You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lenya.apache.org by Jürgen Ragaller <ra...@apache.org> on 2007/11/15 16:02:14 UTC
wrapping of tabbed areas
Hi
| svn-commit: Use display:table-cell instead of float:left for tabbed
areas to avoid wrapping
This now looks nice in firefox - but IE (6 and 7) ignore
display:table-cell
I'll do some investigation about how to avoid the wrapping in IE.
BTW: IE6 still has way too big fonts for the authoring area (and the
default publication) - the reason is the xml Prolog (let's ie 6
switch to quirks mode).
What about removing the prolog as long as ie 6 is still around in
relevant numbers?
Jürgen
null-oder-eins GmbH Zürich
web & graphic design
www.null-oder-eins.ch
ragaller@apache.org
Skype: callto://ragaller
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Juergen Ragaller <ra...@apache.org>.
Hi Jörn
Am 20.11.2007 um 20:56 schrieb Jörn Nettingsmeier:
> Jürgen Ragaller wrote:
>> Hi, Jörn
>> Am 16.11.2007 um 13:01 schrieb Jörn Nettingsmeier:
>>>
>>> over my very dead corpse.
>>> if a certain software vendor cannot be bothered to provide the
>>> most trivial bugfixes and read a f"$§%ing spec, that's really not
>>> our problem.
>>> anyone with half a brain can download and understand the relevant
>>> standards documents, and if the world's largest pile of software
>>> engineers can't be bothered, well, tough luck for their users.
>> Just to make shure we don't misunderstand each other on this one.
>> The removing of the xml declaration (remove only <?xml
>> version="1.0" encoding="UTF-8"?>) would not make the page invalid:
>> http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.webstandards.org%2F
>
> sigh. you are of course right, the xml declaration is not mandatory.
> but it does provide a very unambigous hint as to xml version and
> encoding that makes a whole lot more sense than those grafted-on
> <meta/> tags with their "; content-encoding=FOO". and there is
> absolutely no excuse whatsoever to change one's parsing behaviour
> depending on this processing instruction. if it were, the
> instruction should say <?IE createAFsckingMessOfMyCode="yes"?>.
>
> well, i don't want to punish IE users (although somebody should do
> it, someday), and much less fellow admins who have to get their jobs
> done. if it can be demonstrated that the xml declaration can be
> omitted without the slightest bit of regression in xml-conformant
> browsers (including the encoding setting), then well, let's omit it
> and wait for microsoft and the last of their broken browsers to die
> of old age. should be any day now :)
>
>
> or maybe we could resurrect a server-side browser selector and use
> two different xhtml serializers. but this would imply having a
> centralized final transformation step in the global sitemap, unless
> we want to duplicate this selector all over the place. i have put a
> patch in bugzilla a while ago that tries to accomplish this, but it
> does so by breaking all badly modularized features (bxe, webdav and
> some other ad-hoc stuff currently in the default pub sitemap).
We avoid the xml declaration for all our websites - I am not aware of
any side effects.
My suggestion is to leave the already inserted ie6hacksonly.css in for
2.0 and remove the xml declaration and the quirks mode override part
of ie6hacksonly.css. for 2.0.1.
Jürgen
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jörn Nettingsmeier <ne...@apache.org>.
Jürgen Ragaller wrote:
> Hi, Jörn
>
> Am 16.11.2007 um 13:01 schrieb Jörn Nettingsmeier:
>>
>> over my very dead corpse.
>> if a certain software vendor cannot be bothered to provide the most
>> trivial bugfixes and read a f"$§%ing spec, that's really not our problem.
>> anyone with half a brain can download and understand the relevant
>> standards documents, and if the world's largest pile of software
>> engineers can't be bothered, well, tough luck for their users.
>
> Just to make shure we don't misunderstand each other on this one.
> The removing of the xml declaration (remove only <?xml version="1.0"
> encoding="UTF-8"?>) would not make the page invalid:
>
> http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.webstandards.org%2F
sigh. you are of course right, the xml declaration is not mandatory.
but it does provide a very unambigous hint as to xml version and
encoding that makes a whole lot more sense than those grafted-on <meta/>
tags with their "; content-encoding=FOO". and there is absolutely no
excuse whatsoever to change one's parsing behaviour depending on this
processing instruction. if it were, the instruction should say <?IE
createAFsckingMessOfMyCode="yes"?>.
well, i don't want to punish IE users (although somebody should do it,
someday), and much less fellow admins who have to get their jobs done.
if it can be demonstrated that the xml declaration can be omitted
without the slightest bit of regression in xml-conformant browsers
(including the encoding setting), then well, let's omit it and wait for
microsoft and the last of their broken browsers to die of old age.
should be any day now :)
or maybe we could resurrect a server-side browser selector and use two
different xhtml serializers. but this would imply having a centralized
final transformation step in the global sitemap, unless we want to
duplicate this selector all over the place. i have put a patch in
bugzilla a while ago that tries to accomplish this, but it does so by
breaking all badly modularized features (bxe, webdav and some other
ad-hoc stuff currently in the default pub sitemap).
--
Jörn Nettingsmeier
"One of my most productive days was throwing away 1000 lines of code."
- Ken Thompson.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jürgen Ragaller <ra...@apache.org>.
Hi, Jörn
Am 16.11.2007 um 13:01 schrieb Jörn Nettingsmeier:
>
> over my very dead corpse.
> if a certain software vendor cannot be bothered to provide the most
> trivial bugfixes and read a f"$§%ing spec, that's really not our
> problem.
> anyone with half a brain can download and understand the relevant
> standards documents, and if the world's largest pile of software
> engineers can't be bothered, well, tough luck for their users.
Just to make shure we don't misunderstand each other on this one.
The removing of the xml declaration (remove only <?xml version="1.0"
encoding="UTF-8"?>) would not make the page invalid:
http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.webstandards.org%2F
The patching is possible but is/was an unpleasing experience and will
be for another 2 to 3 years (thrown back to when we were supporting
ie5 that's essentially what ie6 quirks-on disbehaves like).
Here's my take on this
http://issues.apache.org/bugzilla/attachment.cgi?id=21152&action=view
Jürgen
null-oder-eins GmbH Zürich
web & graphic design
www.null-oder-eins.ch
ragaller@apache.org
Skype: callto://ragaller
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jürgen Ragaller <ra...@apache.org>.
Am Nov 16, 2007 um 16:29 schrieb solprovider@apache.org:
>
> I dislike code based on brower-checking because maintenance and
> testing becomes difficult. Simple CSS updates need to change several
> files; did the newly hired developer remember?
I feel different about this. Keeping hacks (only hacks) separated is
IMO a healthy thing. The readability of the main css file is good as
it does not contain any hacks (pure valid css). The hack-stylesheet is
separated and can be deleted when a browser becomes obsolete.
Remembering hacks and there consequences inside a stylesheet is hard
(and the hack removal might even influence the modern compliant
browsers as they «see» the hacks as well).
The separation using conditional comments is especially handy when it
comes to testing - it is well defined which browser reads the
stylesheet. And for non Microsoft-Browsers... only a comment.
Jürgen
Jürgen Ragaller
null-oder-eins GmbH Zürich
ragaller@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by so...@apache.org.
On 11/16/07, Jürgen Ragaller <ra...@apache.org> wrote:
> Am Nov 16, 2007 um 13:39 schrieb Jörn Nettingsmeier:
> > solprovider@apache.org wrote:
> >> On 11/16/07, Jörn Nettingsmeier <ne...@apache.org> wrote:
> >>> if a certain software vendor cannot be bothered to provide the most
> >>> trivial bugfixes and read a f"$§%ing spec, that's really not our
> >>> problem. anyone with half a brain can download and understand the
> >>> relevant standards documents, and if the world's largest pile of
> >>> software engineers can't be bothered, well, tough luck for their
> >>> users.
> >> That certain software vendor's buggy software is still rather popular
> >> (>80%?) Version 6 is still popular because version 7 is even more
> >> buggy.
> solprovider: I disagree here - ie7 is by far less buggy that ie6 - the
> list of bugs that were squashed (in coordination with the webstandards
> group is impressive - well ie6 was in fact impressingly buggy).
IE7 meets standards better than IE6. I think somebody at Microsoft
may have downloaded W3C's HTML and CSS specifications in 2002. I used
"buggy" meaning "crashes often" rather than "not meeting standards."
The most stubborn early-adopting Microsoft-addicted technical person I
know finally surrendered and stopped using IE7 in August so my data is
a few months old. (Greg could not get Vista to use IE6 without
crashing so he installed Firefox! He claims this will be temporary.)
> >> From the evidence, none of the company's programmers have half
> >> a brain, and few are able to read. If our project wants to be used
> >> for websites, we need to make certain our software works with that
> >> buggy non-standard web browser.
> true - and we're close - the fixes for ie6 probably not so hard to do.
IE6 is still the leader. IE6, IE7, and Firefox2 are currently used by
90% of the public. We should not release something that does not
display well in IE6. (That changes if the issue is only with the
adminstration screens because we can force admins to use a modern
standards-following browser.)
> > Not at the cost of standards conformance and proper validation. a
> > stupid bug is a stupid bug no matter the market share. people
> > putting up with this kind of shit have made web designer's lives
> > miserable for ten years now.
> > it's not as if we'd lose those users by being picky about our code.
> I agree in principle - but leaving out the xml-declaration would not
> break xhtml validation.
> But ok - let's leave it in and I'll provide a IE 6 conditional comment
> inserted lessthenequalie6hacksonly.css.
> ... can't do that at home - so I'll start with it next week (no
> windows at home)
> >>> if anyone wants to provide a patch with an IE-specific conditional
> >>> comment, that's fine with me. clean code for the rest of us, and
> >>> it's
> >>> plain to see it's an extra bugfix for IE.
> >> Special CSS for Microsoft Internet Explorer is easy. The browser is
> >> so non-standard that equal signs are treated as colons. All browsers
> >> use the last setting found if the same property is repeated. Try
> >> something like:
> >> display: table-cell;
> >> display=block;
> >> float=left;
> I prefer the separation of style sheet documents
I dislike code based on brower-checking because maintenance and
testing becomes difficult. Simple CSS updates need to change several
files; did the newly hired developer remember?
> > at the risk of sounding like a broken record: this is invalid css.
> > (oh yes, i validate mine :)
> > i don't want to see validation warnings all over the place that
> > obscure the real problems. imnsho, the *only* acceptable way to hack
> > around M$' standards dyslexia are conditional comments. preferably
> > by including files named
> > "really_stupid_microsoft_fuckup_workaround.css".
> > Jörn Nettingsmeier
> Jürgen Ragaller
solprovider
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jürgen Ragaller <ra...@apache.org>.
Hi
Am Nov 16, 2007 um 13:39 schrieb Jörn Nettingsmeier:
> solprovider@apache.org wrote:
>> On 11/16/07, Jörn Nettingsmeier <ne...@apache.org> wrote:
>>> if a certain software vendor cannot be bothered to provide the most
>>> trivial bugfixes and read a f"$§%ing spec, that's really not our
>>> problem. anyone with half a brain can download and understand the
>>> relevant standards documents, and if the world's largest pile of
>>> software engineers can't be bothered, well, tough luck for their
>>> users.
>> That certain software vendor's buggy software is still rather popular
>> (>80%?) Version 6 is still popular because version 7 is even more
>> buggy.
solprovider: I disagree here - ie7 is by far less buggy that ie6 - the
list of bugs that were squashed (in coordination with the webstandards
group is impressive - well ie6 was in fact impressingly buggy).
>> From the evidence, none of the company's programmers have half
>> a brain, and few are able to read. If our project wants to be used
>> for websites, we need to make certain our software works with that
>> buggy non-standard web browser.
true - and we're close - the fixes for ie6 probably not so hard to do.
> Not at the cost of standards conformance and proper validation. a
> stupid bug is a stupid bug no matter the market share. people
> putting up with this kind of shit have made web designer's lives
> miserable for ten years now.
> it's not as if we'd lose those users by being picky about our code.
I agree in principle - but leaving out the xml-declaration would not
break xhtml validation.
But ok - let's leave it in and I'll provide a IE 6 conditional comment
inserted lessthenequalie6hacksonly.css.
... can't do that at home - so I'll start with it next week (no
windows at home)
>
>
>>> if anyone wants to provide a patch with an IE-specific conditional
>>> comment, that's fine with me. clean code for the rest of us, and
>>> it's
>>> plain to see it's an extra bugfix for IE.
>> Special CSS for Microsoft Internet Explorer is easy. The browser is
>> so non-standard that equal signs are treated as colons. All browsers
>> use the last setting found if the same property is repeated. Try
>> something like:
>> display: table-cell;
>> display=block;
>> float=left;
I prefer the separation of style sheet documents
>>
>
> at the risk of sounding like a broken record: this is invalid css.
> (oh yes, i validate mine :)
> i don't want to see validation warnings all over the place that
> obscure the real problems. imnsho, the *only* acceptable way to hack
> around M$' standards dyslexia are conditional comments. preferably
> by including files named
> "really_stupid_microsoft_fuckup_workaround.css".
>
> --
> Jörn Nettingsmeier
>
> "One of my most productive days was throwing away 1000 lines of code."
> - Ken Thompson.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
> For additional commands, e-mail: dev-help@lenya.apache.org
>
Jürgen Ragaller
null-oder-eins GmbH Zürich
ragaller@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jörn Nettingsmeier <ne...@apache.org>.
solprovider@apache.org wrote:
> On 11/16/07, Jörn Nettingsmeier <ne...@apache.org> wrote:
>> if a certain software vendor cannot be bothered to provide the most
>> trivial bugfixes and read a f"$§%ing spec, that's really not our
>> problem. anyone with half a brain can download and understand the
>> relevant standards documents, and if the world's largest pile of
>> software engineers can't be bothered, well, tough luck for their users.
>
> That certain software vendor's buggy software is still rather popular
> (>80%?) Version 6 is still popular because version 7 is even more
> buggy. From the evidence, none of the company's programmers have half
> a brain, and few are able to read. If our project wants to be used
> for websites, we need to make certain our software works with that
> buggy non-standard web browser.
not at the cost of standards conformance and proper validation. a stupid
bug is a stupid bug no matter the market share. people putting up with
this kind of shit have made web designer's lives miserable for ten years
now.
it's not as if we'd lose those users by being picky about our code.
>> if anyone wants to provide a patch with an IE-specific conditional
>> comment, that's fine with me. clean code for the rest of us, and it's
>> plain to see it's an extra bugfix for IE.
>
> Special CSS for Microsoft Internet Explorer is easy. The browser is
> so non-standard that equal signs are treated as colons. All browsers
> use the last setting found if the same property is repeated. Try
> something like:
> display: table-cell;
> display=block;
> float=left;
at the risk of sounding like a broken record: this is invalid css. (oh
yes, i validate mine :)
i don't want to see validation warnings all over the place that obscure
the real problems. imnsho, the *only* acceptable way to hack around M$'
standards dyslexia are conditional comments. preferably by including
files named "really_stupid_microsoft_fuckup_workaround.css".
--
Jörn Nettingsmeier
"One of my most productive days was throwing away 1000 lines of code."
- Ken Thompson.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by so...@apache.org.
On 11/16/07, Jörn Nettingsmeier <ne...@apache.org> wrote:
> Jürgen Ragaller wrote:
> > | svn-commit: Use display:table-cell instead of float:left for tabbed
> > areas to avoid wrapping
> >
> > This now looks nice in firefox - but IE (6 and 7) ignore display:table-cell
> > I'll do some investigation about how to avoid the wrapping in IE.
> >
> > BTW: IE6 still has way too big fonts for the authoring area (and the
> > default publication) - the reason is the xml Prolog (let's ie 6 switch
> > to quirks mode).
> >
> > What about removing the prolog as long as ie 6 is still around in
> > relevant numbers?
>
> over my very dead corpse.
> if a certain software vendor cannot be bothered to provide the most
> trivial bugfixes and read a f"$§%ing spec, that's really not our
> problem. anyone with half a brain can download and understand the
> relevant standards documents, and if the world's largest pile of
> software engineers can't be bothered, well, tough luck for their users.
That certain software vendor's buggy software is still rather popular
(>80%?) Version 6 is still popular because version 7 is even more
buggy. From the evidence, none of the company's programmers have half
a brain, and few are able to read. If our project wants to be used
for websites, we need to make certain our software works with that
buggy non-standard web browser.
> if anyone wants to provide a patch with an IE-specific conditional
> comment, that's fine with me. clean code for the rest of us, and it's
> plain to see it's an extra bugfix for IE.
> Jörn Nettingsmeier
Special CSS for Microsoft Internet Explorer is easy. The browser is
so non-standard that equal signs are treated as colons. All browsers
use the last setting found if the same property is repeated. Try
something like:
display: table-cell;
display=block;
float=left;
HTH,
solprovider
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jörn Nettingsmeier <ne...@apache.org>.
Jürgen Ragaller wrote:
> Hi
>
> | svn-commit: Use display:table-cell instead of float:left for tabbed
> areas to avoid wrapping
>
> This now looks nice in firefox - but IE (6 and 7) ignore display:table-cell
> I'll do some investigation about how to avoid the wrapping in IE.
>
>
> BTW: IE6 still has way too big fonts for the authoring area (and the
> default publication) - the reason is the xml Prolog (let's ie 6 switch
> to quirks mode).
>
> What about removing the prolog as long as ie 6 is still around in
> relevant numbers?
over my very dead corpse.
if a certain software vendor cannot be bothered to provide the most
trivial bugfixes and read a f"$§%ing spec, that's really not our
problem. anyone with half a brain can download and understand the
relevant standards documents, and if the world's largest pile of
software engineers can't be bothered, well, tough luck for their users.
if anyone wants to provide a patch with an IE-specific conditional
comment, that's fine with me. clean code for the rest of us, and it's
plain to see it's an extra bugfix for IE.
--
Jörn Nettingsmeier
"One of my most productive days was throwing away 1000 lines of code."
- Ken Thompson.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jürgen Ragaller <ra...@apache.org>.
Hi Andreas
Am 15.11.2007 um 16:05 schrieb Andreas Hartmann:
> Hi Jürgen
>
> Jürgen Ragaller schrieb:
>
>> Hi
>> | svn-commit: Use display:table-cell instead of float:left for
>> tabbed areas to avoid wrapping
>> This now looks nice in firefox - but IE (6 and 7) ignore
>> display:table-cell
>>
>
> thanks for the hint!
>
>
>> I'll do some investigation about how to avoid the wrapping in IE.
>>
>
> I hope we don't have to use a real table :(
>
uuuhhhh - let's hope not (shudder); I'll give it a try.
>
>
>> BTW: IE6 still has way too big fonts for the authoring area (and
>> the default publication) - the reason is the xml Prolog (let's ie
>> 6 switch to quirks mode).
>> What about removing the prolog as long as ie 6 is still around in
>> relevant numbers?
>>
>
> Or we could set the font sizes with "pt", e.g. 11pt. Would this help?
>
I'd prefer to throw out the xml declaration (starting with <!
DOCTYPE...) - the xhtml is valid without it.
If the xml-declaration stays I think it's best to work around the
(hard to predict) effects of the quirks mode using a ie6-only
stylesheet (with font size corrections and other unquirkings) called
via a conditional comment:
<!--[if lte IE 6]><link rel="stylesheet" href="lteie6hacksonly.css"
type="text/css"/><![endif]-->
Jürgen
null-oder-eins GmbH Zürich
web & graphic design
www.null-oder-eins.ch
ragaller@apache.org
Skype: callto://ragaller
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Jürgen Ragaller <ra...@null-oder-eins.ch>.
Hi Andreas
Am 15.11.2007 um 16:05 schrieb Andreas Hartmann:
> Hi Jürgen
>
> Jürgen Ragaller schrieb:
>> Hi
>> | svn-commit: Use display:table-cell instead of float:left for
>> tabbed areas to avoid wrapping
>> This now looks nice in firefox - but IE (6 and 7) ignore
>> display:table-cell
>
> thanks for the hint!
>
>> I'll do some investigation about how to avoid the wrapping in IE.
>
> I hope we don't have to use a real table :(
uuuhhhh - let's hope not (shudder); I'll give it a try.
>
>> BTW: IE6 still has way too big fonts for the authoring area (and
>> the default publication) - the reason is the xml Prolog (let's ie
>> 6 switch to quirks mode).
>> What about removing the prolog as long as ie 6 is still around in
>> relevant numbers?
>
> Or we could set the font sizes with "pt", e.g. 11pt. Would this help?
I'd prefer to throw out the xml declaration (starting with <!
DOCTYPE...) - the xhtml is valid without it.
If the xml-declaration stays I think it's best to work around the
(hard to predict) effects of the quirks mode using a ie6-only
stylesheet (with font size corrections and other unquirkings) called
via a conditional comment:
<!--[if lte IE 6]><link rel="stylesheet" href="lteie6hacksonly.css"
type="text/css"/><![endif]-->
Jürgen
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: wrapping of tabbed areas
Posted by Andreas Hartmann <an...@apache.org>.
Hi Jürgen
Jürgen Ragaller schrieb:
> Hi
>
> | svn-commit: Use display:table-cell instead of float:left for tabbed
> areas to avoid wrapping
>
> This now looks nice in firefox - but IE (6 and 7) ignore display:table-cell
thanks for the hint!
> I'll do some investigation about how to avoid the wrapping in IE.
I hope we don't have to use a real table :(
> BTW: IE6 still has way too big fonts for the authoring area (and the
> default publication) - the reason is the xml Prolog (let's ie 6 switch
> to quirks mode).
>
> What about removing the prolog as long as ie 6 is still around in
> relevant numbers?
Or we could set the font sizes with "pt", e.g. 11pt. Would this help?
-- Andreas
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org