Thanks to someone's posting here, I discovered this URL:

http://browsershots.org/http://imaginatori...g/shop/misc.htm

Is this as good as any service to use? Or am I missing anything else?

Hmm, the most obvious "error" on this page is that MSIE6 gets just the top 'panel' wrong. I wonder if this is an artefact of the browserfarm? Could anyone with a "genuine" MSIE6 check to see if the top panel comes out the same as the others?

The page is generated by a program, so obviously the structure is exactly the same. I'm using a negative margin to make the white background belonging to the heading text extend down under the thumbnails... (There is one CSS error, which I will fix, but I don't think it's related. See http://jigsaw.w3.org/css-validator/validat...g=1&lang=en


DIV.pageptr h5 {
margin : 0 0 -60px;
border : none;
border-top : 2px solid #0040b0;
padding : 0.3em 1em 65px;
background-color : #f0f0f0;
color : #0040b0;
font-family : Arial, sans-serif;
font-size : 1.2em;
text-align : left;
}

DIV.pageptr P {
margin : 0.2em 0 -60px;
border : none;
padding : 0 1em 65px;
border : none;
background-color : #f0f0f0;
font-family : Arial, sans-serif;
font-size : 1em;
text-align : left;
}

Using Internet Explorer 6 under Windows 2000 I see solid white behind all the panels. Likewise using I.E. 7 under Windows XP. Other browsers I've tried show white behind the top half of each panel.

Thanks...

Can you please clarify: looking at the Browser shots page almost all of them look as I intended: the white bit being the background to the <hN> and <p> bits forming the heading, which sticks out by the negative margin, down to about the middle of the thumbnails.

http://browsershots.org/http://imaginatori...g/shop/misc.htm

The ones that look not quite right are: several I've never heard of, Opera 7 something, and MSIE6. Are you saying that the real MSIE6 looks not like the browser shot, but hase every row "wrong"?

I haven't done much comparative testing: I have been trying to keep things simple, and checking they work in a couple of standards browsers (Opera and Firefox). Is this about standard for Microsoft's efforts?

Are there any well-known standard M$ bugs that would obviously cause negative margin tricks to go wrong?

My first instinct is that I really ought to add one of those "Works best on any standards-compliant browser" comments. If MSIE6 gets this wrong it isn't really a big deal, since it doesn't affect legibility.