Forum Moderators: open
Anyone got any idea how to remedy this?
I did'nt get the oppotunity to view the site in any other mac browsers,
I'd be intrested to know what happens in
ie5.5 ie6 mozilla and safari.
<snip>
thanks
Paul
[edited by: korkus2000 at 4:07 pm (utc) on Aug. 31, 2003]
[edit reason] No site reviews TOS #21 [/edit]
As for other Mac browsers... there is no IE 5.5 and IE 6.0 for MacOS. The latest version of MacIE is version 5.2, which is almost as standards compliant as Gecko and KHTML, but is hideously slow. I've never been able to test my site on a Mac, I just test it in WinIE 6.0, WinIE 5.0, WinIE 4.0, Netscape 7.1, Netscape 6.2, Netscape 6.0, Netscape 4.8, Netscape 4.0, Netscape 3.0, Opera 7.1, Opera 6.0, Opera 4.0 and Konqueror 3.1 and hope for the best.
Please can you be a bit more specific about "falls apart". You've done your layout with tables (rather than position:absolute), so I can't really imagine how it can fall appart, as your layout it actually quite basic. Is it just the case that it doesn't look quite so pretty in MacIE 5.0?
Thanks for the reply hartlandcat,
not so much of "does'nt look so pretty" but more of becomes
unreadlable/unusable, tables get pushed out of synch with each other.
I'am unable to get access to a mac at the moment, so I can't go into a more detailed decription. But what a mean by the term "falls apart" is that users viewing the page on that platform would go elsewhere.
Its quite annoying that a site that validates to W3C standards in both html (version 4.1) and css. Is still able to become unusable in ie5 mac.
It is likely that other mac browsers (safari, mozilla) would
render the pages with such 'orrible' results?
Is the script, I could use to send ie mac viewers to an alternate page/s
cheers
Paul
1. IE5 on macintosh has much weaker error recovery than IE 5 on PC. However, you say the HTML validates, so that shouldn't be the issue.
2. Table layouts are lot more flexible than we tend to think. We put in a width declaration and assume that it will be rendered that width. But the W3C recommendations for user agents allow these attributes to be overruled at rendering time, because it's possible (even for validated HTML) to give impossible or mutually contradictory instructions. Maybe this is what's going on for you?
I had to use pencil and paper to layout the tables, including the width declarations, and then go through it logically. Eventually I discovered that the mac browser was collapsing one cell down to a narrower width than I had declared, because of activity elsewhere in the table. The PC browser was making a different decision. Using a spacer gif in that particular cell fixed the problem.
It is likely that other mac browsers (safari, mozilla) would
render the pages with such 'orrible' results?
I'd say that you need to test to find out - it's much more of a browser issue than a platform issue. There's no consistency from one browser to another just because they're on the same OS.
Is the script, I could use to send ie mac viewers to an alternate page/s
Yes, you can use javascript to sniff for a particular platform and browser and then serve either a different page, or a different stylesheet. But that approach can become quite complex, and also a nasty maintenance headache. I'll bet you can resolve this layout issue in a cross platform way.
What PC browsers have you viewed the page on, besides IE? Is all OK on recent Mozilla or Netscape?
I've just been reading an article at [url=http://www.alistapart.com/stories/doctype/]ALA[/url] about doctypes and was wandering if that could have any thing to do with the problem. (more here at [url]http://macedition.com/cb/ie5macbugs/notbugs.html[/url]
I've just amended the pages in the site to from
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
to
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
apparently mac ie 5 is quite picky about this(?)
and so is netscape 6 which I have'nt yet tested in.
article here
[url]http://www.evolt.org/article/Does_Netscape_6_Break_Your_Table_Layouts/17/4427/[/url]
But I suspect the problem is more likely to do with your
suggestion about width and height declarations.
I haven’t got access to a mac at the moment so can't experiment with finding a solution :(
I need to test in netscape 6 though
//edit works fine in netscape 6
thanks for your comments
Try to remove the doc type declaration completely, that does no good anyway and might be forcing some kind of rendering to happen, putting the browser into compliance mode, removing the doc type declaration will just get rid of some potential display problems, try testing without it altogether.
The odds are excellent that the page will display right on mozilla for mac, since it's the same gecko rendering engine in all cases as far as I know. I haven't seen any noticeable differences between any of the platforms mozilla runs on, except fonts in linux versions, but that problem is fixed now in latest linux distros from what I've seen.
Ie mac 5x is always a problem, I've almost given up supporting it, although I try to get the stuff to more or less work because I have clients that use it, but even they have gotten used to the familiar refrain: it's an ie mac 5x bug, sorry.
Because so many developers and designers do not check their work on a Macintosh, we hear less about support problems for IE5/mac than we do for NN4/pc. But it really is a more important browser at this point in time.
However, I know several businessmen who prefer Macintosh - and they all know that it takes more than one browser to get around the web on a Mac and not miss anything important. A sad but true fact of life in the Apple barrel.
In my logs more than half of Mac users are using Safari. A few months ago most of my Mac users were on IE.
The comment about Mac users using many different browsers is correct. Some still don't and won't, but some are coming around. Safari may be good for most tasks, but some websites (especially banks, school registrations, and govt transactions) ONLY work for IE. (Because of their broken, non-compliant code.)
Or even worse because they have a function that checks if you are using IE, if you don't it automatically blocks you - even though the browser works fine.
I hacked mozilla's user agent string to show that it was internet explorer and a site that said I needed IE to continue let me right in. ;)
As to the support for Mozilla on Mac, Mozilla.org skipped the offcial support for Mac OS 9.x somewhere between 1.2.1 and 1.3 focusing completely on Mac OS X which is very well supported.!However! somebody put together a Mac OS 9.x version of Mozilla 1.3.1 with additional bugfixes over and above the official 1.3.1 - see <http://wamcom.org/> he has more or less taken Mozilla 1.3.1 and branched it.
Personally, I've long since skipped supporting IE5 Mac except by chance. But on the other hand IE5 Mac has a very good standards compliant css support, so I havn't seen that many problems - neither in the javascript department.
I'm coding to Mozilla and the standards. Subsequently I then use the IE only conditional comment to fix the css errors in IE5.x Win & IE6 by loading supplementary css files.