Site critque request plz

On Sat, 29 Jan 2005 02:34:02 +0100, Bernhard Sturm
<sturmnixspam (AT) datacomm (DOT) ch> wrote:

Stephen Poley wrote:

The text is too small for comfortable reading for me, and Verdana is not
a very good idea. See
http://www.xs4all.nl/~sbpoley/webmatters/fontsize.html and
http://www.xs4all.nl/~sbpoley/webmatters/verdana.html

you don't mean this for real, don't you?

Actually I do.

The above two links do state a couple of facts that I would consider as
very dangerous for the least:

I've read your whole post carefully, and I can't for the life of me see
what you think is "very dangerous". What precisely is the danger you are
referring to?

Verdana was designed for best screen legibility, and is one of the very
rare things Microsoft introduced for the benefit of the web

So you judge Microsoft's products by the manufacturer's claims, do you?
I judge them by the results.

Claiming Verdana is a bad font choice because of it's size would neglect
the fact that you can set font sizes with CSS.

If you think I've neglected that then you need to go back and reread the
pages you are criticising.

There is a reason why you
should avoid serif typefaces for small sizes. The reason is that your
screen has a very limited resolution compared to a sheet of paper.
It's a big mistake to think that if serif fonts are good in print they
must be good on the screen as well.. they are not. google a bit about
this issue.

This is actually a side issue; however I have indeed spent quite a bit
of time on this. I have come across quite a few *assertions* that
sans-serif fonts are better on the screen, but I have yet to come across
any *evidence* - i.e. results of carefully set up comparative tests. If
you have found some, by all means give us the URL. (I suspect that these
claims are based on a superficial and unfair comparison of a wide
sans-serif font with a narrow serif font, but I'm happy to look at
evidence to the contrary.)

If you are going to propagate serif fonts,

I really wonder whether you've actually read my pages. Where do I
"propagate serif fonts"? What I propagate is the ability of the reader
to select whatever he/she finds pleasant to read.

then propagate
Georgia, but forget Times (or as a windows user Times New Roman), they
were not designed for screen readability,

And yet Times New Roman is the default font in Internet Explorer and
indeed most other browsers. So one can hardly ignore it when considering
such matters as font sizes. And personally I find it a pleasant font to
read on the screen; you may not.

but are great typfaces for
print purposes.
In your quoted links are some other fundamental errors:
There are no special fonts defined in a UA.
Fonts are available on the users system but not on the UA itself.
snip

I can't work out what you are referring to here. Do I claim otherwise?

Another thing is your definition of 100% font sizes... you better use em
or en for text-size

Oh really? Why? What do you suppose is the difference between 100% and
1em? And where in the CSS spec does it specify the en as a valid unit?

(the equivalent width of the uppercase letter M or N
as a referenceof the relevant font).

Wrong. Normally I would let this pass. But in the context of a
vituperative post such as yours, you really ought to get your facts
right.

1em is defined as the equivalent of
16px by the W3C.

Utter nonsense.

Something striking for those of you who always thought that px is an
absolute length unit: according to the W3C it is a relative unit (like
em),

I get the impression that your understanding of pixels is superficial,
to put it mildly. In fact I get the impression that your understanding
of the whole subject area is pretty superficial.

--
Stephen Poley

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

Stephen Poley wrote:

I get the impression that your understanding of pixels is superficial,
to put it mildly. In fact I get the impression that your understanding
of the whole subject area is pretty superficial.

Seconded.

--
Ben M.

Stephen Poley wrote:

Claiming Verdana is a bad font choice because of it's size would neglect
the fact that you can set font sizes with CSS.


If you think I've neglected that then you need to go back and reread the
pages you are criticising.

"Incidentally it is sometimes said that using Verdana in a user
stylesheet is sound, but even this is doubtful: an inoffensive author
style specification like font-family: serif; font-size: 100% will leave
you with text that is too small."

It's not a crime to define 120% font-size values in CSS.
so I suggest you have once a look at the 264 screenshots done by Owen
Briggs in order to get your text sizing problem solved :-)

http://www.thenoodleincident.com/tut...nt/method.html

This is actually a side issue; however I have indeed spent quite a bit
of time on this. I have come across quite a few *assertions* that
sans-serif fonts are better on the screen, but I have yet to come across
any *evidence* - i.e. results of carefully set up comparative tests. If
you have found some, by all means give us the URL. (I suspect that these
claims are based on a superficial and unfair comparison of a wide
sans-serif font with a narrow serif font, but I'm happy to look at
evidence to the contrary.)

okay. you asked for it, so I am afraid you have to read it as well:
http://psychology.wichita.edu/hci/pr...r%20Adults.pdf

If you are going to propagate serif fonts,


I really wonder whether you've actually read my pages. Where do I
"propagate serif fonts"? What I propagate is the ability of the reader
to select whatever he/she finds pleasant to read.

Your title of the page reads "Why you should avoid the Verdana font",
and then a bit below:
"Note how much bigger Verdana is than the others, even though the point
size is the same. This is particularly so in comparison to Times New
Roman, which is a common browser default font."
So I assumed you prefer Times New Roman instead of Verdana, which would
make it serif vs. sans-serif.
On your page you ponder about the various font-sizes if you define the
same pt height. This is a very common effect in typography. Although a
typeface size is defined in pt, there are no standardised definitions
where to measure those pts. Some typographers measure the height from
the p-line to the x-line or up to the k-line. Others measure the height
of the 'Versalien' (uppercase) from the baseline to the H-line in order
to define the size in pt. That's why you will always get variable
font-sizes depending on the typeface you use.

And yet Times New Roman is the default font in Internet Explorer and
indeed most other browsers. So one can hardly ignore it when considering
such matters as font sizes. And personally I find it a pleasant font to
read on the screen; you may not.

what is your 'default' font? Times New Roman will be chosen from your
browser if a 'serif' font is being used. This has nothing to do with
'default' font. If you avoid all font statements in your HTML/CSS
document then this serif font will be selected. But this doesn't mean
that it's good practice. (or as good old David Siegel would put it:
first generation websites would have worked this way ;-)

Oh really? Why? What do you suppose is the difference between 100% and
1em?

No technical difference but a practical one: IE has an ems cascading bug
AFAIK.

(the equivalent width of the uppercase letter M or N
as a referenceof the relevant font).


Wrong. Normally I would let this pass. But in the context of a
vituperative post such as yours, you really ought to get your facts
right.

so read this: http://webdesign.crissov.de/Typographie#Geviert.
1 em is the width of the uppercase letter M (quad, German: Geviert)

1em is defined as the equivalent of
16px by the W3C.


Utter nonsense.

True they don't state it explicitly on their pages. But 16px are assumed

by most visual UAs as the equivalent of 16px.

Something striking for those of you who always thought that px is an
absolute length unit: according to the W3C it is a relative unit (like
em),


I get the impression that your understanding of pixels is superficial,
to put it mildly.

I've only stated what the W3C says (you dont' have to believe it :-)
(http://www.w3.org/TR/REC-CSS2/syndat...absrel-units): px are like
ems relative. relative to the viewing device. Why is this a superficial
notion?

In fact I get the impression that your understanding

of the whole subject area is pretty superficial.

thanx... you seem to be a typographer as well?



--
www.daszeichen.ch
remove nixspam to reply


http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Bernhard Sturm wrote:

Stephen Poley wrote:

This is actually a side issue; however I have indeed spent quite a bit
of time on this. I have come across quite a few *assertions* that
sans-serif fonts are better on the screen, but I have yet to come across
any *evidence* - i.e. results of carefully set up comparative tests. If
you have found some, by all means give us the URL. (I suspect that these
claims are based on a superficial and unfair comparison of a wide
sans-serif font with a narrow serif font, but I'm happy to look at
evidence to the contrary.)

okay. you asked for it, so I am afraid you have to read it as well:
http://psychology.wichita.edu/hci/pr...r%20Adults.pdf

Lol, it appears you pulled that page from a google search without
reading it properly. If you had, you'd realise that it is a comparison
of only four specific fonts and hence makes no statement to support the
argument that "sans-serif fonts have better onscreen legibility than
serif fonts".

Oh really? Why? What do you suppose is the difference between 100% and
1em?

No technical difference but a practical one: IE has an ems cascading bug
AFAIK.

Oh so evilly snipped:

Bernhard Sturm wrote:

Stephen Poley wrote:
Bernhard Sturm wrote:

Another thing is your definition of 100% font sizes... you better use em
or en for text-size

Oh really? Why? What do you suppose is the difference between 100% and
1em? And where in the CSS spec does it specify the en as a valid unit?

No technical difference but a practical one: IE has an ems cascading bug
AFAIK.

So, why had Stephen "better use em or en for text-size"? And /how/ can
he use "en" for text size?

Sometimes questions asked are for /your/ benefit, not others.

(the equivalent width of the uppercase letter M or N as a referenceof
the relevant font).

Wrong. Normally I would let this pass. But in the context of a
vituperative post such as yours, you really ought to get your facts
right.

so read this: http://webdesign.crissov.de/Typographie#Geviert.
1 em is the width of the uppercase letter M (quad, German: Geviert)

<translated>
However, a Geviert (in English, quad or em) sometimes becomes as simple
as the width of the uppercase M, or like the em unit in CSS when
specifying the height of a character.
</translated>

<emphasis>
A Geviert becomes like the em unit in CSS when specifying the height of
a character.
</emphasis>

Perhaps /you/ should have read it. Not reading the references supporting
your arguments will make them flimsy and easily to blow over.

Something striking for those of you who always thought that px is an
absolute length unit: according to the W3C it is a relative unit
(like em),

I get the impression that your understanding of pixels is superficial,
to put it mildly.

I've only stated what the W3C says (you dont' have to believe it :-)
(http://www.w3.org/TR/REC-CSS2/syndat...absrel-units): px are like
ems relative. relative to the viewing device. Why is this a superficial
notion?

_The notion_ is not superficial, _your understanding_ is.

--
Ben M.

Ben Measures wrote:

Lol, it appears you pulled that page from a google search without
reading it properly. If you had, you'd realise that it is a comparison
of only four specific fonts and hence makes no statement to support the
argument that "sans-serif fonts have better onscreen legibility than
serif fonts".

"As seen in Figure 5, both males and females greatly preferred as their
first or second choice the 14-point sans serif fonts Verdana and Arial
to the 14-point serif fonts, Georgia and Times. Thus, it appears that
greatest preference differences among the typefaces at this size are not
between the print and the computer-displayed fonts, but between the
serif and sans serif fonts." (p 7)

But don't tell me now that *preferring* doesn't mean legibility ;-)

No technical difference but a practical one: IE has an ems cascading bug
AFAIK.


So, why had Stephen "better use em or en for text-size"? And /how/ can
he use "en" for text size?

That was a bit confusing, I admit, as 'ems' do just belong to the same
unit family as 'en' or 'ex'... CSS allows you to use 'em' and 'ex' for
your font-size. So this was confusing, but the core thing is correct:
W3C suggests the use of 'em' over percentage.

so read this: http://webdesign.crissov.de/Typographie#Geviert.
1 em is the width of the uppercase letter M (quad, German: Geviert)


translated
However, a Geviert (in English, quad or em) sometimes becomes as simple
as the width of the uppercase M, or like the em unit in CSS when
specifying the height of a character.
/translated>¨
outch.. your translation is not quite correct: 'einfach' in German is

not always translated into 'simple'. And 'Definiert' ist not always
translated as 'becomes' :-), but I assume you are a native German
speaker so you know this :-)
Another source:
"The font height is in it self such a significant unit of length for a
typesetter, that it has been used to define a universal unit of relative
length by the name of one em, where the name may stem from the fact that
the letter M in the Roman alphabet did in fact have a width that filled
the full width of the square based on the height of the type in use."
(http://css.nu/articles/typograph1-en.html)

or another one:
It's German again, sorry for that:
http://aktuell.de.selfhtml.org/artikel/css/fontsize/#a7

hmm maybe my understanding of this is not good enough, but I have read
this in many typography books, and was taught by typographers about
this, and I get a bit nervous if someone just comes, and tells, that
this is plain wrong without telling me, what exactly is wrong. I do get
also a bit nervous if there are simple accusations without arguments
supporting the accusation. Telling me that I am superficial is okay, but
as far as I can see, I was the only one providing some sources in order
to prove my point. However, I do accept if you prove me wrong, no
problem with that. As I am here to learn things. (and I have learnt a
lot in NGs, but insults are never a good starting point to start a
teaching lesson)
So don't take it personal, just wanted to make that clear. ;-)

--
www.daszeichen.ch
remove nixspam to reply


http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Bernhard Sturm wrote:

Ben Measures wrote:

Lol, it appears you pulled that page from a google search without
reading it properly. If you had, you'd realise that it is a comparison
of only four specific fonts and hence makes no statement to support
the argument that "sans-serif fonts have better onscreen legibility
than serif fonts".

"As seen in Figure 5, both males and females greatly preferred as their
first or second choice the 14-point sans serif fonts Verdana and Arial
to the 14-point serif fonts, Georgia and Times. Thus, it appears that
greatest preference differences among the typefaces at this size are not
between the print and the computer-displayed fonts, but between the
serif and sans serif fonts." (p 7)

A preference for Verdana and Arial over Georgia and Times _does not_
mean that the preference is for san-serif typefaces. That's like saying
a preference for a Harley over a Fiat means a preference for motorbikes
over cars.

There are many factors involved in legibility, arguably the greatest of
which is the x-height (in proportion to the overall size). This gives
Verdana *huge* advantage over Times, invalidating any further
serif/sans-serif conclusions drawn from these two fonts.

So, why had Stephen "better use em or en for text-size"? And /how/ can
he use "en" for text size?

That was a bit confusing, I admit, as 'ems' do just belong to the same
unit family as 'en' or 'ex'... CSS allows you to use 'em' and 'ex' for
your font-size. So this was confusing, but the core thing is correct:
W3C suggests the use of 'em' over percentage.

Where? I'd be very surprised to find an official line on this,
especially considering IE's broken implementation, as you yourself
pointed out.

so read this: http://webdesign.crissov.de/Typographie#Geviert.
1 em is the width of the uppercase letter M (quad, German: Geviert)

translated
However, a Geviert (in English, quad or em) sometimes becomes as
simple as the width of the uppercase M, or like the em unit in CSS
when specifying the height of a character.
/translated

Another source:
"The font height is in it self such a significant unit of length for a
typesetter, that it has been used to define a universal unit of relative
length by the name of one em, where the name may stem from the fact that
the letter M in the Roman alphabet did in fact have a width that filled
the full width of the square based on the height of the type in use."
(http://css.nu/articles/typograph1-en.html)

hmm maybe my understanding of this is not good enough, but I have read
this in many typography books, and was taught by typographers about
this, and I get a bit nervous if someone just comes, and tells, that
this is plain wrong without telling me, what exactly is wrong.

Okay, I'll spell out for you what's wrong:

You consistently refer to the em as a _width_ of a character, when *all*
of the references you've provided use it as a measurement for the
relative _height_.

Thus, saying "1 em is the width of the uppercase letter M" is plain
wrong. That's like saying a house is two storeys wide.

Telling me that I am superficial is okay,

Nobody said /you/ were superficial, only your understanding.

but as far as I can see, I was the only one providing some sources in
order to prove my point.

Your references do not prove your point, frequently they prove the
opposite. Take some effort to understand them before you try to use them
in your arguments. Accept that they may not say what you want them to
say and be prepared to change your opinions accordingly.

--
Ben M.

Ben Measures wrote:

Bernhard Sturm wrote:

Another source:
"The font height is in it self such a significant unit of length for a
typesetter, that it has been used to define a universal unit of
relative length by the name of one em, where the name may stem from
the fact that the letter M in the Roman alphabet did in fact have a
width that filled the full width of the square based on the height of
the type in use."
(http://css.nu/articles/typograph1-en.html)


Okay, I'll spell out for you what's wrong:

You consistently refer to the em as a _width_ of a character, when *all*
of the references you've provided use it as a measurement for the
relative _height_.

<quoteformabove>...letter M in the Roman alphabet did in fact have a
<emphasise>width</emphasise> that filled the full width...</quotefromabove>

OMG: let's stop it here. You just don't get it, don't you? em is a
measurement for the font-size, and that's the height. But it has its
reference in the *width* of the character M.... (*Laufweite* in German,
as refered to in the second URL I posted) Is that so difficult to get?
Read the sources. That's with all units: they are founded somewhere. em
is founded in the width of the character M, but ems are used to describe
the font height... I can see where your confusion comes from.

Thus, saying "1 em is the width of the uppercase letter M" is plain
wrong.

*sigh* at least you could have read my sources...

That's like saying a house is two storeys wide.

The unit meter [m] can be measured horizontally and vertically. I think
you don't have a problem with that? The same goes for ems:

M M
MM MM
M M M
M M
----- <- this is the definition of 1 'em' (5 x -), the width or
*LAUFWEITE" of the letter M

now (not to scale, just for illustration purposes):

OOO |
O O |
O O |
O O |
OOO |

This letter 'O' has the size of 1 'em' (5 x - or the equivalent of the
width of the letter 'M')

Your references do not prove your point, frequently they prove the
opposite. Take some effort to understand them before you try to use them
in your arguments. Accept that they may not say what you want them to
say and be prepared to change your opinions accordingly.

I leave that as it is.
Cheers and have a good night
bernhard


--
www.daszeichen.ch
remove nixspam to reply


http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Bernhard Sturm wrote:

em is founded in the width of the character M

http://doubletype.sourceforge.net/?about%20em

--
Ben M.

Ben Measures wrote:

Bernhard Sturm wrote:


em is founded in the width of the character M


http://doubletype.sourceforge.net/?about%20em

cheers I knew this source :-) and thought, you might come up with it.
It's not a bad one. The author quotes various sources in order to prove
his point that today the em is no longer used as the reference to the
width of the letter M, and we could close this argument here, that
historically the em does indeed stem from the width of the letter M, but
not in nowadays typfaces. AnOther source which would prove us both as
correct ist from the W3:

http://www.w3.org/WAI/GL/css2em.htm

(at least you have to admit that my statements were not just *wrong,
wrong* as you claimed them to be). Now off to enjoy the sunday. I
suggest you do that as well. there is life out here.

;-)


--
www.daszeichen.ch
remove nixspam to reply


http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Bernhard Sturm <sturmnixspam (AT) datacomm (DOT) ch> wrote:

If you
use a windows OS then your system has most certainly installed: Arial,
Verdana, Georgia, Times New Roman.

Verdana was first supplied with W98, Georgia afaik isn't even available
for W9x, it's not on my system (W98+IE6+Office97 fonts). Assuming that
clients are using a certain OS is a fundamental mistake to begin with,
assuming that a client system has certain fonts installed even if they
are using a Windows OS is equally folly since users have full control
over which fonts are installed or not.

--
Spartanicus