Stylesheets - CSS Menu Rendering Problem in Opera

Author

CSS Menu Rendering Problem in Opera

Lachlan Hunt

2004-08-19, 1:21 pm

Hi,
I've been trying, but failing to work out what is causing Opera to
render my drop down menu incorrectly on my site.

http://www.lachy.id.au/

For some reason, there seems to be extra margin added below each list
item, that gets bigger for each subsequent item. Can someone please
take a look and see if they can find my problem? It works almost
perfectly in Firefox so you can see how it should look. It degrades
quite a lot in IE, there's only one hack in there to make it at least
readable.

The style sheet is:
http://www.lachy.id.au/lib/style/basic/coolblue
(It does validate, with the exception of -moz-border-radius which I'm
just experimenting with)

Finally, if someone could let me know how good or bad it looks in
Safari, that would be good too. But I'll fix those problems when I buy
my first Mac in a few weeks.

--
Lachlan Hunt
http://www.lachy.id.au/

Please direct all spam to abuse@127.0.0.1
Thank you.

Frostillicus

2004-08-19, 1:21 pm

I don't have Opera installed (I'm a Firefox man) but your site looks good.
You're not wrong about it degrading horribly in IE :-)

Which town in central NSW are you from? I'm in Tamworth (and I don't like
country music, either ;-)

Lachlan Hunt

2004-08-19, 1:21 pm

Frostillicus wrote:
> I don't have Opera installed (I'm a Firefox man) but your site looks good.

Thank you.

> You're not wrong about it degrading horribly in IE :-)

Yeah, I know. That's why I call it Internet Exploder â˜º

> Which town in central NSW are you from? I'm in Tamworth (and I don't like
> country music, either ;-)

Well, it's off topic, but ok, I'm originally from Parkes, but have since
moved to Sydney for work.

--
Lachlan Hunt â˜º
http://www.lachy.id.au/

Please direct all spam to abuse@127.0.0.1
Thank you.

Els

2004-08-19, 1:21 pm

Lachlan Hunt wrote:

> Hi,
> I've been trying, but failing to work out what is
> causing Opera to
> render my drop down menu incorrectly on my site.
>
> http://www.lachy.id.au/
>
> For some reason, there seems to be extra margin added below
> each list item, that gets bigger for each subsequent item.
> Can someone please take a look and see if they can find my
> problem?

Haven't looked at all the possibilities, but if you change:
#sitenav li { margin: 0; padding: 0; text-align: center;min-
height:100%;}
to:
#sitenav li { margin: 0; padding: 0; text-align: center;min-
height:40%;}
the problem seems to disappear (in my 'only the menu'-test).

Could well be that this has to do with one of the other
properties somewhere in the code though.

> It works almost perfectly in Firefox so you can
> see how it should look. It degrades quite a lot in IE,
> there's only one hack in there to make it at least
> readable.

In my IE6.0 WinXP, I get a 406 : Not accepted. My own local
copy does work however...
Could enhance it a tiny bit more by centering the page though
:-)

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

Lachlan Hunt

2004-08-21, 4:34 am

Els wrote:

> Lachlan Hunt wrote:
>
>
> Haven't looked at all the possibilities, but if you change:
> #sitenav li { ... min-height:100%;}
> to:
> #sitenav li { ... min-height:40%;}
> the problem seems to disappear (in my 'only the menu'-test).

Thank you, that was essentially the problem. I ended up just taking it
out, and setting the height to 2em instead. The menu now works
perfectly well in Opera, though I can't work out how to get the text for
each menu item to be vertically centered within it's block without
adding supurflous <span>s to the mark up.

The method I was using that works in Firefox:

#sitenav li a { ... padding: .4em 0; line-height: 1.2em; }

which should add to 2em, thus vertically centering within a box that is
2em high, was being miscalculated by Opera due to what looks like a
rounding error, leaving a 1px gap between each menu item.

Oh well, unless someone can suggest another method, I'll just live with
it. I tried display: table; vertical-align: middle; but it caused other
problems.

--
Lachlan Hunt
http://www.lachy.id.au/

Please direct all spam to abuse@127.0.0.1
Thank you.

Els

2004-08-21, 4:34 am

Lachlan Hunt wrote:

> Els wrote:
>
>
> Thank you, that was essentially the problem. I ended up
> just taking it out, and setting the height to 2em instead.
> The menu now works perfectly well in Opera, though I can't
> work out how to get the text for each menu item to be
> vertically centered within it's block without adding
> supurflous <span>s to the mark up.

Setting the line-height to the same as the height I think.

> The method I was using that works in Firefox:
>
> #sitenav li a { ... padding: .4em 0; line-height: 1.2em; }
>
> which should add to 2em, thus vertically centering within a
> box that is 2em high,

Ah, you did that already :-)

> was being miscalculated by Opera due
> to what looks like a rounding error, leaving a 1px gap
> between each menu item.

And what if you anticipate that error and put the line-height
to 2.1em? Or take the .4em padding out? If the line-height is
equal to the height, padding should be obsolete, I think?

What I don't get, actually, is what you are seeing in Opera.
I've just taken out the min-height:40% that I had in it, and
afaics the menu items are all vertically centered in their
respective little boxes:

http://locusmeus.com/temp/lachlan.html

(Opera 7.51)

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

Lachlan Hunt

2004-08-21, 12:22 pm

Els wrote:
> ... If the line-height is
> equal to the height, padding should be obsolete, I think?

I didn't realise that was how line-height worked â€” it's one of those
properties that has always confused me. I made the change, the padding
is gone and line-height is now 2em and it works perfectly.

> What I don't get, actually, is what you are seeing in Opera.
> I've just taken out the min-height:40% that I had in it, and
> afaics the menu items are all vertically centered in their
> respective little boxes:
>
> http://locusmeus.com/temp/lachlan.html

I thought I tried that, but it seemed to cause another strange bug
firefox. I guess I must have inadvertently made a syntax error or
something. I don't know what was causing the 1px gap in Opera, I
haven't been able to replicate it since.

Thanks again, it now works exactly the way I intended. I even figured
out how to fix the overlapping borders in Opera. So, except for the
unavoidable rounding error between the edges of the sidebar and menu,
it's all pixel-perfect in both Opera and Firefox... Safari's my next
problem, but that will have to wait until I get around to buying my
first Mac.

One thing I don't understand is what caused IE to change the way it was
rendering the menu? I had to update the IE hack so it didn't cover up
any content, but I guess it's a slight improvement for it anyway. If
you want to see what I mean, remove the width from the * html hack at
the end of the file. However, I think it's just one of those
unexplainable mysteries of the universe like nearly every other bug in IE.

--
Lachlan Hunt
http://www.lachy.id.au/

Please direct all spam to abuse@127.0.0.1
Thank you.

Els

2004-08-21, 12:22 pm

Lachlan Hunt wrote:

> One thing I don't understand is what caused IE to change
> the way it was rendering the menu? I had to update the IE
> hack so it didn't cover up any content, but I guess it's a
> slight improvement for it anyway. If you want to see what
> I mean, remove the width from the * html hack at the end of
> the file. However, I think it's just one of those
> unexplainable mysteries of the universe like nearly every
> other bug in IE.

What I still don't get, is that I still get a 406 in IE...
Your server and my IE are in war or something? The local copy
"works" in IE. (on Apache 2.0.49.0 under WinXP)

--
Els http://locusmeus.com/
Sonhos vem. Sonhos vão. O resto é imperfeito.
- Renato Russo -
Now playing: Christina Aquilera - Get Mine, Get Yours

Alan J. Flavell

2004-08-21, 12:22 pm

On Sat, 21 Aug 2004, Els wrote:

> Lachlan Hunt wrote:
>

(IE moves in wondrous ways...)
[color=darkred]
> What I still don't get, is that I still get a 406 in IE...

Are you talking about http://www.lachy.id.au/ ?

> Your server and my IE are in war or something?

That server is saying:

| Vary: negotiate,accept

which indicates (at least if all is working to plan) that the answer
is to be sought in the Accept: header that you are presenting. (I.e
that any Accept-charset, Accept-language ... which might be presented
are not germane to the issue).

If I provoke a 406 from that server, its response headers include
this:

Server: Apache/1.3.26 (Unix) Debian GNU/Linux PHP/4.1.2
Alternates: {"index.html" 1 {type text/html} {language en-au}},
{"index.xhtml" 1 {type application/xhtml+xml} {language en-au}}

which kind-of confirms that language isn't the issue, since
the only language variants available are all en-au. (This seems
to be working OK - that server is still returning the en-au variant
even if I tell it that I only accept Afrikaans.[1])

I think we need to know what Accept: header your IE is presenting
(can you find out?).

I know that mine presents this, without any interference on my part
(beyond having installed MS Office, I mean):

HTTP_ACCEPT = application/vnd.ms-excel, application/msword,
application/vnd.ms-powerpoint, image/gif, image/x-xbitmap, image/jpeg,
image/pjpeg, */*

which means that under the provisions of the wildcard (*/*) it is
willing to receive text/html -or- any of the various XHTML
content-types, even though IE has little idea what to do with those
when it gets them (some would say much the same about its reaction to
text/html, but I digress).

But that header can be doctored (I'm told, by those with a willingness
to go tinkering with the registry...). Possibly it also gets adjusted
by one or other "fix" from galactic HQ, I don't know.

cheers

[1] No special reason for choosing Afrikaans, except that it's right
at the top of the menu ;-)

Els

2004-08-21, 12:22 pm

Alan J. Flavell wrote:

> On Sat, 21 Aug 2004, Els wrote:
>
>
> (IE moves in wondrous ways...)
>
>
> Are you talking about http://www.lachy.id.au/ ?

Yes.

>
> That server is saying:
>
>| Vary: negotiate,accept
>
> which indicates (at least if all is working to plan) that
> the answer is to be sought in the Accept: header that you
> are presenting. (I.e that any Accept-charset,
> Accept-language ... which might be presented are not
> germane to the issue).

And how would I change that?

> If I provoke a 406 from that server, its response headers
> include this:
>
> Server: Apache/1.3.26 (Unix) Debian GNU/Linux PHP/4.1.2
> Alternates: {"index.html" 1 {type text/html} {language
> en-au}}, {"index.xhtml" 1 {type application/xhtml+xml}
> {language en-au}}
>
> which kind-of confirms that language isn't the issue, since
> the only language variants available are all en-au. (This
> seems to be working OK - that server is still returning the
> en-au variant even if I tell it that I only accept
> Afrikaans.[1])
>
> I think we need to know what Accept: header your IE is
> presenting (can you find out?).

How?

> I know that mine presents this, without any interference on
> my part (beyond having installed MS Office, I mean):
>
> HTTP_ACCEPT = application/vnd.ms-excel, application/msword,
> application/vnd.ms-powerpoint, image/gif, image/x-xbitmap,
> image/jpeg, image/pjpeg, */*
>
> which means that under the provisions of the wildcard (*/*)
> it is willing to receive text/html -or- any of the various
> XHTML content-types, even though IE has little idea what to
> do with those when it gets them (some would say much the
> same about its reaction to text/html, but I digress).

But this is the very first time I ever had a 406 from anywhere
in the world...

> But that header can be doctored (I'm told, by those with a
> willingness to go tinkering with the registry...).
> Possibly it also gets adjusted by one or other "fix" from
> galactic HQ, I don't know.
>
> cheers
>
> [1] No special reason for choosing Afrikaans, except that
> it's right at the top of the menu ;-)

Plus I'd understand it ;-)

--
Els http://locusmeus.com/
Sonhos vem. Sonhos vão. O resto é imperfeito.
- Renato Russo -
Now playing: Diana Ross & Supremes - Motown Hits medley

Alan J. Flavell

2004-08-21, 12:22 pm

On Sat, 21 Aug 2004, Els wrote:

> Alan J. Flavell wrote:
>
>
> And how would I change that?

You already quoted me saying:

which was supposed to mean "it should be possible to find out, but I
don't have the answer at my fingertips". OK?
[color=darkred]
>
> How?

Assuming that you've got a local version of Apache (which is always a
good idea to have around, whether you run Win32 or linux), you'd write
a little CGI script to dump the environment. I'd do it with Perl:
here's a very simple version that would suffice for this purpose

___
/

#!/usr/bin/perl -wT

print "Content-type: text/plain\n\n";

foreach $name (sort keys (%ENV)) {
print "$name = $ENV{$name}\n";
}

\___

- but don't put that script on the *public* web in that form, as it
can reveal some internal details of the web server which *might* be
useful to hostile visitors.

Then you'd visit the script's URL using the browser in question, and
look for the item which says HTTP_ACCEPT

> But this is the very first time I ever had a 406 from anywhere
> in the world...

We don't yet know whose side the anomaly is on, but without a bit of
diagnosis we aren't going to get much further.

>
> Plus I'd understand it ;-)

I think I knew that, which was why I was explaining why I had chosen
it.

Groete

Els

2004-08-21, 12:22 pm

Alan J. Flavell wrote:

>
> Assuming that you've got a local version of Apache (which
> is always a good idea to have around, whether you run Win32
> or linux), you'd write a little CGI script to dump the
> environment. I'd do it with Perl: here's a very simple
> version that would suffice for this purpose
>
> - but don't put that script on the *public* web in that
> form, as it can reveal some internal details of the web
> server which *might* be useful to hostile visitors.
>
> Then you'd visit the script's URL using the browser in
> question, and look for the item which says HTTP_ACCEPT

Did that, but my Apache isn't configured to use my Perl...
Have tried that before but didn't manage to get that to work.

So I did the uploading to the web, checked the details and
took it off again :-)

HTTP_ACCEPT = */*
HTTP_ACCEPT_ENCODING = gzip, deflate
HTTP_ACCEPT_LANGUAGE = nl

> Groete

Houdoe ;-)

--
Els http://locusmeus.com/
Sonhos vem. Sonhos vão. O resto é imperfeito.
- Renato Russo -
Now playing: Ellen Foley - We belong to the night

Lachlan Hunt

2004-08-21, 12:22 pm

Alan J. Flavell wrote:

As far as I'm concerned that's the correct way to discriminate against IE.
[color=darkred]

Perhaps my prejudice against IE has rubbed off on my server. It seems
to have developed a personality and become a /browserist/ â˜¹. I tried to
coerce it into at least sending this document [1] when it's not accepted
using:

ErrorDocument 406 /error/406

But it damn well seems insistent on further discrimination â€” it refused
to send it when I provoked it into a 406. It wasn't happy at all â˜¹.
[color=darkred]
> That server is saying:
>
> | <del>Vary: negotiate,accept</del>

<ins>Go back to Redmond, your blue e isn't welcome here! â˜º</ins>

Perhaps I should set the server to do something useful instead, like
redirecting you to Get Firefox [2] or Browse Happy [3]. That's easy
enough, I can just set this in .htaccess

ErrorDocument 406 http://getfirefox.com/

> HTTP_ACCEPT = application/vnd.ms-excel, application/msword,
> application/vnd.ms-powerpoint, image/gif, image/x-xbitmap, image/jpeg,
> image/pjpeg, */*

How did you get this information? It's strange that it doesn't include
text/html as accepted variant. However, that sort of explains why it
accepts plain text over HTML. I just tested it, but the index.txt
document served as text/plain contained HTML, and as most of you know,
it sniffed the content and rendered it as HTML. â˜º

I may aswell provide it plain text if that's what it prefers â€” it
doesn't can't handle valid HTML correctly, though some people think it
does handles XHTML betterâ€½ â˜º [4]... and I even question it's ability to
even handle plain text with that ridiculous content sniffing.

> But that header can be doctored (I'm told, by those with a willingness
> to go tinkering with the registry...). Possibly it also gets adjusted
> by one or other "fix" from galactic HQ, I don't know.

I asked the same question a few days ago in a private e-mail to Jim Ley,
and he tole me this:

# Yep, it's a registry setting.
# HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Internet
# Settings\Accepted Documents

[1] http://www.lachy.id.au/error/406
[2] http://getfirefox.com/
[3] http://browsehappy.com/
[4] http://www.lachy.id.au/blogs/nettwi...e-really-better

--
Lachlan Hunt
http://www.lachy.id.au/

Please direct all spam to abuse@127.0.0.1
Thank you.

Alan J. Flavell

2004-08-21, 12:22 pm

On Sat, 21 Aug 2004, Els wrote:

> HTTP_ACCEPT = */*
> HTTP_ACCEPT_ENCODING = gzip, deflate
> HTTP_ACCEPT_LANGUAGE = nl

Rats! I think I was fooled before by a cached copy, and on top of
that, I fooled myself by assuming that if accept-language isn't in
the Vary header, then accept-language is not germane to the issue.

I've just repeated the experiment, and I've got a 406 by setting
the Accept-language (to e.g Afrikaans, or indeed to nl).

[I'm still learning about the nooks and crannies of negotiation,
evidently. Yup, I can see why that might indeed be correct, when
in fact there's only a single language available at the server,
even though language negotiation is enabled.]

I think if you add any kind of English later in your language
selection list, his server will send you the page. This isn't quite
the behaviour that's expected in the RFC, but it's within acceptable
parameters. Pedantically, a page which is in en-au should /not/
satisfy a request for en-gb (British English), although it's OK in
satisfying a request for en (generic English). But the rfc says it's
OK to return one of the available languages if there's otherwise
nothing that matches the client's language preferences (the
alternative being indeed to return 406). Seems that whoever
configured this server decided to allow en-xx to match, but to return
406 if there was no kind of en-* amongst the available client choices.
Something like that, anyway.

In general you should add generic English later in your accept
languages, if you are willing to accept any kind of English as a last
resort (irrespective of whether you have some specific English higher
in the list).

He should however give some thought to his 406 custom error handling,
because otherwise he gets:

Additionally, a 406 Not Acceptable error was encountered while
trying to use an ErrorDocument to handle the request.

as you probably noticed.

tot ziens

Brian

2004-08-21, 12:22 pm

Alan J. Flavell wrote:

[on MSIE accept header]

> Possibly it also gets adjusted by one or other "fix" from
> galactic HQ, I don't know.

Funny, I was in Washington State recently, and I never noticed
the Death Star hovering overhead. ;-)

--
Brian (remove ".invalid" to email me)
http://www.tsmchughs.com/

Alan J. Flavell

2004-08-21, 12:22 pm

On Sat, 21 Aug 2004, Lachlan Hunt wrote:

>
> How did you get this information?

(See other posting.)

> It's strange that it doesn't include text/html as accepted variant.

You noticed! IE has to make do with it via the */* selection.

But that means (given normal Apache settings) that if the server has a
..doc, .ppt or .xls version of the page, it will get chosen in
preference to any WWW-compatible version. Which kind-of says it all,
as far as Redmond's view of the "World" is concerned, hmmm?

> I asked the same question a few days ago in a private e-mail to Jim
> Ley, and he tole me this:
>
> # Yep, it's a registry setting.
> # HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Internet
> # Settings\Accepted Documents

Thanks. I think Jim told me earlier, too, but the grey cells had
faded and I couldn't quite put my finger on it when Els asked.

Brian

2004-08-21, 7:25 pm

Lachlan Hunt wrote:

Els wrote

Alan J. Flavell wrote:[color=darkred]

[color=darkred]
> How did you get this information?

If you have access to Apache logs, just look at access logs; you may
need to configure the log to include the info you want. Otherwise, you
could try A.Flavell's useful-looking PERL script. (I tried it, then
remebered that I run Widows -- Oh yeah! -- and don't have PERL available.)

> It's strange that it doesn't include text/html as accepted variant.

Strange, or perverted, take your choice. You seem to have missed some
entertaining discussions in ciwasd about just that strange fact. Might
want to have a look.

> However, that sort of explains why it accepts plain text over HTML.

Or MS Word over text/html, as I learned when I first tried content
negotiation. MSIE's behavior was so strange to me, and buggy, and
potentially confusing to visitors, that I abandoned the idea altogether.

> I just tested it, but the index.txt document served as text/plain
> contained HTML, and as most of you know, it sniffed the content and
> rendered it as HTML.

Content sniffing is now copied in Firefox and Opera :-( albeit only
to "fix" css incorrectly served as text/plain.

--
Brian (remove ".invalid" to email me)
http://www.tsmchughs.com/

Alan J. Flavell

2004-08-21, 7:25 pm

On Sat, 21 Aug 2004, Brian wrote:

> If you have access to Apache logs, just look at access logs; you may
> need to configure the log to include the info you want.

True, You almost certainly -will- need to configure the log - this
infmration isn't logged by default (and could become quite voluminous
if you log it routinely)

> Otherwise, you could try A.Flavell's useful-looking PERL script. (I
> tried it, then remebered that I run Widows -- Oh yeah! -- and don't
> have PERL available.)

Indigoperl is a handy bundled package containing Apache, PERL and PHP.
I'm told it's just the thing for developers whose OS is in thrall to
Bill ;-) And now with new improved Apache 2.0.*

http://www.indigostar.com/indigoperl.htm

"Just unzip and run setup". But take care you're not offering Apache
services to the whole Internet ;-)

If I hadn't already, for various reasons at various times, installed
Perl (Win32 5.6 and 5.8, plus Cygwin PERL 5.8) and Apache (1.3 and
2.0) separately on the Win2K laptop, then Indigoperl is what I would
do.

Brian

2004-08-21, 7:25 pm

Alan J. Flavell wrote:

> He should however give some thought to his 406 custom error handling,
> because otherwise he gets:
>
> Additionally, a 406 Not Acceptable error was encountered while
> trying to use an ErrorDocument to handle the request.

This has to be the most unfortunately named error in the spec. Even if
language does not play into it, it's hard to create a 406 error message
that would not confuse the general public. "Not Acceptable" tends to
connote that the user did something that the server found unacceptable,
instead of that the server could not find anything acceptable to the user.

--
Brian (remove ".invalid" to email me)
http://www.tsmchughs.com/

Lachlan Hunt

2004-08-21, 7:25 pm

Brian wrote:
>
> Or MS Word over text/html...

Gee, that's secure. I could send a word document with a macro virus to
IE users... That could be fun :-) It'd probably work too, since a lot
of users don't update virus scanners either.

> Content sniffing is now copied in Firefox and Opera :-( albeit only
> to "fix" css incorrectly served as text/plain.

And Safari too, apparently:
http://diveintomark.org/archives/20...ontent-sniffing

However, sniffing for CSS served as text/plain is not as bad when it can
be determined from the context of the request that it should be CSS.
ie. when a <link> element specifies the type to be text/css. Although
it breaks the rules of the HTTP Content-Type header, it can be fairly
confidently guessed that it is meant to be CSS.

It is, however, an unfortunate result of Internet Exploder destroying
every standard in every possible way you can imagine, and thus allowing
authors and site administrators to get away with it.

Content sniffing like MSIE does, and the way Safari appears to in that
article is bad because there is no other contextual information to
obtain that information from, so the UA has no right to guess, at least
without the user's consent.

However, Mozilla does actually issue an error in the JavaScript console
when CSS is served as text/plain shows. I just tested it, though it is
a strange place for non-scripting error message.

Error: The stylesheet http://localhost/lib/style/basic/coolblue was
not loaded because its MIME type, "text/plain", is not "text/css".

Also, it didn't actually apply the style sheet for my test, though I
have seen this error before, with the style sheet applied.

--
Lachlan Hunt
http://www.lachy.id.au/

Please direct all spam to abuse@127.0.0.1
Thank you.

Brian

2004-08-21, 7:25 pm

Lachlan Hunt wrote:

> Brian wrote:
>
>
> Gee, that's secure. I could send a word document with a macro virus
> to IE users...

I hadn't thought of that. FYI, MSIE does not launch MS Word as a
separate application; instead, it opens a Word session inside the
browser chrome. That may make a differenece in security, or perhaps not.
In any case, I imagine it makes it more confusing for visitors, which is
why I turned off content negotiation on a site I maintain.

> http://diveintomark.org/archives/20...ontent-sniffing

Server does not respond. (I get that a lot with that site.)

> Mozilla does actually issue an error in the JavaScript console when
> CSS is served as text/plain shows. I just tested it, though it is a
> strange place for non-scripting error message.

I thought the same thing when I first encountered such an error.

> Error: The stylesheet http://localhost/lib/style/basic/coolblue was
> not loaded because its MIME type, "text/plain", is not "text/css".
>
> Also, it didn't actually apply the style sheet for my test,

Strange.

> though I have seen this error before, with the style sheet applied.

As have I, with the error that the stylesheet was applied after a mime
fix. Perhaps this is a "quirks" v. "standards" mode thing.

--
Brian (remove ".invalid" to email me)
http://www.tsmchughs.com/

Lachlan Hunt

2004-08-21, 7:25 pm

Brian wrote:
>
> Strange.
>
>
> As have I, with the error that the stylesheet was applied after a mime
> fix. Perhaps this is a "quirks" v. "standards" mode thing.

Yep, that it. I just did another test, and it applies it when in Quirks
Mode, but not when in Standards, or Almost Standards Mode.

>
> As long as you're in communication with Him, perhaps you could ask him
> to smite IE when he has a moment? :-)

Perhaps I'm just tired, since it's 04:00 here and I'm watching the
Olympics, but I couldn't stop laughing at that. I'll see what do
though. :-D

--
Lachlan Hunt
http://www.lachy.id.au/

Please direct all spam to abuse@127.0.0.1
Thank you.

Alan J. Flavell

2004-08-23, 7:19 pm