Re: [css-d] font-family font sizes

2007-05-18 Thread Bruno Fassino
On 5/18/07, Philippe Wittenbergh [EMAIL PROTECTED] wrote:

 On May 18, 2007, at 6:44 AM, Jukka K. Korpela wrote:

  On Thu, 26 Apr 2007, Philippe Wittenbergh wrote:
 
  The only problem I have atm is finding the aspect value for a given
  font. The font contains that information, but I haven't found an
  utility to tell me that value, nor any resource for it - especially
  for fonts I don't have.
 
  This sounds like an odd situation. Are we all expected to find out
  such things by ourselves?


I don't know if the following is accurate or not, anyway:

- The aspect ratio is defined as ratio between x-height and font-size.
- The x-height is the unit used in CSS lengths expressed in 'ex'.
- Gecko seems able to use an accurate 'ex' unit (I guess it is
extracted from the font in use, even version not supporting
font-size-adjust seem to have a good 'ex')

Putting together the above, if we trust Gecko,  the aspect ratio of a
font can be obtained simply observing the size of an element
dimensioned using 'ex' units.
I've created a quick and dirty test page [1], with includes due
elements sized one in 'em' and another in 'ex'. A javascript gets the
size of the elements in pixels and computes their ratio. This should
be the aspect value of the font in use.  To minimize errors a big font
size is used (and the computation is repeated with some different font
size / widths, the results compared, and discrepancies signaled.)

I get reasonable results, maybe with the exception of Verdana, where I
get as aspect value 0.55 (not 0.58)
In the page I've also added a red line drawn at 'baseline - x-height'
to visually check the accuracy of the computed x-height.  And there is
the possibility to choose a second font family, which is then
displayed (with and without font-size-adjust) together with the main
one (Again: the page makes sense only if using Gecko.)

Bruno

[1] http://brunildo.org/test/xheight.pl
__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-05-18 Thread Jukka K. Korpela
On Fri, 18 May 2007, Philippe Wittenbergh wrote:

 It seems that font-size-adjust helps in some sizing issues on
 Firefox 2 (Windows)
- -
 Yeah, it helps readability of articles, like when you mix two font-
 families in the same run of text:
 p text text codecode/code text text/p
 where code uses a monospaced font and p uses a sans-serif font.

Interesting. It seems to help with the problem that monospace fonts and 
proportional fonts are quite often too differently sized. I wonder if we 
can (and should) simply set
* { font-size-adjust: 0.5; }
where 0.5 is to be replaced by the relative x-height (aspect ratio) of the 
primary font suggested for copy text.

So far I haven't found any risks with this idea, but there are surely some 
nasty bugs lurking around somewhere.

 On a slightly un-related note: may I assume that those have the full
 range of unicode characters available when installed on XP or Vista ?

No, the coverage is still rather limited, but it includes Unicode 5 
coverage _for some scripts_ (Latin, Greek, Cyrillic, some others). There 
is _no_ font that covers all of Unicode. (Code2000 probably comes closest 
to covering BMP.)

This might be relevant if you use special characters that have to be taken 
from fonts other than the copy text font. If a character is designed in a 
small size relative to the font size, then the font size adjustment may 
turn it into too small (or too large). But this is just a possibility and 
not a very serious one, as compared with all the other problems you have 
with special characters and insufficient coverage in fonts.

 Afaik, upgrading to the latest MS Office installs those fonts.

Yes, and most of them are also available from microsoft.com for some older 
versions of Windows, such as XP.

-- 
Jukka Yucca Korpela, http://www.cs.tut.fi/~jkorpela/

__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-05-17 Thread Jukka K. Korpela
On Thu, 26 Apr 2007, Philippe Wittenbergh wrote:

 font-size-adjust works on the first specified font-family in the
 stylesheet.

 You know that the value for Verdana is 0.58 [1]. You specify that.
 If you have 'Verdana', no problems arise, as you say. If 'Verdana' is
 not available, the size of the font that is  actually used will be
 enlarged or reduced to match the aspect value of 'Verdana'.

I'm returning to this discussion we had a few weeks ago, since I learned a 
lot from it but did not quite get some specifics.

It seems that font-size-adjust helps in some sizing issues on Firefox 2 
(Windows) and does not hurt when it does not work, so it's reasonable to 
use it fairly often. Moreover, the new sans-serif fonts in Vista seem to 
have fairly small aspect ratios, so that there will be some problems when 
you write, say,
font-family: Calibri, Arial, sans-serif;

 The only problem I have atm is finding the aspect value for a given
 font. The font contains that information, but I haven't found an
 utility to tell me that value, nor any resource for it - especially
 for fonts I don't have.

This sounds like an odd situation. Are we all expected to find out such 
things by ourselves? I guess I can get to sufficiently accurate results 
by, say, using to copies of letter x side by side, one in Verdana in a 
very large font size, the other in the font being investigated. Then I can 
tune the font size of the latter x so that the x's are equally tall, and 
then I simply divide the font sizes and multiply the result by 0.58. But 
this is rather clumsy and boring.

-- 
Jukka Yucca Korpela, http://www.cs.tut.fi/~jkorpela/

__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-05-17 Thread Philippe Wittenbergh

On May 18, 2007, at 6:44 AM, Jukka K. Korpela wrote:

 On Thu, 26 Apr 2007, Philippe Wittenbergh wrote:

 font-size-adjust works on the first specified font-family in the
 stylesheet.

 You know that the value for Verdana is 0.58 [1]. You specify that.
 If you have 'Verdana', no problems arise, as you say. If 'Verdana' is
 not available, the size of the font that is  actually used will be
 enlarged or reduced to match the aspect value of 'Verdana'.

 I'm returning to this discussion we had a few weeks ago, since I  
 learned a
 lot from it but did not quite get some specifics.

 It seems that font-size-adjust helps in some sizing issues on  
 Firefox 2
 (Windows) and does not hurt when it does not work, so it's  
 reasonable to
 use it fairly often.
Yeah, it helps readability of articles, like when you mix two font- 
families in the same run of text:
p text text codecode/code text text/p
where code uses a monospaced font and p uses a sans-serif font.
As you say, no damage done when the browser doesn't support font-size- 
adjust. And bonus points for the browser that does support it.

 Moreover, the new sans-serif fonts in Vista seem to
 have fairly small aspect ratios, so that there will be some  
 problems when
 you write, say,
 font-family: Calibri, Arial, sans-serif;
I haven't tested those fonts yet. I have them available for install  
on my OS X machines (thanks anonymous donor), but haven't had time to  
do more. Heard good things about them, though.
---
On a slightly un-related note: may I assume that those have the full  
range of unicode characters available when installed on XP or Vista ?  
Afaik, upgrading to the latest MS Office installs those fonts.
---

 The only problem I have atm is finding the aspect value for a given
 font. The font contains that information, but I haven't found an
 utility to tell me that value, nor any resource for it - especially
 for fonts I don't have.

 This sounds like an odd situation. Are we all expected to find out  
 such
 things by ourselves? I guess I can get to sufficiently accurate  
 results
 by, say, using to copies of letter x side by side, one in Verdana  
 in a
 very large font size, the other in the font being investigated.  
 Then I can
 tune the font size of the latter x so that the x's are equally  
 tall, and
 then I simply divide the font sizes and multiply the result by  
 0.58. But
 this is rather clumsy and boring.

That is the kind of exercise I've been doing on and off - kind of  
boring as you say.
I created a two line string of text, duplicated it, and compared:
p {font-family:;}
p.adjust {font-size-adjust: value;} /* start filling in values here  
until both paragraphs match*/

Better do this at large font-sizes to really notice differences (I  
used 30px and 50px). Not scientific at all, of course.
Some people have been doing with illustrator+ photoshop, and counting  
pixels. There are  a couple of hits on the first 3-4 pages when  
asking Uncle google.

Philippe
---
Philippe Wittenbergh
http://emps.l-c-n.com




__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-04-25 Thread Jukka K. Korpela
On Tue, 24 Apr 2007, Philippe Wittenbergh wrote:

 font-size-adjust is support by Gecko (Firefox) on Windows.
 Firefox 3.0/Gecko 1.9 will have support for it on all platforms.
 Unfortunately, no other browser support it at the time of writing.

That's interesting news. But presumably there won't be any support in IE 
in the foreseeable future.

And I don't quite get the idea. The property is supposed to help in 
situations like

font-family: Verdana, Arial, sans-serif;

so that I can handle the problem that Verdana looks larger than Arial of 
the same size, so if I set the size suitable for Verdana, it will be 
too large for Arial. But it is supposed to help so that the browser 
analyzes the properties (basically, x height) of the fonts listed, the 
adjusts the font size accordingly. How can this help? If my system has 
Verdana, it will be used, and the problem does not arise. If my system has 
no font under that name, how is the browser expected to know the 
properties of the font?

So it seems that this could only help when the primarily suggested font 
(and thus its font information) _is_ available but its use has been 
prevented. This is a possible scenario, but hardly a common one.

-- 
Jukka Yucca Korpela, http://www.cs.tut.fi/~jkorpela/

__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-04-25 Thread Philippe Wittenbergh
Jukka K. Korpela wrote:

 On Tue, 24 Apr 2007, Philippe Wittenbergh wrote:

 font-size-adjust is support by Gecko (Firefox) on Windows.
 Firefox 3.0/Gecko 1.9 will have support for it on all platforms.
 Unfortunately, no other browser support it at the time of writing.

 That's interesting news. But presumably there won't be any support  
 in IE
 in the foreseeable future.

 And I don't quite get the idea. The property is supposed to help in
 situations like

 font-family: Verdana, Arial, sans-serif;

 so that I can handle the problem that Verdana looks larger than  
 Arial of
 the same size, so if I set the size suitable for Verdana, it will be
 too large for Arial. But it is supposed to help so that the browser
 analyzes the properties (basically, x height) of the fonts listed, the
 adjusts the font size accordingly. How can this help? If my system has
 Verdana, it will be used, and the problem does not arise. If my  
 system has
 no font under that name, how is the browser expected to know the
 properties of the font?

 So it seems that this could only help when the primarily suggested  
 font
 (and thus its font information) _is_ available but its use has been
 prevented. This is a possible scenario, but hardly a common one.

font-size-adjust works on the first specified font-family in the  
stylesheet.

You know that the value for Verdana is 0.58 [1]. You specify that.
If you have 'Verdana', no problems arise, as you say. If 'Verdana' is  
not available, the size of the font that is  actually used will be  
enlarged or reduced to match the aspect value of 'Verdana'.

Here is a test case. The first paragraph doesn't have a value for  
font-size-adjust specified; the second has.
http://dev.l-c-n.com/_temp/font-size-adjust.html
Remove/disable 'Verdana' on your system.
It then looks like this on my system (OS X, Gecko latest trunk build):
http://dev.l-c-n.com/_b/font-size-adjust.png

The only problem I have atm is finding the aspect value for a given  
font. The font contains that information, but I haven't found an  
utility to tell me that value, nor any resource for it - especially  
for fonts I don't have.

[1] http://www.w3.org/TR/REC-CSS2/fonts.html#propdef-font-size-adjust
http://www.w3.org/TR/css3-fonts/#font-size-props

Philippe
---
Philippe Wittenbergh
http://emps.l-c-n.com




__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


[css-d] font-family font sizes

2007-04-24 Thread David Sharp
Take the site www.syntony.org

I have specified garamond for the titles because it most closely matches 
the logo I was given, and georgia as a fall-back. However it strikes me 
that georgia appears quite a bit larger than garamond even at the same 
specified font size. Although I realise we don't have total control over 
font sizes (nor should we), however because I use a sans-serif font for 
the main text, the proportions of the type are affected.

Is there a way to specify different font sizes depending on which font 
is actually used?

regards,
D#
__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-04-24 Thread Simon White
 Take the site www.syntony.org
 
 I have specified garamond for the titles because it most 
 closely matches the logo I was given, and georgia as a 
 fall-back. However it strikes me that georgia appears quite a 
 bit larger than garamond even at the same specified font 
 size. Although I realise we don't have total control over 
 font sizes (nor should we), however because I use a 
 sans-serif font for the main text, the proportions of the 
 type are affected.
 
 Is there a way to specify different font sizes depending on 
 which font is actually used?

http://www.w3.org/TR/REC-CSS2/fonts.html#font-size-props

font-size-adjust would appear to help : quoting from the above page

In bicameral scripts, the subjective apparent size and legibility of a font are 
less dependent on their 'font-size' value than on the value of their 
'x-height', or, more usefully, on the ratio of these two values, called the 
aspect value (font size divided by x-height). The higher the aspect value, the 
more likely it is that a font at smaller sizes will be legible. Inversely, 
faces with a lower aspect value will become illegible more rapidly below a 
given threshold size than faces with a higher aspect value. Straightforward 
font substitution that relies on font size alone may lead to illegible 
characters.

For example, the popular font Verdana has an aspect value of 0.58; when 
Verdana's font size 100 units, its x-height is 58 units. For comparison, Times 
New Roman has an aspect value of 0.46. Verdana will therefore tend to remain 
legible at smaller sizes than Times New Roman. Conversely, Verdana will often 
look 'too big' if substituted for Times New Roman at a chosen size. 

The problem is whether that's supported by browsers... according to 
w3schools.com, it's in the CSS2 spec but currently unsupported 
(http://www.w3schools.com/css/css_reference.asp#font)

-Simon
__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-04-24 Thread Philippe Wittenbergh

On Apr 24, 2007, at 7:31 PM, Simon White wrote:

 Is there a way to specify different font sizes depending on
 which font is actually used?

 http://www.w3.org/TR/REC-CSS2/fonts.html#font-size-props

 font-size-adjust would appear to help : quoting from the above page

 [...]

 The problem is whether that's supported by browsers... according to  
 w3schools.com, it's in the CSS2 spec but currently unsupported  
 (http://www.w3schools.com/css/css_reference.asp#font)

font-size-adjust is support by Gecko (Firefox) on Windows.
Firefox 3.0/Gecko 1.9 will have support for it on all platforms.
Unfortunately, no other browser support it at the time of writing.

Philippe
---
Philippe Wittenbergh
http://emps.l-c-n.com




__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/


Re: [css-d] font-family font sizes

2007-04-24 Thread David Sharp
Philippe Wittenbergh wrote:
 On Apr 24, 2007, at 7:31 PM, Simon White wrote:

   
 Is there a way to specify different font sizes depending on
 which font is actually used?
   
 http://www.w3.org/TR/REC-CSS2/fonts.html#font-size-props

 font-size-adjust would appear to help : quoting from the above page
 

 font-size-adjust is support by Gecko (Firefox) on Windows.
 Firefox 3.0/Gecko 1.9 will have support for it on all platforms.
 Unfortunately, no other browser support it at the time of writing.


   
Philippe and Simon,

Thanks for the detailed replies - looks like I have a bit more reading 
to do. I'm curious as to what happens then with Safari swapping fonts to 
simulate italic text...
http://alistapart.zeldman.com/2006/11/27/safari-beats-firefox/

What if the next font on the list appears noticably different in size 
to the current? Is this something anyone else has come across?

Regards,
D#
__
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7 information -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/