Presumably IE8 has a long way to go.

[dorayme]

I was getting some browsershots of a page and noticed that IE8 was
pretty different in how it rendered:

<http://dorayme.890m.com/alt/ie8/ie8.html>

I was expecting IE6 to play silly buggers but it does not do too bad. I
have not looked at why IE8 might be doing what it does or what might be
done to forestall it on this layout.

 

[Harlan Messinger]

I was getting some browsershots of a page and noticed that IE8 was
pretty different in how it rendered:

<http://dorayme.890m.com/alt/ie8/ie8.html>

I was expecting IE6 to play silly buggers but it does not do too bad. I
have not looked at why IE8 might be doing what it does or what might be
done to forestall it on this layout.

Thank goodness for

<meta http-equiv="X-UA-Compatible" content="IE=EmulateIE7" />

or, for use on an entire website, the HTTP header

X-UA-Compatible: IE=EmulateIE7

 

[+mrcakey]

Thank goodness for

<meta http-equiv="X-UA-Compatible" content="IE=EmulateIE7" />

or, for use on an entire website, the HTTP header

X-UA-Compatible: IE=EmulateIE7

Good find!

+mrcakey

 

[dorayme]

Thank goodness for

<meta http-equiv="X-UA-Compatible" content="IE=EmulateIE7" />

With the slash, yes, it is fine except for the niggle over formal
invalidity. Putting your meta statement into an IE comment conditional
fixes all as far as I see.

or, for use on an entire website, the HTTP header

X-UA-Compatible: IE=EmulateIE7

Now this is more like what I would want. Where my files using this sort
of layout are destined I have little control of the public server but I
can ask the administrators to do something. I might tell them to do what
you have indicated or I might just stick in the conditional into an
include. It is a Windows server. (I must try to get this changed at some
stage, it causes me extra work! I will try to persuade the client to
host elsewhere)

 

[Harlan Messinger]

With the slash, yes, it is fine except for the niggle over formal
invalidity. Putting your meta statement into an IE comment conditional
fixes all as far as I see.

You're also most welcome to exclude the slash.

Now this is more like what I would want. Where my files using this sort
of layout are destined I have little control of the public server but I
can ask the administrators to do something. I might tell them to do what
you have indicated or I might just stick in the conditional into an
include.

This option, I had just come across before mentioning it here and my
comment just relayed what I'd read. Certainly you can also do this
per-page. In ASP.NET: Response.AddHeader("X-UA-Compatible",
"IE=EmulateIE7").

 

[dorayme]

You're also most welcome to exclude the slash.

Thank you. I had anticipated this offer - perhaps a little
presumptively! I see now that I seem not to have said that without the
slash IE8 does not behave better on my tests. Maybe I was shy of saying
this because it might reveal too greatly some terrible ignorance.

You can see my trials at

<http://dorayme.890m.com/alt/ie8/>

And you can see the results via

<http://browsershots.org>

I am grateful for your input because it was niggling me about launching
a new section of a commercial website if it was going to look bad in
some major browser. I have no idea how popular IE8 is already or whether
they will fix things. I don't trust MS on IE and my only hope is that we
have a man on the inside (Gerard Talbot! He has been communicating with
those guys as I understand it and at least that is cause for hope. In my
opinion, he should go to whatever meetings he goes to with them,
equipped with an equaliser like Dirty Harry so they take notice...)

 

[cwdjrxyz]

Thank you. I had anticipated this offer - perhaps a little
presumptively! I see now that I seem not to have said that without the
slash IE8 does not behave better on my tests. Maybe I was shy of saying
this because it might reveal too greatly some terrible ignorance.

You can see my trials at

<http://dorayme.890m.com/alt/ie8/>

And you can see the results via

<http://browsershots.org>

Thanks for the link to browshots.org. Since it has so many browsers
and 3 OSs I decided to see which browsers on which OSs can handle true
xhtml served properly as application/xhtml+xml. I used my page served
as true xhtml at http://www.cwdjr.info/test/formtestX.xhtml for the
test. If the page works , you should see a form box which works, but
takes you nowhere. I only tested the most recent version of a browser
on an OS when more than one was available.

Linux OS Browsers That Support True Xhtml:
Konqueror 3.5 ; Firefox 3.0.1 ; Opera 9.5.1 ; Epiphany 2.22.3 ; Flock
2.0 ; Galeon 2.0.6 ; Iceape 1.1.11 ; Iceweasel 3.0 ; Kazehakase
0.5.5 ; Minefield 3.1 ; Shivetoke 3.1 .

MAC OS Browsers That Support True Xhtml:
Safari 3.1 ; Firefox 3.0

Windows OS Browsers That Support True Xhtml:
Minefield 3.1 ; K Melon 1.1.6 ; Firefox 3.0 ; Flock 2.0 ; Opera 9.51;
Safari for Windows 3.1 ; Seamonkey 2.0

Non-Support of true Xhtml:
IE7 and all lower version will not support true xhtml. The Dillo 0.8.6
Browser on a Linux OS shows a blank page. The IE8 browser would not
load my page, even after extending the time to over 1 hour.

So far as I can tell, IE8 has not been released yet as an official
stable version, and the Windows download site for their XP OS, at
least in the US, is not yet offering IE8. Thus there could still be
changes in the final version. Also I would want to view my page
directly on a computer that had IE 8 installed. But even just
considering the IE 7, it is apparent that IE is out of tune with
nearly all other recent popular browsers, and many rather obscure
ones, when it comes to support of true xhtml. I can think of only 2
possible reasons for this. 1. The programmers at Microsoft who write
browser programs are rather stupid or 2. The administration at
Microsoft does not care about standards support or even perhaps does
not want it, even though they likely donate more funds to the W3C than
anyone else. Can anyone think of another reason why Microsoft has been
so tardy in true xhtml support?

 

[Chaddy2222]

Thank you. I had anticipated this offer - perhaps a little
presumptively! I see now that I seem not to have said that without the
slash IE8 does not behave better on my tests. Maybe I was shy of saying
this because it might reveal too greatly some terrible ignorance.

You can see my trials at

<http://dorayme.890m.com/alt/ie8/>

And you can see the results via

<http://browsershots.org>

I am grateful for your input because it was niggling me about launching
a new section of a commercial website if it was going to look bad in
some major browser. I have no idea how popular IE8 is already or whether
they will fix things. I don't trust MS on IE and my only hope is that we
have a man on the inside (Gerard Talbot! He has been communicating with
those guys as I understand it and at least that is cause for hope. In my
opinion, he should go to whatever meetings he goes to with them,
equipped with an equaliser like Dirty Harry so they take notice...)

Hi dorayme.
Mate far as I am aware I don't think IE 8 is out of beta yet. So it
might be a while before all the bugs are sorted out.
Or perhaps I should write, it might be a while before all the
"features" are added.

 

[Chaddy2222]

Thanks for the link to browshots.org. Since it has so many browsers
and 3 OSs I decided to see which browsers on which OSs can handle true
xhtml served properly as application/xhtml+xml. I used my page served
as true xhtml at http://www.cwdjr.info/test/formtestX.xhtml for the
test. If the page works , you should see a form box which works, but
takes you nowhere. I only tested the most recent version of a browser
on an OS when more than one was available.

Linux OS Browsers That Support True Xhtml:
Konqueror 3.5 ; Firefox 3.0.1 ; Opera 9.5.1 ; Epiphany 2.22.3 ; Flock
2.0 ; Galeon 2.0.6 ; Iceape 1.1.11 ; Iceweasel 3.0 ; Kazehakase
0.5.5 ; Minefield 3.1 ; Shivetoke 3.1 .

MAC OS Browsers That Support True Xhtml:
Safari 3.1 ; Firefox 3.0

Windows OS Browsers That Support True Xhtml:
Minefield 3.1 ; K Melon 1.1.6 ; Firefox 3.0 ; Flock 2.0 ; Opera 9.51;
Safari for Windows 3.1 ; Seamonkey 2.0

Non-Support of true Xhtml:
IE7 and all lower version will not support true xhtml. The Dillo 0.8.6
Browser on a Linux OS shows a blank page. The IE8 browser would not
load my page, even after extending the time to over 1 hour.

So far as I can tell, IE8 has not been released yet as an official
stable version, and the Windows download site for their XP OS, at
least in the US, is not yet offering IE8. Thus there could still be
changes in the final version. Also I would want to view my page
directly on a computer that had IE 8 installed. But even just
considering the IE 7, it is apparent that IE is out of tune with
nearly all other recent popular browsers, and many rather obscure
ones, when it comes to support of true xhtml. I can think of only 2
possible reasons for this. 1. The programmers at Microsoft who write
browser programs are rather stupid or 2. The administration at
Microsoft does not care about standards support or even perhaps does
not want it, even though they likely donate more funds to the W3C than
anyone else. Can anyone think of another reason why Microsoft has been
so tardy in true xhtml support?

Cause it a waste of time and money, especially when HTML 4.01 Strict
works fine. You can also use XML to create HTML so that argument is
also not a good one.

 

[dorayme]

Hi dorayme.
Mate far as I am aware I don't think IE 8 is out of beta yet. So it
might be a while before all the bugs are sorted out.
Or perhaps I should write, it might be a while before all the
"features" are added.

Thanks Chaddy, I am relieved to hear confirmation of this. Since the
meta avoider has now come to my attention via Harlan, and at least
browsershot screenshot tells me it works, I probably will chuck it in
with a conditional for ie8 (via an include for other things to make it
easy).

 

[Chaddy2222]

Thanks Chaddy, I am relieved to hear confirmation of this. Since the
meta avoider has now come to my attention via Harlan, and at least
browsershot screenshot tells me it works, I probably will chuck it in
with a conditional for ie8 (via an include for other things to make it
easy).

That sounds like the way to go. I use PHP includes for just about
everything now a days.
In fact I am not quite sure how I lived before discovering PHP
includes, or PHP in general really.

 

[Mark A. Boyd]

cwdjrxyz posted in alt.html:

So far as I can tell, IE8 has not been released yet as an official
stable version, and the Windows download site for their XP OS, at
least in the US, is not yet offering IE8. Thus there could still be
changes in the final version. Also I would want to view my page
directly on a computer that had IE 8 installed. But even just
considering the IE 7, it is apparent that IE is out of tune with
nearly all other recent popular browsers, and many rather obscure
ones, when it comes to support of true xhtml. I can think of only 2
possible reasons for this. 1. The programmers at Microsoft who write
browser programs are rather stupid or 2. The administration at
Microsoft does not care about standards support or even perhaps does
not want it, even though they likely donate more funds to the W3C than
anyone else. Can anyone think of another reason why Microsoft has been
so tardy in true xhtml support?

Or 3. MS buearocracy gets in the way of nearly everything the programmers
want/need to do.

I suspect a combination of 2 and 3 with *heavy* emphasis on 2.

Just look how long it's taken them to support 32bit .png files in IE. More
than a decade.

 

[Neredbojias]

cwdjrxyz posted in alt.html:


Or 3. MS buearocracy gets in the way of nearly everything the programmers
want/need to do.

I suspect a combination of 2 and 3 with *heavy* emphasis on 2.

Just look how long it's taken them to support 32bit .png files in IE. More
than a decade.

Whatever the reason, when you lose touch with what the people want, you're
headed for doom. Microsoft's biggest problem is their aloofness.

 

[Harlan Messinger]

Thank you. I had anticipated this offer - perhaps a little
presumptively! I see now that I seem not to have said that without the
slash IE8 does not behave better on my tests. Maybe I was shy of saying
this because it might reveal too greatly some terrible ignorance.

You can see my trials at

<http://dorayme.890m.com/alt/ie8/>

I just tried these out in IE8 and found they were identical--and looked
fine, as far as I could tell. Then I looked at the source code and
realized I'd forgotten to mention one important point: IE requires that
the meta tag be the very first tag to appear in the head section of the
document. Egotistical little tag, isn't it?

 

[BootNic]

On Mon, 18 Aug 2008 10:07:02 -0400

dorayme wrote:
dorayme wrote:
I was getting some browsershots of a page and noticed that IE8 was
pretty different in how it rendered: [snip]

<meta http-equiv="X-UA-Compatible" content="IE=EmulateIE7" />

http://msdn.microsoft.com/en-us/library/cc288325(VS.85).aspx

[snip]

I just tried these out in IE8 and found they were identical--and looked
fine, as far as I could tell. Then I looked at the source code and
realized I'd forgotten to mention one important point: IE requires that
the meta tag be the very first tag to appear in the head section of the
document. Egotistical little tag, isn't it?

I think it is really just too early to try and make IE 8 work. It's
just still beta and still too buggy.

At most I would suggest if one has conditional comments, to limit the
comments to [if lte IE 7], that is if there are any [if IE] then change
them to [if lte IE 7].

--

BootNic Mon Aug 18, 2008 11:01 am
I thought you should know that you didn't invent anything new.
*Bergamot*

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)

iEYEARECAAYFAkipjt8ACgkQylMUzZO6jeI2ygCeOlWtye78wVFoScPfE1B9aLuX
FuoAniQBPn7/JFoJafzcSR3iofI5P6qq
=A4l/
-----END PGP SIGNATURE-----

 

[Harlan Messinger]

With the slash, yes, it is fine except for the niggle over formal
invalidity. Putting your meta statement into an IE comment conditional
fixes all as far as I see.

Wait--why is it invalid? Both HTML 4.01 Strict and XHTML 1.0 Strict
provide for a meta tag with both http-equiv and content attributes.

http://www.w3.org/TR/html4/struct/global.html#edef-META
http://www.w3.org/TR/xhtml1/dtds.html#dtdentry_xhtml1-strict.dtd_meta

 

[Sherm Pendley]

Wait--why is it invalid?

As I'm sure you're aware, we can't properly serve XHTML to IE. IE will
parse this as HTML, in which the slash is invalid.

sherm--

 

[Harlan Messinger]

As I'm sure you're aware, we can't properly serve XHTML to IE. IE will
parse this as HTML, in which the slash is invalid.

Well, yes, but we're mixing modes here. The browser neither knows nor
cares if the code is valid, and if the browser is treating the code as
HTML, then the trailing slash will be ignored just fine, as it will be
treated as a bad attribute, and the spec says the user agent should
gracefully ignore a bad attribute.

As for validating the code whether to satisfy oneself or to fulfill some
requirement or to follow good practice, if you want it to validate as
HTML, drop the slash, and if you want it to validate as XHTML, leave the
slash in. Either way, the tag will not prevent the document from passing
validation.

 

[dorayme]

Well, yes, but we're mixing modes here. The browser neither knows nor
cares if the code is valid, and if the browser is treating the code as
HTML, then the trailing slash will be ignored just fine, as it will be
treated as a bad attribute, and the spec says the user agent should
gracefully ignore a bad attribute.

As for validating the code whether to satisfy oneself or to fulfill some
requirement or to follow good practice, if you want it to validate as
HTML, drop the slash, and if you want it to validate as XHTML, leave the
slash in. Either way, the tag will not prevent the document from passing
validation.

Before your post saying "IE requires that the meta tag be the very first
tag to appear in the head section of the document." my tests showed my
original test page (in 4.01 Strict) with your suggested tag, displayed
properly in IE8 when the slash was there but *not* without it. I also
noticed validators being unhappy. With the slash, and it *not* being the
the very first tag, it displayed in IE 8 like in IE7 or FF, at least in
my tests via browsershots.

The validation is no problem as I mentioned in a later post. What I have
not done yet is test to see if a slashless tag that is the first tag
works in IE8.

 

[cwdjrxyz]

dorayme wrote:




Before your post saying "IE requires that the meta tag be the very first
tag to appear in the head section of the document." my tests showed my
original test page (in 4.01 Strict) with your suggested tag, displayed
properly in IE8 when the slash was there but *not* without it. I also
noticed validators being unhappy. With the slash, and it *not* being the
the very first tag, it displayed in IE 8 like in IE7 or FF, at least in
my tests via browsershots.

The validation is no problem as I mentioned in a later post. What I have
not done yet is test to see if a slashless tag that is the first tag
works in IE8.

The meta must be self closed (<blah />) in xhtml as must img etc,
since everything must be closed in xhtml. If you serve the xhtml page
properly as application/xhtml+xml the xml parser of the browser rather
than the html one is used. Then the usual response is to give you an
error message rather than a view of the page. If you mis-serve an
xhtml page as html (text/html) then the self closing tag is unknown to
html. However the html parser of the browser that is used in such a
case lets you get away with much sloppy and invalid code. Some may
result in error messages or warnings when you validate at W3C, but not
an error message from the browser parser instead of a view of the
page. An html page can be, and often is, extremely sloppy with dozens
of validation errors, but it can still be viewed, although it may not
be exactly what you had in mind. Now when it comes to Microsoftese,
one can hardly be surprised at any response on an IE browser.

If you are going to use a php include for header exchange of some
sort, it must be at the very top of the page. If this include is used,
you may specify most things you would use meta tags for in it and not
include meta tags on the page, but usually you can do what is most
convenient for you.