[EMAIL PROTECTED] wrote:

>page w/ absolute position works fine in most or all browsers, except IE 6
>is off by one pixel right, and one pixel bottom... anybody know why?
>
>http://www.fatgraffix.com/testing/test-absolute-position/absolute-test.html
>  
>
Hi Francis,
I didn't investigate this 1px question, for I think there has te be 
thought about some (IMO) more important other things first...

Oops-Help!
If I visit the page in Internet Explorer in my screen resolution of 
1280x1024, I see very very small font sizes. And I cannot enlarge them 
in IE. - So for lots of visitors the page will be inaccessible...

    * Screenshot IE at 1280x1024.
      
http://home.tiscali.nl/developerscorner/css-discuss/images/screenshot-herbalife-1280x1024.gif
    * About 80-85% of the visitors is using IE:
      http://www.thecounter.com/stats/2006/September/browser.php
    * About 20% is surfing with a 1280x1024 screen size:
      http://www.thecounter.com/stats/2006/September/res.php
    * And more and more people are buying and using screens > 1280x1024px.
    * Then 16-17% of the visitors can have severe accessibility problems
      with the fixed font sizes of 11 px and 9 (!) px.

On the other hand, if I visit the page in a resolution of 800x600px, the 
page is walking out of the screen on the right side, and I need to use 
the horizontal scrollbar  to read what is on the page.

    * About 16% of the visitors is still using 800x600 screens:
      http://www.thecounter.com/stats/2006/September/res.php
    * Then these 16% has less usability because of the fixed width of 961px.

On the third hand, if I visit the page in other browsers than IE, in a 
resolution of 1280x1024px, I can enlarge the font size. But scaling the 
font-size upwards, already in a few steps the text is extending the 
header area.

    * Screenshot Firefox at 1280x1024, font-size client side enlarged
      with 3 steps:
      
http://home.tiscali.nl/developerscorner/css-discuss/images/screenshot-herbalife-1280x1024-enlarged.gif

All together, I agree cordially with the recommendation of the BrowserNews:

    * Recommendation.
      A good way to ensure that sites will work for as many resolutions
      as possible is to design sites to be resolution-independent, i.e.
      not to specify widths in absolute units (e.g. pixels) /unless/ a
      width is that of a fixed-width object, e.g. a GIF, JPG, or PNG image.
    * http://www.upsdell.com/BrowserNews/stat_trends.htm#res
    * As a rule of thumb, I use em's for font-sizes, and % for the width
      of a container (and: mostly no div-height! Then the height of the
      boxes can grow or shrink together with the font-sizes).
    * In the same way, I try to avoid Absolute Positioning as much as
      possible: it is fixing the elements; then they cannot shift
      upwards or downwards if needed.

To rebuild a site with all fixed styles into a flexible css is not easy. 
But everything can be made with css! [1]
So I hope you will consider to make a fresh start ... ;-)

Success and greetings,
francky

[1]
Illustration of flexibility: a page in the Belgian CSS Zen garden 
"Gigastyle":

    * http://www.gigastyle.be/?cssfile=41/elastico.css
      It is not for the beauty of the design, just a test page for
      showing possibilities of fluid css coding. You can try different
      screen sizes, different window sizes, different font-sizes, and
      different browsers.
      The page and the comments in the stylesheet are in Dutch, the css
      is universal.  :-)

Reading stuff for instance:

    * The Sizing Text pages of the css-Wiki: http://css-discuss.incutio.com/
    * The Float Layouts page of the css-Wiki:
      http://css-discuss.incutio.com/?page=FloatLayouts
    * http://www.cssliquid.com/
    * Not enough? Google for "liquid style css"!


______________________________________________________________________
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/

Reply via email to