NM Public wrote: >>> <http://deflexion.com/>
> Thank you very much Georg! After adding your suggested style rule, I > get these results at Browsershots.org: > > Screenshot request group 5 > <http://browsershots.org/http://deflexion.com/> > > MSIE 5.5 and MSIE 6.0 on Windows 2000 now both have the left sidebar > - yay! But the text in the sidebar stops part way down the page (and > then, I assume, continues below the mainClm, (it's not shown in the > image)). > > For MSIE 7.0 on Windows XP, the left sidebar is there, but it's > empty. I assume that all the text appears below the mainClm text > (again this isn't shown in the image). Incomplete loading of sidebar content in that browsershots-IE7 it seems. My IE7/winXP present a complete page with sidebar-content as intended, but indicates script-errors in the page that _may_ cause failure in browsershots. On narrow windows parts of sidebar-content will start to drop - in all browsers - simply because there's not enough space for the long and unbreakable link-text and other parts in there. The sidebar is "in the flow" - not floating, so it won't drop in one piece but bit by bit from the "too wide" part(s) as window gets narrower. Testing and fixing some of the "too wide" parts will improve things a lot. > So, it is better than before [*], but still not great. I wonder if I > should include a standards mode Doctype in this page, e.g.: > > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" > "http://www.w3.org/TR/html4/strict.dtd";> > > Or if that will cause more problems than it solves? The validator flags 967 errors when checking that page as HTML 4.01 Transitional, and 2430 errors when checking as XHTML 1.0 Transitional, and the markup shows a syntax-mix of both. As HTML 4.01 Strict it flags 1172 errors. Clearly that page isn't ready for a "standard mode triggering doctype", and needs serious clean-up before that is an option - IMO. For more, see: <http://www.gunlaug.no/contents/wd_additions_25.html> > And I wonder why Blogger didn't use a Doctype in this template in the > first place?! Have no idea, but it wouldn't help if the markup isn't improved. > Thank you again for your help. I know the best course is for me to > redsign the entire page using standards mode and clean CSS, but if > you have any thoughts about my wonderings above. Or suggestions for > temporary fixes, I appreaciate them. As mentioned: I don't think there are as many problems for the last iteration as browsershots points out. At least I can't see them, but I don't have all those browsers to test on locally. I'd advice you to leave what you have more or less as it is, until you have a complete re-design ready and tested. Generally: IE6 won't gain much from going into standard mode, and older IE/win versions _can't_ go into standard mode. IE7 may gain some. IE6 and IE7 support the "W3C box model" when in standard mode, but you still have to feed corrections to the older versions. For more, see: <http://www.gunlaug.no/contents/wd_additions_16.html> Opera, Firefox, Safari etc. don't care much about mode - as long as the markup and CSS is good. Mode-differences are few in these browsers since they for a long time have been developed close to what we now know as standards, and are only keeping their various quirks mode rendering in order to support old legacy pages. Only IE/win has really problematic mode-related differences, and that won't change. It is the the quality of the markup and CSS - and the choice of layout-solution - that'll make the real difference, so I would start planning for a re-design in accordance with standards. Marking up in accordance with a "Strict" standard will certainly be the right choice, and there are a number of quite well-functioning two-columns solutions around to learn and borrow from. regards Georg -- http://www.gunlaug.no ______________________________________________________________________ css-discuss [EMAIL PROTECTED] http://www.css-discuss.org/mailman/listinfo/css-d List wiki/FAQ -- http://css-discuss.incutio.com/ List policies -- http://css-discuss.org/policies.html Supported by evolt.org -- http://www.evolt.org/help_support_evolt/
