fortunately, I have my web browser's min.
font size to save you. You can set as many different font sizes as
you like below that, and they'll all come out the same for me. Of
course, when I'm asked to actually _review_ a web page, I have to
remember to turn the min. font size off. And then it's Room 101 for
most of the reviewed pages, because they come out little better than
fly droppings for me.

Noozer / 2004-01-22 01:56:[color=blue][color=green][color=darkred]
>>>>>>body { font-size: 100%; sans-serif; }[/color][/color][/color]
^^^^^^^^^^
[color=blue]
> Why not just omit the 100% altogether?
>
> Leave my fonts ALONE![/color]

The page author should set body { font-size: 100% } if he's going to
set font-size of any element (say h1, h2, h3, span.footnote etc
etc). This is because MSIE has a bug that causes View:Text Size
control to freak out if any font size is defined with percentage
unit without specifying font size of body element. The point is,
body { font-size: 100% } should already be the browser default, but
somehow, it makes difference in case of MSIE.

--
Mikko

Mikko Rantalainen <mira@st.jyu.fi> wrote:[color=blue]
>Noozer / 2004-01-22 01:56:[color=green]
>>
>> Why not just omit the 100% altogether?
>>[/color]
>The page author should set body { font-size: 100% } if he's going to
>set font-size of any element (say h1, h2, h3, span.footnote etc
>etc). This is because MSIE has a bug that causes View:Text Size
>control to freak out if any font size is defined with percentage
>unit without specifying font size of body element.[/color]

Not true. The bug in IE is when font-size is set in ems then the size
difference is vastly inflated when the user has anything other than
Medium set.

So the solution is to either:
1. Set 100% on the body and then use ems
2. Use percentages throughout in which case you don't need an explicit
size set on the body.

Compare http://steve.pugh.net/test/test67a.html and
http://steve.pugh.net/test/test67b.html to see the bug in action.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <steve@pugh.net> <http://steve.pugh.net/>

[color=blue]
> Not true. The bug in IE is when font-size is set in ems then the size
> difference is vastly inflated when the user has anything other than
> Medium set.
>
> So the solution is to either:[/color]
[color=blue]
> 1. Set 100% on the body and then use ems[/color]
^ see below.[color=blue]
> 2. Use percentages throughout in which case you don't need an explicit
> size set on the body.[/color]

http://www.thenoodleincident.com/tut...phy/index.html

( and study this pages css )
http://www.thenoodleincident.com/tut.../template.html

http://www.alistapart.com/articles/elastic/

--
Mark

Mark Woodward <markwoodwardHATES_SPAM@internode.on.net> wrote:[color=blue]
>[color=green]
>> Not true. The bug in IE is when font-size is set in ems then the size
>> difference is vastly inflated when the user has anything other than
>> Medium set.
>>
>> So the solution is to either:[/color]
>[color=green]
>> 1. Set 100% on the body and then use ems[/color]
> ^ see below.[/color]

Okay, "100% or any other non-em explicit size". Could be 80%, could be
22px, could be 12cm. But in the context that was being discussed 100%
is the logical example to give.
[color=blue][color=green]
>> 2. Use percentages throughout in which case you don't need an explicit
>> size set on the body.[/color]
>
>http://www.thenoodleincident.com/tut...phy/index.html[/color]

Agress with what I said except that he seems to think that smaller
than 100% for main body text is a good idea. I disagree. His page is
also unreadably small when IE is set to Smallest, which is the very
thing we're trying to avoid.

He's also obsessed with making the text the same size in all browsers
which is frankly a waste of time. My primary browser is Opera 7 and
doesn't show the percentages the same size a his screenshot of Opera
7.
[color=blue]
>( and study this pages css )[/color]

Yes, his CSS says that 76% is in his opinion the smallest you can
practiaclly use. So he's using the smallest possible font size for his
main body text. What's he going to use for his footnotes and legalese?

He's also using Verdana which as any long time reader of this
newsgroup knows causes most of the problems...
[color=blue]
>http://www.thenoodleincident.com/tut.../template.html[/color]

Triggers my minimum font size setting in Opera. Not a good indication.
Some of the text becomes unreadably small at just a small reduction to
90% zoom in Mozilla.
[color=blue]
>http://www.alistapart.com/articles/elastic/[/color]

Would work just as well (or badly as it also advocates much smaller
than default font sizes) with percentages used for all font sizes.
Some people seem obsessed by ems and don't seem to realise that for
font-size percentages are exactly the same thing.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <steve@pugh.net> <http://steve.pugh.net/>

On Thu, 22 Jan 2004, Mark Woodward wrote:
[color=blue]
> http://www.thenoodleincident.com/tut...phy/index.html[/color]

Was that supposed to be a recommendation? That page is only rescued
by my min. font size setting - if I turn that off, then I get
microfonts.

As, indeed, you'd expect from what it says in the text. The author
requires readers to select a font size much larger than what they want
to see, so as to let the author size it down for them. Logically
absurd (yes, yes, I'm perfectly well aware of Win/IE): calling that
"sane" is, well, petitio principii.

"Alan J. Flavell" <flavell@ph.gla.ac.uk> wrote in message
news:Pine.LNX.4.53.0401212314080.16106@ppepc56.ph. gla.ac.uk...[color=blue]
> On Wed, 21 Jan 2004, Harlan Messinger wibbled:
>[color=green]
> > Any of these. It varies person by person. It is presumptuous of you to
> > assume he is happy with the status quo,[/color]
>
> Equally, it is presumptuous of you to assume that his vendor has
> provided him with something that he cannot use.[/color]

This is what Mikko said to me in the first place. I was responding to *him*
that *he* was being equally presumptuous. Is there a point to going around
in circles?
[color=blue]
>[color=green]
> > or that he has even considered whether he's happy with it.[/color]
>
> Yes, I've seen this line of argument before.
>
> 1. "the vendor has provided the user with something horrible"
> (unproven)
>
> 2. "the reader is too dumb to do anything about it" (unproven)
>
> 3. "everybody else sets an absolute font size, so the reader has
> never seen their vendor's default font size. They would be appalled
> if I were to use it" (unproven)
>
> 4. "if everyone else is delivering crap, then my only option is
> to deliver the same crap". (Oh yes, no argument here!)[/color]

Just like every book that has ever been published in other than the one and
only and unique perfect optimal font size for the average human being has
been crap. I feel that the view you are presenting of the narrowness of
human beings' requirements is grossly and needlessly simplistic, and that if
the situation were as dire as you paint it to be, it would be astounding
that the human species has managed as long as it has.

"Noozer" <postmaster@127.0.0.1> wrote in message
news:_kEPb.215223$ts4.14712@pd7tw3no...[color=blue][color=green][color=darkred]
> > > >>>body { font-size: 100%; sans-serif; }[/color][/color]
>
>
> Why not just omit the 100% altogether?
>
> Leave my fonts ALONE![/color]

I take it you refuse to read any book, magazine, or newspaper that is
printed in other than your one and only and immutable font size, without any
tolerance for even the most minute deviation.

I should point out that there are usability benefits to smaller fonts. For
those whose visual acuity is not a barrier, they greatly improve reading
speed an comprehension, because (obviously) the larger chunks of text are
within the main field of view and the brain is able to process them better.
Also, more material fits on the screen, making it easier for the user to
mentally relate parts of the display, follow the material, and find one's
place. Visual acuity is not the only consideration.

On Thu, 22 Jan 2004, Harlan Messinger wrote:
[color=blue]
> I should point out that there are usability benefits[/color]

....in choosing an optimal font size, yes.
[color=blue]
> to smaller fonts.[/color]

But not "smaller" than the reader's optimal choice!

By all means set *your* browser to an appropriate font size, and/or
lobby your vendor for better font sizing arrangements.

Recognise this? :-

The font size chosen by the user as a comfortable default (1 em)
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.

That comes from a Real Typographer who understood the difference
between printed books and the web (as you have just demonstrated you
do not).
[color=blue]
> For those whose visual acuity is not a barrier, they greatly improve
> reading speed an comprehension, because (obviously) the larger
> chunks of text are within the main field of view and the brain is
> able to process them better. Also, more material fits on the screen,
> making it easier for the user to mentally relate parts of the
> display, follow the material, and find one's place.[/color]

(Talk about stating the bleeding obvious...)
[color=blue]
> Visual acuity is not the only consideration.[/color]

If the author has made the text too small to be read comfortably - or
at all, then it *is* the primary consideration, nevertheless. And the
notorious operating system component (the very one that all the
argument rages around) does not support a min. font size setting.
Unlike quite a number of WWW-compatible browsers. So we mustn't be
too surprised that its vendor errs on the side of caution and sets its
initial default font size too large for many people's tastes and
needs.

Steve Pugh wrote:[color=blue]
> Mark Woodward <markwoodwardHATES_SPAM@internode.on.net> wrote:
>[color=green]
>>http://www.alistapart.com/articles/elastic/[/color]
>
> it also advocates much smaller than default font sizes[/color]

A List Apart's site is really user unfriendly. I played around one day
making it better with user css. The following produces reasonable
results in Mozilla 1.3. Here it is in case anyone else wants to fix
their site.

body[onload*="A List Apart"] * {
font-size: 100% !important ;
font-family: lucida, arial, helvetica, sans-serif !important;
}

body[onload*="A List Apart"] div#wrapper {
margin: 10px 25px !important;
width: auto !important;
}

body[onload*="A List Apart"] div#maincontent {
margin: auto !important;
width: 38em !important;
}

body[onload*="A List Apart"] div#pagebody,
body[onload*="A List Apart"] div#sidebar,
body[onload*="A List Apart"] div#subnav {
width: auto !important;
margin: auto !important;
}

body[onload*="A List Apart"] div#menu li#five a {
border-right: 1px solid #ccc !important;
}

body[onload*="A List Apart"] div#footer {
font-size: 90% !important ;
}

body[onload*="A List Apart"] a:link, body[onload*="A List Apart"]
a:visited {
text-decoration: underline !important;
}

body[onload*="A List Apart"] a:visited {
color: #03c !important ;
}

body[onload*="A List Apart"] a:hover, body[onload*="A List Apart"]
a:focus {
color: #e30 !important ;
background-color: #bbb !important ;
text-decoration: none !important;
}

body[onload*="A List Apart"] a:active {
color: #33d !important ;
background-color: #bbb !important ;
}



And if anyone can improve on this, I'd appreciate knowing how.

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Steve Pugh wrote:[color=blue]
> The bug in IE is when font-size is set in ems then the size
> difference is vastly inflated when the user has anything other than
> Medium set[/color]
[color=blue]
> Use percentages throughout in which case you don't need an explicit
> size set on the body.[/color]

Are you sure of this? This is not intended to be read in a smart-alec
way. I too was under the (mistaken?) impression that you had to use
body {font-size: 100%} to avoid bugs. But I no longer use ems for
font-size. The thing is, I have only IE 5.01/Win2k to test on, and I'd
hate to have things break badly. So I just want to be sure before
removing that body font-size declaration. :)

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Brian <usenet2@julietremblay.com.invalid-remove-this-part> wrote:
[color=blue]
>A List Apart's site is really user unfriendly. I played around one day
>making it better with user css. The following produces reasonable
>results in Mozilla 1.3. Here it is in case anyone else wants to fix
>their site.[/color]

[snip]
[color=blue]
>And if anyone can improve on this, I'd appreciate knowing how.[/color]

http://steve.pugh.net/taming.html

I concentrated on removing the fixed widths as Opera's minimum font
size setting had already taken care of the tiny text.

I was also writing with the aim of creating a cross-browser user
stylesheet hence my use of the #sectionone as a selector rather than
the attribute selector you use which whilst betterin many ways isn't
supported by IE.

Fixing the lack of distinction between :link and :visted is a good
idea. I may add that to my stylesheet as well.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <steve@pugh.net> <http://steve.pugh.net/>

Steve Pugh wrote:[color=blue]
>
> The bug in IE is when font-size is set in ems then the size
> difference is vastly inflated when the user has anything other than
> Medium set.
>
> Compare http://steve.pugh.net/test/test67a.html and
> http://steve.pugh.net/test/test67b.html to see the bug in action.[/color]

BTW, there is no difference in the a and b tests in MSIE 5.01/Win2k.
In both cases, 80% is different than .8em. I tested on "large" and
"small."

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Harlan Messinger wrote:[color=blue]
> I feel that the view you are presenting of the narrowness of
> human beings' requirements is grossly and needlessly simplistic[/color]

Uh, that's backwards. We who argue that font-size: 100% for body text
is correct are assuming that "one-size fits all" is *not* going to
work. font-size: 100% means that lots of text-sizes will be used, and
that's a good thing.

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Harlan Messinger wrote:[color=blue]
> "Noozer" <postmaster@127.0.0.1> wrote in message
> news:_kEPb.215223$ts4.14712@pd7tw3no...
>[color=green]
>> Leave my fonts ALONE![/color]
>
> I take it you refuse to read any book, magazine, or newspaper that is
> printed in other than your one and only and immutable font size[/color]

I take it you don't see the difference between print and the web?

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Brian <usenet2@julietremblay.com.invalid-remove-this-part> wrote:
[color=blue]
>Steve Pugh wrote:[color=green]
>>
>> The bug in IE is when font-size is set in ems then the size
>> difference is vastly inflated when the user has anything other than
>> Medium set.
>>
>> Compare http://steve.pugh.net/test/test67a.html and
>> http://steve.pugh.net/test/test67b.html to see the bug in action.[/color]
>
>BTW, there is no difference in the a and b tests in MSIE 5.01/Win2k.
>In both cases, 80% is different than .8em. I tested on "large" and
>"small."[/color]

Would have helped if I'd put font-size: 100% instead of just size:
100%.

Now working as I described originally in IE 5.01, IE 5.5 and IE6 on
WinXP.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <steve@pugh.net> <http://steve.pugh.net/>

Brian <usenet2@julietremblay.com.invalid-remove-this-part> wrote:
[color=blue]
>Steve Pugh wrote:[color=green]
>> The bug in IE is when font-size is set in ems then the size
>> difference is vastly inflated when the user has anything other than
>> Medium set[/color]
>[color=green]
>> Use percentages throughout in which case you don't need an explicit
>> size set on the body.[/color]
>
>Are you sure of this? This is not intended to be read in a smart-alec
>way. I too was under the (mistaken?) impression that you had to use
>body {font-size: 100%} to avoid bugs.[/color]

I've never come across a bug with % sized text in Internet Explorer
(apart from when it screws up inheritence, e.g. with tables, but that
applies equally to ems).

Opera 5 and 6 (and very early versions of 7?) had a bug whereby 100%
was rendered 1px too small and thus inherited 100% would actually make
the font size increasingly small. This is why some people suggest
using 101% instead. But this bug was fixed in Opera a fairly early 7.x
release. Netscape 4 has a funny idea of what a % is, but Netscape 4
has a funny idea of what everything is.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <steve@pugh.net> <http://steve.pugh.net/>

Steve Pugh wrote:[color=blue]
> Brian wrote:
>[color=green]
>> A List Apart's site is really user unfriendly. I played around
>> one day making it better with user css.
>> And if anyone can improve on this, I'd appreciate knowing how.[/color]
>
> http://steve.pugh.net/taming.html[/color]

clever name :)

From A List Apart (which you quoted):
<quote>
ALA’s layout is CSS-based and our articles share a common id attribute
to the body tag, so if your browser provides the option to "Use My
Style Sheet" you can adjust ALA's look and feel to fit your fancy.
</quote>

Balderdash! They offer no such common id. Some pages use "sectionone",
some use "sectionthree". I don't know how many sections there are.
That's why I had to resort to the onload attrib selector.
[color=blue]
> I concentrated on removing the fixed widths as Opera's minimum font
> size setting had already taken care of the tiny text.[/color]

I don't use that on Mozilla because I want to be able to critique
sites without opening up prefs and turning things off and on too often.
[color=blue]
> I was also writing with the aim of creating a cross-browser user
> stylesheet hence my use of the #sectionone as a selector[/color]

But what do you do when your browse to a different section of the site?
[color=blue]
> rather than the attribute selector you use which whilst betterin
> many ways isn't supported by IE.[/color]

My solution will only work in Moz and possibly Opera. I figure that
for the ciwas audience, that will help lots of people. If there were a
common id, it could work in MSIE, too. And it would have been easier
for me to write it. (I had some trouble getting the attrib selector to
work.)
[color=blue]
> Fixing the lack of distinction between :link and :visted is a good
> idea. I may add that to my stylesheet as well.[/color]

:)

I wonder if showing them the user css would embarass them?

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/