Previous Message
Next Message

Gilder/Levin IR Method IE5mac Solution

Sent by Michael Ryznar on 12 March 2004 07:07


I spent some time studying the various IR (Image Replacement) Techniques 
and decided upon using the Gilder/Levin IR Method based on a number of 
factors. It seems to be the only method that works when css is turned on 
and images are turned off in a browser. CSS-on/images-off probably 
doesn't happen often but it is good to know that the IR method would 
function in those cases. As well, it is easy to add links to a replaced 
header image using the Gilder/Levin IR Method.

Unfortunately there were a few problems with the technique when viewed 
in IE5mac. Any following elements would slide under the replaced 
heading. Here is a solution I came up with [1] as well as a few other 
modifications to the code to make it compatible in IE5mac.

The only disadvantage to the Gilder/Levin IR Method is that it doesn't 
work with transparent images. (For IR with transparent images, LIR [3] 
is probably the best IR method to use in those cases IMHO.) But like all 
other IR methods, except the Gilder/Levin IR Method, the LIR technique 
fails in CSS-on/images-off situations. This is why the Gilder/Levin IR 
Method might be the most successful IR method (IMHO).

There are about three versions of the Gilder/Levin IR Method on the 
internet. This one [2] is Levin's most successful version. I based my 
code on this version.

I worked at this for some time trying to understand IE5mac's strange 
behaviour of not respecting the height value on h.x tags which seems to 
be the central root of this problem. It would seem IE5mac can't handle 
height values in h.x tags that contain absolute spans - unless the 
absolute spans come first in the h.x tag. IE5mac never has this problem 
with divs - its only h.x tags that it chokes on (but only when the h.x 
tags contain absolute spans).

Could anyone please let me know if they find any problems with this 
solution. It seems to work in IE5mac on both OS9 and OSX as well as Mozilla.

Steve Clay did some IR tests too [3] which are very interesting to 
study. Thanks for this Steve.


[1] http://www.ryznardesign.com/pages/web_page_coding/gilder_levin_ir_mac/

[2] http://levin.grundeis.net/files/20030809/alternatefir.html
[3] http://www.kryogenix.org/code/browser/lir/


--
Mike Ryznar


Ryznar Design
Burnaby, British Columbia
Canada

http://www.ryznardesign.com

______________________________________________________________________
css-discuss [EMAIL-REMOVED]]
http://www.css-discuss.org/mailman/listinfo/css-d
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/
Previous Message
Next Message