Suckerfish CSS drop-down menu problem (major IE inconsistency!)

Greetings!

When I use the Suckerfish drop-down menu in my new site
(http://publishing.kabis.net/), it only works in non-IE browsers.

If you look at the following page:
http://publishing.kabis.net/menu.html
in both IE as well as Mozilla, both browsers work flawlessly.

However, if you look at:
http://publishing.kabis.net/index.aspx
in both browsers, only a non-IE browser works properly. Internet
Explorer doesn't work with the drop-down menu at all. Strange thing is,
both pages use identical CSS.

That's right. Identical CSS.

So why does one page work and another doesn't?

FYI, I am using ASP.NET to dynamically bring in a random
background-image of the H1 and a random banner below the menu.

TIA!
....Geshel
--
************************************************** ********************
My reply-to is an automatically monitored spam honeypot. Do not use it
unless you want to be blacklisted by SpamCop. Please reply to my first
name at my last name dot org.
************************************************** ********************

Neo Geshel wrote:

Greetings!

When I use the Suckerfish drop-down menu in my new site
(http://publishing.kabis.net/), it only works in non-IE browsers.

If you look at the following page:
http://publishing.kabis.net/menu.html
in both IE as well as Mozilla, both browsers work flawlessly.

However, if you look at:
http://publishing.kabis.net/index.aspx
in both browsers, only a non-IE browser works properly. Internet
Explorer doesn't work with the drop-down menu at all. Strange thing is,
both pages use identical CSS.

That's right. Identical CSS.

So why does one page work and another doesn't?

FYI, I am using ASP.NET to dynamically bring in a random
background-image of the H1 and a random banner below the menu.

TIA!
...Geshel

I've just confirmed that the problem also exists in IE for the Mac.
Except in this version, the first-level links keep on shifting and
vanishing behind each other, as well as their backgrounds keep on
vanishing and switching with other backgrounds - even those backgrounds
that aren't even attached to that part of the web page!! (the
background-image for the H1 element has appeared inside the drop-down
list!!)

As with the PC version of IE, the raw suckerfish drop-down menu
(menu.html) works flawlessly. It's only when it's embedded inside my web
page that it goes nuts. As before, the CSS between the two are exact.

I believe that the problem experienced by the Mac and PC versions of IE
is related.

TIA
....Geshel
--
************************************************** ********************
My reply-to is an automatically monitored spam honeypot. Do not use it
unless you want to be blacklisted by SpamCop. Please reply to my first
name at my last name dot org.
************************************************** ********************

rf wrote:

"Neo Geshel" <gotcha (AT) geshel (DOT) org> wrote in message
news:bOiIc.73888$WB5.10031 (AT) pd7tw2no (DOT) ..

Greetings!

When I use the Suckerfish drop-down menu in my new site
(http://publishing.kabis.net/), it only works in non-IE browsers.

If you look at the following page:
http://publishing.kabis.net/menu.html
in both IE as well as Mozilla, both browsers work flawlessly.

However, if you look at:
http://publishing.kabis.net/index.aspx
in both browsers, only a non-IE browser works properly. Internet
Explorer doesn't work with the drop-down menu at all. Strange thing is,
both pages use identical CSS.

That's right. Identical CSS.

So why does one page work and another doesn't?


Because of the complexity of the page, including all javascript and other
stuff you have in there, it's nothing to do with the CSS. Hint, look atthat
top image, it reloads from the server on every mousein/mouseout. This is not
a good thing. What is doing this and why?

It has a "flicker" because of how IE handles a background-image. I am
using that to make it accessible for screen readers. If you check out
http://www.mezzoblue.com/tests/revis...e-replacement/, you will
notice that I have made use of a modified form of the Gilder/Levin
Method to make the h1 header accessible to screen readers.

It could also be because you are trying to feed XHTML to IE. IE does not
clearly understand XHTML.

True, but I have never had a problem with raw XHTML. At least, not more
than IE's buggy, non-standards-compliant rendering engine normally spits
out on a normal day anyways. I swear, IE is today what Netscape 4.x was
two years ago... a buggy and viciously horrible piece of coding that is
better left in the trash heap of history.

Why do you have an <object ...> in your <head>? This, being content,
*should* cause the browser (IE at least) to close the head and open a body.
The stuff below the object (the script elements) then become part of your
body with what sort of results?

I am trying to automatically enable cleartype smoothing for everyone
that comes to the site with NT 5.1 (Windows “XP�). Microsoft does it
themselves on their own site:
http://www.microsoft.com/typography/...pe/tuner/1.htm . At least I
use ASP.NET to filter the user-agent for all of the following: NT 5.1,
Internet Explorer, and IE version 6+. I filter for ALL of the values
because you can now install more than one version of IE on NT 5.1 at the
same time:
http://www.insert-title.com/web_desi...s/dev/multi_IE , and
I don't want someone with IE 5.x on NT 5.1 to be getting that object, as
it will just crash their browser (I know, I've tried!). If you use
another OS, another browser, or a <6 version of IE on NT 5.1, you will
not get that object item along with the web page.

And *what* produced this? It a bit like the standard cargo cult stuff but
I've never seen it this complex before.
quote
style type="text/css" media="screen"
!--/*--><![CDATA[<!--*/ h1 a#h1 { background:
url("/images/header02.jpg") no-repeat; } /*]]>*/--
/style
/quote

What you see is the XHTML-compliant method of introducing inline style
sheets. The <!CDATA[ ... ]]> is required by XHTML rules, but causes
older browsers to crash or not understand its contents (the CSS). So it
has to be contained within the fancy array: <!--/*--><![CDATA[<!--*/ ...
/*]]>*/--> so that both older and newer browsers can access the contents
with the contents remaining XHTML compliant.

Besides, the only reason that I have that chunk of CSS inline instead of
inside the style.css stylesheet is because I have the h1
background-image (your first question and my answer) dynamically
modified by ASP.NET. I am using a server-side (ASP.NET) random number
generator to randomly choose a number between two values and then place
it between "header" and ".jpg". This allows me to have a random
background-image via CSS without having to rely on JS. BTW, ASP.NET's
random image generator wouldn't have worked here because it only works
with image tags (not CSS), and I already tried calling a custom CSS file
that was a modified ASP.NET file (Example: <link type="style/css"
media="screen" rel="stylesheet"
src="/css/imageswap.aspx?img1=02&img2=05"></link>). It didn't work.It
wasn't seen by the browsers as a CSS file because it didn't end with
".css", and therefore was ignored.

Start removing stuff until only the bits to do with the menu are left and
then see what happens. When you put the removed stuff back examine exactly
why you are putting each bit back.

quote
This site makes use of standards-compliant XHTML and CSS
/quote
Why then does the page not validate at the CSS validator?

Are you talking about this link:
http://jigsaw.w3.org/css-validator/v...css/screen.css

Strange. Validated without any errors for me.

Why did you crosspost this to alt.html.critique?. If you want a critique it
will not be a pretty one :-)

alt.discuss.html removed, not on my server.

Good points, rf. But I've done my homework.

...Geshel
--
************************************************** ********************
My reply-to is an automatically monitored spam honeypot. Do not use it
unless you want to be blacklisted by SpamCop. Please reply to my first
name at my last name dot org.
************************************************** ********************

brucie wrote:

in post: <news:QirIc.77217$WB5.36619@pd7tw2no
Neo Geshel <gotcha (AT) geshel (DOT) org> said:


I am trying to automatically enable cleartype smoothing for everyone


you can just fuck off thank you very much. i'm truly amazed. how dare
you try to or think you have the right to fuck around with peoples
computers.

fuck!

not to mention cleartype has a huge system overhead and can make text
unreadable so its really hard to turn off again if you cant read the
text.

fuck again!

Nice language, Brucie. Glad to see you taking the high moral and
linguistic road.

Cleartype was designed to increase legibility on monitors of all types,
but was originally designed to produce its maximum effect on flat panel
LCD's. A decrease in legibility (especially in small type sizes) is a
clear indicator of a malfunctioning monitor or video card. If you have a
problem with ClearType's legibility on your computer, the problem lies
with your computer, not ClearType.

While it does have some system overhead, you will only notice a
performance hit on something less than a Pentium-II 233. And how many
people have XP installed on a P-II 233? Not many.

I have installed XP on several dozen computers of all types over the
last two years, and have seen what effects cleartype has on a system. I
wouldn't have enabled it if it affected the performance of more than 2%
of the XP users out there.

And turning it off is as simple as pie. Control Panel -> Display ->
Appearance -> Effects -> uncheck ClearType.

As I said, I do my homework.

PS, leave the righteous indignation for those scum that place spyware
and other nasty software on computers through Internet Explorer's
continent-sized security holes. Going after me for Cleartype is like
throwing the baby out with the bathwater.

....Geshel
--
************************************************** ********************
My reply-to is an automatically monitored spam honeypot. Do not use it
unless you want to be blacklisted by SpamCop. Please reply to my first
name at my last name dot org.
************************************************** ********************

Neo Geshel wrote:

Besides, the only reason that I have that chunk of CSS inline instead of
inside the style.css stylesheet is because I have the h1
background-image (your first question and my answer) dynamically
modified by ASP.NET. [snip] I already tried calling a custom CSS file
that was a modified ASP.NET file (Example: <link type="style/css"
media="screen" rel="stylesheet"
src="/css/imageswap.aspx?img1=02&img2=05"></link>). It didn't work. It
wasn't seen by the browsers as a CSS file because it didn't end with
".css", and therefore was ignored.

Did you try the link with the correct type="text/css" attribute? (Really?)

--
Ben M.

Ben Measures wrote:

Neo Geshel wrote:


Besides, the only reason that I have that chunk of CSS inline instead
of inside the style.css stylesheet is because I have the h1
background-image (your first question and my answer) dynamically
modified by ASP.NET. [snip] I already tried calling a custom CSS file
that was a modified ASP.NET file (Example: <link type="style/css"
media="screen" rel="stylesheet"
src="/css/imageswap.aspx?img1=02&img2=05"></link>). It didn't work. It
wasn't seen by the browsers as a CSS file because it didn't end with
".css", and therefore was ignored.


Did you try the link with the correct type="text/css" attribute? (Really?)

So shoot me. These days, I recycle the entire META section between web
projects. Here I was simply typing out the link tag from memory clear
off the top of my head and in the heat of the moment while trying to
remember a possible attribute string that the ASP.NET code would have
added to it.

It's been a year or two since I've had to type out a link tag without a
direct reference, and quite a few months since I've had to type out the
full link tag at all.

Good point, though. It just goes to show how easily a mistake can creep in.

....Geshel
--
************************************************** ********************
My reply-to is an automatically monitored spam honeypot. Do not use it
unless you want to be blacklisted by SpamCop. Please reply to my first
name at my last name dot org.
************************************************** ********************

In message <bOiIc.73888$WB5.10031@pd7tw2no>, Neo Geshel
<gotcha (AT) geshel (DOT) org> writes

Greetings!

When I use the Suckerfish drop-down menu in my new site
(http://publishing.kabis.net/), it only works in non-IE browsers.

If you look at the following page:
http://publishing.kabis.net/menu.html
in both IE as well as Mozilla, both browsers work flawlessly.

However, if you look at:
http://publishing.kabis.net/index.aspx
in both browsers, only a non-IE browser works properly. Internet
Explorer doesn't work with the drop-down menu at all. Strange thing is,
both pages use identical CSS.

That's right. Identical CSS.

So why does one page work and another doesn't?

FYI, I am using ASP.NET to dynamically bring in a random
background-image of the H1 and a random banner below the menu.

TIA!
...Geshel

As a matter of interest, what assistive technology (AT) UAs (screen
readers, talking browsers) have you tested your menu system with? It
doesn't work with IBM's HPR.

"......It is also Section 508 compliant, and conforms to Level Triple-A
of the W3C Web Accessibility Initiative...."

Incidentally, although your page validates -- IMHO, it's not WAI-AAA
compliant; and assuming that the layout would be the same on other
pages, it's not 508 compliant, either.

regards.

--
Jake

Toby Inkster wrote:

Neo Geshel wrote:


What you see is the XHTML-compliant method of introducing inline style
sheets. The <!CDATA[ ... ]]> is required by XHTML rules


Big hairy bollocks it is.

style type="text/css"
body {
background: #ccc;
color: #000;
}
/style

Perfectly valid XHTML.

Oh, really?

http://www.w3.org/TR/xhtml1/diffs.html#h-4.8

Straight from the horse's mouth, too. Problem is, the raw usage of it
causes embedded style sheets to break in older browsers. Hence the
complicated arrangement of regular <!-- --> and css /* */ comments.

As I said, I do my homework. As such, I seem to me a majority of one in
this NG. Next time, try learning XHTML from the source (the W3C), and
not from the back of a Crackerjack box. Trust me, it's for your own good.

....Geshel
--
************************************************** ********************
My reply-to is an automatically monitored spam honeypot. Do not use it
unless you want to be blacklisted by SpamCop. Please reply to my first
name at my last name dot org.
************************************************** ********************

Neo Geshel wrote:

As I said, I do my homework.

In that case, what are you doing here?!
If you know everything better than anyone here, what's the point
in coming in to ask a question?
Anything you can ask here, and that's really *anything*, could
just as well be found somewhere on the net, it must be somewhere
with the rest of your homework.

--
Els
http://locusmeus.com/
Sonhos vem. Sonhos vão. O resto é imperfeito.
- Renato Russo -

Els wrote:

Neo Geshel wrote:


As I said, I do my homework.


In that case, what are you doing here?!
If you know everything better than anyone here, what's the point
in coming in to ask a question?
Anything you can ask here, and that's really *anything*, could
just as well be found somewhere on the net, it must be somewhere
with the rest of your homework.

I might be an XHTML *guru*, but I am only a CSS *expert*. I am not a CSS
*genius* or a CSS *guru*. Hence my appeal to help when apparently
identical CSS works in one page but not in another. I didn't find when I
required on the net, so I was hoping a CSS *genius* or *guru* could take
a quick look at my CSS, and in a potential "a-ha!" moment, be able to
point out where I went wrong.

Not much to ask for, is it? So far everyone has bitched about my XHTML.
No-one has tried to take a stab at the CSS.

So, as I have said, since everyone keeps focusing on the XHTML, I can
say that I've done my homework. Because with XHTML, I have.

....Geshel
--
************************************************** ********************
My reply-to is an automatically monitored spam honeypot. Do not use it
unless you want to be blacklisted by SpamCop. Please reply to my first
name at my last name dot org.
************************************************** ********************