|
Posted by GTalbot on November 11, 2007, 6:13 pm
Please log in for more thread options
> GTalbot wrote:
>
> >> IE6 has serious problems with positioning, and with floats in
> >> particular. Adding position:relative to a container can resolve many i=
ssues.
>
> > Adding more css rules or css declarations is never a good
> > recommendable way to debug or to workaround bugs as these are often
> > not forward-compatible and these increase complexity, leading and
> > bringing often other bugs surfacing.
>
> Adding 1 standard property to 1 container hardly constitutes complexity,
Hello Berg,
It certainly is not reducing complexity. This issue is furthermore
relevant and important if the original webpage was already bloated,
over-coded, over-constrained, divitis-like, classitis-like, over-
contained, etc... A wide DOM tree of nodes and a deep DOM tree of
nodes is a very frequent "aspect" of a wide majority of current
webpages on the web.
The more a DOM tree of nodes is wide and deep, the more complex the
interaction of elements can be. Overconstrained coding is also very
very frequent in webpages and is the cause layout bugs.
In my opinion, the most important thing (after validating markup code
and CSS code and choosing a strict DTD) when debugging a webpage is to
reduce that DOM tree complexity, remove unneeded and over-nesting
(over-containing) nodes, unneeded and over-declared CSS declarations
and CSS rules.
> or an issue with forward compatibility. It is always wise, however, to
> make sure such things don't have negative impacts on browsers that are
> better than IE.
>
Agreed.
> > Removing declarations, reducing DOM tree (width and depth), relying on
> > inheritance and on browser defaults are things I recommend.
>
> Great. Go for it. If, however, you expect to have something a bit more
> stylish than gtalbot.org, you may have to reconsider that position.
>
> > People using IE6
> > should just switch to a better browser otherwise at least upgrade to
> > IE7
>
> Good luck with that, too. IE7 requires minimum Windows XP, which a lot
> of people still don't have.
A lot of people (approx. 90% of all Windows users) have Windows XP or
Windows Vista according to all the webstats sources I know of
(w3counter.com, thecounter.com, etc).
> I'm no IE lover, but I'm of the opinion that
> it's bloody arrogant of web authors to tell their visitors what browser
> they should be using.
>
My first, best recommendation is not to upgrade to IE7 but to switch
(Firefox 2, Opera 9, Safari 3, Konqueror 3.5, even Icab).
Inviting IE6 users to switch is not arrogant by itself. It all depends
on how a web author is doing that, how the invitation is done. Manners
count. Forcing, annoying, telling with arrogance are not ways, manners
I had in mind. Such an invitation is often proposed, implicit in help,
contact us, code conformance webpages of big websites, professional
websites. I found a few good ones on this particular issue. I'm not
referring to a dry, blunt or agressive "upgrade" message/slogan here.
> >> Paragraphs have some default
> >> margins but browsers don't have to agree
>
> > Opera 9, Safari 3, Firefox 2, Icab 3 and Hv3 TKHTML alpha 16 (and
> > other acid2-test-compliant browsers) agree on how much and when to
> > collapse adjoining margins. The problem is with IE6 and IE7, period.
>
> IE being different in this respect is not a bug.
No. IE has bugs regarding how much (margins) and when collapsing
adjoining margins. Definitely. Absolutely.
> There is no spec that
> says how much or where margins go.
margin: auto should be resolved as 0: the CSS 2.1 spec says that for
block-level elements. Internet Explorer 6 and Internet Explorer 7 do
not do that. So, web authors are confused, baffled as they have no
idea what is causing the vertical gap, vertical spacing between
elements in Internet Explorer 6 and in Internet Explorer 7:
http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/OrdinaryPargWithMarginA=
uto.html
and you can even remove the black 3px border around the container and
the bug (CSS 2.1 spec. violation) is still there.
> It's nice that some browsers agree,
> but it's not a requirement, nor should you count on it.
>
When the container is not floated in Internet Explorer 6 and in
Internet Explorer 7, a nested block has default margin-top and margin-
bottom. When the container is floated in Internet Explorer 6 and in
Internet Explorer 7, a nested block no longer has default margin-top
and margin-bottom.
http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/DefaultMarginDisappear.=
html
If that is not a bug, then it certainly is a non-documented feature
(swindle) in Internet Explorer 6 and in Internet Explorer 7 ...
Regards, G=E9rard
| 
XML Sitemap